A preview of this full-text is provided by Wiley.
Content available from Tectonics
This content is subject to copyright. Terms and conditions apply.
1. Introduction
The Ordos Block is located northeast of the Tibetan Plateau (Figure1a; Deng etal.,1999). Since the Ceno-
zoic, due to the rapid expansion of the Tibetan Plateau to the northeast (Gaudemer et al., 1995; Tapponnier
etal.,1982, 1986) and subduction of the Pacific plate to the west (Northrup etal.,1995; Uyeda & Kanamo-
ri,1979), at least four M≥8 large earthquakes have been triggered in the basin on the periphery of the Ordos
Block. These earthquakes were the 1303 Hongdong M 8.0 earthquake in the Shanxi Basin in the eastern block
(Deng & Xu,1994; Jiang etal.,2004), the 1556 Huaxian M 81/4 earthquake in the Weihe Basin in the southern
block (Deng,2007; Feng etal.,2020; Hou etal.,1998), the 1739 Yinchuan-Pingluo M 8.0 earthquake in the Yin-
chuan Basin (Lei etal.,2015; Middleton etal.,2016), and the 1920 Haiyuan M 8.5 earthquake in the southwestern
block (Ou etal.,2020; Zhang etal.,1987). The distribution of these large earthquakes indicates that around the
Ordos Block, only the northern Hetao Basin lacks a historical record of M≥8 earthquakes (Figure1a). According
to drilling and petroleum geological section data, the thickness of the Cenozoic sediments in the Linhe Basin
in the western part of the Hetao Basin reaches 12,000m, with Quaternary sediments accounting for 2,400m
(AppendixC; Li & Nie,1987; State Seismological Bureau [SSB],1988). Since the late Pleistocene, the sediment
thickness has reached 240m (Figure1a). These sediments are much thicker than those in other basins around
Ordos from the same period, indicating that the faults in the Linhe Basin should have a higher subsidence rate.
The Linhe Basin must have very large active faults to explain the fast subsidence.
According to previous studies, the deformation of the Linhe Basin is mainly distributed on the northern boundary
faults (Wen,2014). The NE-striking Langshan piedmont fault (LPF) is the northwestern boundary of the basin,
the E-W-striking western section of the Seertengshan piedmont fault (WSPF) is the northern boundary of the ba-
sin, and the NW-striking eastern section of the Seertengshan piedmont fault (ESPF) is the northeastern boundary
of the basin. These three faults together form a flat roof-like northern boundary of the Linhe Basin (Figure1b).
Previous researchers, through the study of trenches and structural geomorphology, confirmed that these three
faults have been highly active since the late Pleistocene (Cheng etal.,2006; Chen, Ran, & Chang,2003; Chen,
Abstract The activity of major active faults around the Ordos Block is of great interest to seismologists, as
at least four M≥8 earthquakes have been recorded. However, the Linhe Basin, which has the thickest Cenozoic
sediments, has no record of large earthquakes. Does this basin have the structural conditions required for
large earthquakes? The northern boundary fault of the Linhe Basin is composed of the NE-striking Langshan
piedmont fault (LPF), E-W-striking western section of the Seertengshan piedmont fault (WSPF), and NW-
striking eastern section of the Seertengshan piedmont fault (ESPF). Based on large-scale active fault mapping,
this article analyzes data from 23 trenches, using an unmanned aerial vehicle to measure the faulted landform,
and combines these data with Quaternary dating methods to acquire the paleoearthquake sequences of the LPF,
WSPF, and ESPF. Furthermore, this article explores their rupture modes and discusses the structural evolution
of two intersection points. Through paleoseismic comparison, seven trenches revealed historical earthquakes
from 7 BC, with a common magnitude of M 8.1. The trenches also revealed seven paleoseismic events since
the Holocene, which conformed to the periodic model with a period of 1.37±0.11ka. The elapsed time since
the latest event (2.0ka) has exceeded the earthquake recurrence period; thus, the area is currently at risk of an
M 7.4–8.0 earthquake. The NE-striking LPF and E-W-striking WSPF are connected by two left-stepping small
fault segments. The E-W-striking WSPF and NW-striking ESPF are connected by a large triangular relay ramp.
LIANG ET AL.
© 2021. American Geophysical Union.
All Rights Reserved.
Joint-Rupture Pattern and Newly Generated Structure of
Fault Intersections on the Northern Margin of the Linhe Basin,
Northwestern Ordos Block, China
Kuan Liang1 , Zhongtai He1 , Baoqi Ma1, Qinjian Tian2, and Shao Liu3
1National Institute of Natural Hazards, Ministry of Emergency Management of China, Beijing, China, 2Institute of Earthquake
Forecasting, China Earthquake Administration, Beijing, China, 3Sichuan Earthquake Administration, Chengdu, China
Key Points:
• Seven trenches revealed 7 BC
historical earthquakes, with a
calculated magnitude of M 8.1
• Since the elapsed time of the latest
event (2.0ka) has exceeded the
recurrence period, the Linhe Basin is
at risk of an M 7.4–8.0 earthquake
• The western section and the eastern
section of the Seertengshan piedmont
fault are connected by a large
triangular relay ramp
Supporting Information:
Supporting Information may be found in
the online version of this article.
Correspondence to:
Z. He,
hezhongtai@126.com
Citation:
Liang, K., He, Z., Ma, B., Tian, Q., &
Liu, S. (2021). Joint-rupture pattern
and newly generated structure of fault
intersections on the northern margin
of the Linhe Basin, northwestern
Ordos Block, China. Tectonics,
40, e2021TC006845. https://doi.
org/10.1029/2021TC006845
Received 30 MAR 2021
Accepted 28 NOV 2021
10.1029/2021TC006845
RESEARCH ARTICLE
1 of 29
Tectonics
LIANG ET AL.
10.1029/2021TC006845
2 of 29
Ran, & Yang,2003; He etal.,2014; Jiang,2002; Li etal.,2015; Li & Nie,1987; Ran etal.,2003; Sun etal.,1990;
Wang etal.,1984; Yang etal., 2002,2003). According to the study by Dong etal.(2018), the LPF has expe-
rienced three events that ruptured a whole section (160km) since the Holocene. According to the relationship
between the length of the rupture and the magnitude of the earthquake, the magnitude of the earthquake was
estimated to be M ∼ 7.8 (Dong etal.,2018; Rao etal.,2016). The WSPF and ESPF have obvious segmentation
(Chen, Ran, & Chang,2003; He etal.,2018), and no research has shown that a great earthquake caused the com-
plete rupture of the Seertengshan piedmont fault (SPF). Therefore, neither the LPF nor the SPF currently seem
to have the conditions required for an M 8.0 earthquake alone. Therefore, is it possible for a joint rupture of the
LPF and one or more segments of the SPF to occur and form an M 8.0 earthquake? Moreover, what is the tectonic
environment at their intersection points?
Figure 1. Regional tectonic setting. (a) Distribution of faulted basins, active faults and M≥5.5 medium-large earthquakes along the periphery of the Ordos Block
(modified from Deng,2007). The inset map shows that the Ordos Block is located on the northeastern side of the Tibetan Plateau. Historic earthquakes are labeled (by
size) as small dots, and red dots indicate the epicenter of large M≥8.0 earthquakes. The blue area shows the range of the Jilantai-Hetao megalake. (b) Distribution
of active faults in the Linhe Basin in the western part of the Hetao Basin. The focal mechanism solution data originated from Harvard’s CMT catalog. Red stars:
geological and geomorphological observation sites. Black triangle: optically stimulated luminescence (OSL) sampling point of terraces. The white numbers correspond
to Table1. LPF: Langshan piedmont fault; WSPF: western segment of the Seertengshan piedmont fault; ESPF: eastern segment of the Seertengshan piedmont fault;
WPF: Wulashan piedmont fault.
Tectonics
LIANG ET AL.
10.1029/2021TC006845
3 of 29
In normal fault development areas, multiple geomorphic surfaces of different sizes, such as planation surfaces,
piedmont erosion surfaces, and river terraces, often record surface dynamic processes on different time scales
(Burbank & Anderson,2011; Commins etal., 2005; Cowie etal.,2006; Hopkins & Dawers,2015; Kirby &
Whipple,2012; Pezzopane & Weldon,1993; Whittaker & Walker,2015; Yang et al.,2020.; Zhang, Molnar, &
Xu,2007). In field investigations, understanding the genetic mechanism and distribution characteristics of var-
ious geomorphological surfaces and accurately identifying them are important components for the comparison
of surfaces along faults (Maddy etal.,2000,2001; Starkel,2003; Sun,2005). This article uses paleoseismology,
geomorphology, and Quaternary dating methods to obtain the paleoseismic sequences of the LPF and SPF seg-
ments, to explore the connection between them and to discuss the tectonic evolution of the northwestern corner
(intersection of the LPF and WSPF) and northeastern corner (intersection of the WSPF and ESPF) of the Linhe
Basin.
Location U (μg/g) Th (μg/g) K (%) Environmental Dose (Gy/ka) Equivalent Dose (Gy) Age (ka) Terrace
Dongquanzicun (3) 3.27 10.7 2.23 4.78 37.72±2.27 8.32±0.96 T1
Dongquanzicun (3) 2.28 9.86 2.08 4.20 41.09±4.46 9.79±1.44 T1
Xishuidaobei (9) 1.82 9.96 1.88 3.84 44.25±1.58 11.54±1.23 T1
Wujiahe (6) 2.31 10.2 1.84 3.99 37.13±2.41 9.29±1.11 T1
Shuiquancun (7) 1.12 5.12 2.38 3.59 80.39±4.46 22.42±2.56 T2
Fanrongwushe (1) 1.83 9.29 2.23 4.13 136.32±15.44 33.01±1.99 T2
Langshankou (2) 2.82 10.1 2.23 4.55 108.16±11.39 23.97±3.45 T2
Mabozi (8) 1.54 11.4 1.79 3.79 165.13±16.93 23.25±2.88 T2
Alagaitu (4) 1.18 6.04 2.19 3.51 184.78±9.32 52.69±5.90 T3
Xiliushuiquanzi (5) 2.08 8.21 2.41 4.30 309.82±28.31 72.05±9.76 T3
Shuiquancun (7) 1.4 5.21 2.28 3.60 217.40±18.61 60.34±7.94 T3
Dashetai (10) 1.65 11.4 2.28 4.01 286.35±26.71 71.38±9.76 T3
Bafenzidi (11) 2.08 9.07 1.76 3.71 247.50±16.58 66.72±8.03 T3
Wulanhudong (12) 1.97 9.46 1.91 3.87 307.39±21.29 79.50±9.67 T3
Delingshan 1.53 7.34 1.73 3.30 98.30±3.85 29.81±3.20 Trench
Delingshan 1.04 4.65 2.27 3.35 125.16±13.50 37.40±5.50 Trench
Delingshan 0.66 3.45 2.5 3.28 144.62±18.44 44.03±7.13 Trench
Delingshan 0.932 5.32 2.16 3.22 107.68±14.64 33.45±5.64 Trench
Delingshan 1.08 5.26 2.27 3.41 130.97±10.31 38.37±4.88 Trench
Delingshan 1.55 7.87 2.19 3.84 5.78±0.48 1.51±0.20 Trench
Delingshan 1.04 6.33 1.72 2.99 66.55±6.23 22.25±3.05 Trench
Delingshan 0.773 4.9 2.1 3.07 214.67±13.24 69.88±8.21 Trench
Note. The sampling locations of the terraces are shown in Figure1b by black triangles. The numbers behind the names in Table1 correspond to the numbers in the
black triangles in Figure1b.
Table 1
Optically Stimulated Luminescence Ages of Samples
Field Code Laboratory ID Method Conventional age Description
DWGG-TC-14C-01 543356 AMS-Standard 5,450±30 BP Organic sediment; Figure5
DWGG-TC-14C-02 543357 AMS-Standard 2,440±30 BP Organic sediment; Figure5
Table 2
Radiocarbon Age Data for Samples
Tectonics
LIANG ET AL.
10.1029/2021TC006845
4 of 29
2. Regional Tectonic Setting
The LPF is the northwestern boundary of the Linhe Basin. The fault starts near Hashatu in the south, extends NE
through Wulanhushu, Qingshanzhen, and Wugaizhen, and connects to the WSPF in Dongwugaigou (Figure1b).
The LPF has a total length of 160km and an overall strike of N55°E. The fault dips to the southeast at ∼60°–70°
(Dong,2016; Dong etal.,2018; Rao etal.,2016). Dong(2016) states that the fault does not have rupture segmen-
tation, and the entire fault is broken when a large earthquake occurs. The SPF forms the northern and northeastern
boundary of the Linhe Basin. According to the fault strike, the SPF can be divided into the WSPF and the ESPF.
Figure 2. (a–c) Formation model of two fault terraces P1 and P2, which are distributed along the fault (revised from Zhang, Molnar, & Xu,2007). (d–h) Formation
model of four flood terraces T1–T4 at the exit of the gully, which are distributed perpendicular to the fault strike (revised from He etal.,2015). (i) Geomorphological
development model of the study area. Two joint geomorphic faces are formed.
Figure 3. Geomorphology and topography of the Sanguibulong site (Site 1). (a) Map showing the interpretation of the Sanguibulong site landscape. The gully shows
evidence that four terraces are developed in this area. (b) Terrace profile in the NE direction. (c) Ground fissures developed on P1, and the mound left by tree planting
was faulted by the ground fissure. (d) Fault spring in the gully. (e) Multiple faults developed in the northwestern wall of the artificially excavated trench. Due to the
activity of the faults, the dip of the strata is opposite to that of the faults. (f) Multiple collapse wedges appear in the northern wall of the trench. (g) A number of faults
are developed in the eastern wall of the trench. These faults cut each other and indicate multiple slip events.
Tectonics
LIANG ET AL.
10.1029/2021TC006845
5 of 29
The WSPF strikes nearly E-W, starting near Dongwugaigou, extending eastward through Shilanji, Hulesitai, and
Wujiahe, and continues for several kilometers beyond Wubulankou. The length of the fault is ∼105 km. The
strike of the ESPF is ∼300°, and it extends southeastward through Delingshan, Dashetai, and Wulanhudong and
gradually disappears from the ground surface near Tailiang, with a total length of ∼85km (Figure1b). Previous
researchers discussed the segmentation of the SPF according to the geometry and activity of the fault (Chen, Ran,
& Chang,2003; Chen, Ran, & Yang,2003; Yang etal.,2002,2003). The latest research divides the SPF into four
segments, with the WSPF being divided into the Wujiahe segment and the Hongqicun segment and the ESPF
being divided into the Kuluebulong segment and the Dashetai segment (He etal.,2017; Long etal.,2017; Zhang
etal.,2017). The maximum slip rate of the SPF since 65ka has been calculated to be 1.81mm/a on the Wujiahe
segment, 1.66mm/a on the Hongqicun segment, and 0.94mm/a on the Kuluebulong segment (Zhang etal.,2017).
Previous studies indicated that a “Jilantai-Hetao megalake” existed. The lake reached its highest surface (at
an altitude of 1,080m) at ∼60–50ka (Chen, Fan, Chun, Madsen, Oviatt, Zhao, etal.,2008; Chen, Fan, Chun,
Madsen, Oviatt, Zhao, & Yang,2008). The Jilantai-Hetao megalake extended west to the southwest of Jilantai
Salt Lake, east to eastern Hohhot, south to the northern edge of the Ordos Block, and north to the southern
piedmont of Bayanwulashan Mountain, Langshan Mountain, Seertengshan Mountain, Wulashan Mountain, and
Daqingshan Mountain (Figure1a). The average depth under modern terrain conditions is ∼50m, the lake area is
∼34,757km2, and the entire lake basin can reach 6,000km3 (Yang,2008). Along the LPF and SPF, many lacus-
trine layers appear on the profiles, which are mainly composed of yellow-green fine silt sand. This layer is a good
marker layer to calculate the vertical slip rates of the faults (Liang etal.,2019; Zhang etal.,2017).
Topographically, a multilevel terrace landscape is present along the foothills of the Langshan Mountains and
Seertengshan Mountains. According to previous studies, the formation of terraces is closely related to the inter-
mittent activity of piedmont faults (Cheng etal.,2006; He etal.,2014; Jia etal.,2015,2016; Liu etal.,2016).
Therefore, it is feasible to explore the characteristics of tectonic activity in the study area through the comparison
of terraces. The NE-striking LPF and E-W-striking WSPF intersect in Dongwugaigou at an angle of ∼145°. The
E-W-trending WSPF and SE-trending ESPF intersect near Wubulangkou at an angle of ∼150°. Their intersection,
evolution process and ability to generate large earthquakes are worth further study.
3. Study Methods
3.1. Unmanned Aerial Vehicle Measurement
To map active fault strands and geomorphic surfaces and to precisely measure offsets, an unmanned aerial vehicle
(UAV) was used to measure the faulted topography and terraces. A PHANTOM 4 real-time kinematic (RTK)
Figure 4. Geological and geomorphological maps of the Tuanjiegacha site (Site 2). (a) Interpreted map of the Tuanjiegacha geomorphology surface. (b and c)
photograph and sketch interpretation of fault outcrops (Rao etal.,2016). (d) Terrace cross section.
Tectonics
LIANG ET AL.
10.1029/2021TC006845
6 of 29
Figure 5. Geological and geomorphological interpretation map of the Dongwugaigou site (Site 3). (a) Geomorphic
interpretive map of the Dongwugaigou site. (b) One trench has been excavated on the low terrace (see a). (c) Cross-sectional
view of the terrace near the trench. The heights of the high terrace and low terrace from the basin are 15.3 and 3.3m,
respectively. (d and e) The Dongwugaigou trench reveals three earthquake events. A detailed stratum description of this
trench can be found in AppendixB1.
Tectonics
LIANG ET AL.
10.1029/2021TC006845
7 of 29
UAV was used to measure the microtopography. The device was equipped with a D-RTK 2 high-precision Global
Navigation Satellite Service (GNSS) mobile station, which allowed real-time differential measurements between
the aircraft and the local GNSS system or a RTK base station, and the measurement precision was on the order of
Figure 6. Geological and geomorphological interpretation of the Wubulangkou site (Site 4). (a) Interpreted Google Earth image of the Wubulangkou site. The width of
the terrace between the bedrock fault and the latest active fault at the front of the terrace is ∼500–800m. Two ground fissures are present in front of the terrace. (b) A
cross section of the terraces shows that P1 and P2 are ∼13.9 and 30.8m above the basin. (c) A fault and two ground fissures are present in the transition zone between
P1 and the basin. (d) The trench profile excavated perpendicular to the southern ground fissure reveals the existence of multiple faults. Marker layers M1 and M2 are cut
by multiple faults. (f) The trench reveals 10 faults, F1–F10, that were active after the formation of the lacustrine strata. A detailed stratum description of this trench can
be found in AppendixB2.
Tectonics
LIANG ET AL.
10.1029/2021TC006845
8 of 29
the centimeter level. Altizure software was used to precisely control the aircraft’s flight height, course, measure-
ment range, and real-time image transmission. The flight height was set to 80–110m to avoid ground disturbanc-
es and to acquire high-resolution images. The three-dimensional reconstructions were based on structure from
motion (SfM) algorithms. The SfM algorithms automatically detect feature points to be matched in disordered
digital images, optimize the mutual positions between the cameras and the target, and acquire three-dimensional
spatial information on the target (Johnson etal., 2014; Snavely etal., 2008; Uysal etal., 2015). In this study,
Pix4Dmapper software was used to process photographs and generate digital orthophoto map (DOM) and digital
elevation model (DEM) data for the measurement area. In this study, 6 sites were chosen to observe their geomor-
phological characteristics. Among them, Site 1, Site 2 and Site 3 are near the intersection of the LPF and WSPF,
and Site 4, Site 5 and Site 6 are near the intersection of the WSPF and ESPF (Figure1b).
3.2. Trench Excavation
Trenching is the most direct and effective method to verify the existence of faults, reveal the spatial parameters
of faults (such as strike, dip direction, and dip angle), and obtain the timing of earthquake occurrences, coseismic
displacement and elapsed time (Crone,1987; M.Liang etal., 2018; Ran etal.,2015; Wallace,1981; Yeats & Pren-
tice,1996). The trenching technique relies on three key assumptions: (a) the strata completely preserve geological
records of paleoearthquake events; (b) it is possible to correctly and completely identify those paleoearthquake
events; and (c) it is possible to collect samples that effectively constrain the age of these events (Deng etal.,2004;
Ran etal.,2012). To confirm the location, geometry, and latest activity of the faults, three trenches are excavated
in Dongwugaigou, Wubulangkou, and Delingshan. Among them, the Dongwugaigou Trench is located at the in-
tersection of the LPF and WSPF, and the Wubulangkou Trench and Delingshan Trench are located at the intersec-
tion of the WSPF and ESPF. According to the development characteristics of the normal fault scarp, the trenches
are excavated perpendicular to the scarp and from the middle of the scarp toward the basin. The dimensions of
the trenches were ∼20m long×6m wide×5m deep. Thus, the spatial orientation of the fault can be revealed,
and the strata on both sides of the fault can be exposed for comparison. At the same time, the strata and the fault
on both sides of the trench can be compared and verified. In this study, 21 trenches from previous studies and
2 trenches (Dongwugaigou trench and Delingshan trench) from this study are used to discuss the paleoseismic
sequences of the LPF, WSPF, and ESPF (TableA1).
Figure 7. Geological and geomorphological interpretations of the Shenhuabei site (Site 5), (a) Geomorphological interpretation of the Shenhuabei site. Faulted terraces
P1, P2 and P3 are observed. (b) P1, P2 and P3 are 5.2, 12.1 and 19m higher than the basin, respectively. (c) The fault plane was exposed where P1 was artificially
excavated. The lacustrine layer is exposed in the lower part of P1. (d) The abandoned fluvial fan has developed a steep linear ridge at the fault. (e) The cross section
shows a vertical offset of 2.85 m.
Tectonics
LIANG ET AL.
10.1029/2021TC006845
9 of 29
4. Results
4.1. Dating and Geochronology Constraints of the Geomorphic Terraces
To obtain the ages of the terraces and paleoseismics, we collected age data obtained by previous researchers in
this area and collected optically stimulated luminescence (OSL) samples and radiocarbon (14C) samples from the
sediments near the tops of the terraces and in the trenches. The OSL samples were analyzed at the Key Laboratory
of Crustal Dynamics (National Institute of Natural Hazards, Ministry of Emergency Management of China). The
radiocarbon (14C) samples were analyzed by the BETA laboratory in the United States. The results of the OSL
samples are given in Table1, and the results of the 14C samples are given in Table2.
From the results of the age samples obtained from the terraces in front of the mountain range, the formation
age of T1 is 8.32± 0.96–11.54 ± 1.23 ka, corresponding to the early Holocene. The formation age of T2 is
22.42±2.56–33.01±1.99ka, corresponding to the late Pleistocene. The formation age of T3 is 60.34±7.94–
79.50±9.67 ka, corresponding to the middle of the late Pleistocene. We did not obtain age data on T4. He
etal.(2015) found the OSL age of the T4 terrace in the Langshan area to be 113.33±12.31ka, corresponding
to the early late Pleistocene.
According to the genesis mechanism and distribution characteristics, the geomorphic surfaces in the study area
can be divided into two types: fault terraces P and flood terraces T. Fault terrace P refers to the geomorphic sur-
faces that result from the uplift and erosion of the fault scarp and are distributed along the strike of the fault. The
formation mechanism is shown in Figures2a–2c. The differential movement of the fault blocks on either side of
the fault forms a fault scarp that is distributed along the fault (Figure2a). After a reduction in tectonic activity, the
fault scarp is eroded by surface flow, forming P1 (Figure2b). When the fault becomes active again, P1 is uplifted,
and a new fault scarp is formed. The multiple phases of intermittent activity of the fault have formed a multilevel
stepping fault-related terrace on the ground surface that is nearly parallel to the fault (Figure2c). There are two
important things that need to be noted. (a) P1 and P2 are connected by a receding erosion ridge, not a fault. (b)
The formation age of the layer on top of P1 is older than that of P2 and cannot represent the formation age of P1.
Event no. Age (ka) TrenchaRupture segmentationbRupture length (km)cMagnituded
Eb1 2.0 (7 BC) 1, 2, 3, 6, 9, 10, and 18 L, W, and K 272 8.1
Eb2 3.11±0.25–3.59±0.42 1, 19, 20, and 22 L, K, and D 140 7.7
Eb3 3.64±0.29–4.41±0.03 1, 16, and 19 L, H, and K 132 7.7
Eb4 5.45±0.03–5.57±0.45 3, 8, 9, 12, 18, and 19 L, W, and K 272 8.1
Eb5 6.71±0.03–6.8±0.4 3, 4, 5, 8, 16, 20, and 23 L, H, and D 238 8.0
Eb6 8.2±0.4–8.36±0.64 7, 8, 11, 14, 20, and 23 L, W, and D 273 8.1
Eb7 10.45±0.32–10.55±0.82 18, 19, and 23 K and D 85 7.4
a“Trench” refers to the trench that reveals the earthquake event. The numbers of trenches are consistent with those in Figure11. The name of each trench can be seen in
Figure10a. bIn the “Rupture segmentation” column, we use abbreviations. L, LPF; W, Wujiahe segment; H, Hongqicun segment; K, Kuluebulong segment; D, Dashetai
segment. cThe “rupture length (km)” is calculated by the length of each segment. L, 160km; W, 70km; H, 35km; K, 42km; D, 43km. The earthquake event EW2 in
the LPF faulted ∼55km south of the LPF (Dong etal.,2018). d“Magnitude” is calculated from the “rupture length (km)” (Well & Coppersmith,1994).
Table 3
Results of the Paleoearthquake Comparison for the LPF, WSPF, and ESPF
Sanguibulong Dongwugaigou Tuanjiegacha Wubulangkou Shenhuabei Delingshan
T1 4.4 4.2 5.4
T2/P1 13.8 13.3 13.4 13.9 5.2 13.5
T3 20.2 24.8 24.8
T4/P2 38 46.5 52.8 30.8 12.1
Note. Here, “height” refers to the height of the terraces relative to the riverbeds and the unit is in m.
Table 4
The Height of the Flood Terraces/Fault Terraces at Each Site
Tectonics
LIANG ET AL.
10.1029/2021TC006845
10 of 29
In this case, it may be better to choose the exposure age of cosmogenic nuclide samples. The flood terrace T refers
to the remaining landform surfaces resulting from earlier fluvial fans that have since been eroded by the gully.
They are often distributed in steps on both sides of the gully. The development process of the four-level flood
terraces T1–T4 in the study area is shown in Figures2d–2h. When the fault is in a calm period, floods flow out
of the mountain pass, and the velocity decreases, resulting in sediment deposition and the formation of a fluvial
fan (Figure2d). When the fault enters an active period, the previously accumulated fan is uplifted and eroded by
water flow. As a result, the fan-shaped body is destroyed, and only the fan root remains as a flood accumulation
Figure 8. Geological and geomorphological features of the Delingshan site (Site 6). (a) The eastern section of the SPF intersects the western section at the
Wubulangkou site (Site 4); (b and c) north of Delingshan, two ground fissures are present at the front of the terrace and cut the man-made farmland ridges and a cement
road. (d) Two scarps developed at the ground fissures, which are 8 and 13.5m above the basin. (e) A trench excavated perpendicular to the southern ground fissure (He
etal.,2017). (f) Unit 3 and underlying strata are obviously faulted by F1–F4. A detailed stratum description of this trench can be found in AppendixB3.
Tectonics
LIANG ET AL.
10.1029/2021TC006845
11 of 29
terrace (Figure2e). According to the dating results (Table1), the abandonment ages of T4–T1 in the study area
were ∼113.33±12.3, 66.72±8.03, 23.97±3.45, and 9.79±1.44ka BP, respectively.
The study area features two fault terraces and four flood terraces (Figure2i). Among them, flood terraces T1 and
T3 are distributed on both sides of the gully. The elevations of flood terrace T2 and fault terrace P1 are compa-
rable, as are those of flood terrace T4 and fault terrace P2. They form two-level joint geomorphic surfaces in the
area. Here, a joint geomorphic surface refers to two spatially connected planes: the plane left by erosion of the
fault scarp and the deposition fan surface. When the fault is reactivated, the two are raised to form a comparable
height plane (Zhang, Molnar, & Xu,2007). The characteristics of the joint geomorphic surface include horizon-
tal changes in lithology. Taking the joint geomorphic surface formed by P1 and T2 as an example, terrace T2 is
composed mainly of sediments with coarse grains, while P1 comprises an upper part with an alluvial sandy gravel
layer and a lower part composed of yellow-green lacustrine deposits. The age of the joint geomorphic surface can
be determined based on the age of sediments at the top of T2 or the exposure age of gravels at the surface of P1
and T2. Because the incision rate of the gully flow is greater than the erosion rate of surface flow and the retreat
rate of the fault scarp, flood terraces are more sensitive to intermittent fault activity, and the number of levels of
Figure 9. Fault distribution at the intersection between the western section of the Seertengshan piedmont fault (WSPF) and the eastern section of the Seertengshan
piedmont fault (ESPF). (a) The distribution of bedrock faults. The pre-Q WSPF passes through Wubulangkou and continues to the ESE. The fault connects to the ESPF
by a central bend with a width of 12km. (b) The distribution of the latest active fault. The ESPF passes through Mabozi and continues to the NW. The Kuluebulong
segment is an overlapping fault area. (c) Section A–A′, which is perpendicular to the fault, reveals a faulted lacustrine layer (in yellow), and the fault activity
migrates from Fc to Fb and Fa. Fa, Fb, and Fc refer to the ESPF, WSPF, and pre-Q WSPF, respectively. (d) Topographic profile B–B′ of the terrace front edge of the
Kuluebulong segment shows that the terrace elevation gradually decreases from SE of Mabozi to the NW and reaches the same elevation as the basin at Delingshan.
Tectonics
LIANG ET AL.
10.1029/2021TC006845
12 of 29
flood terraces is often more than that of fault terraces distributed along the fault in an area with active normal
faults, such as T1 and T3 (Figure2i; Zhang, Molnar, & Xu,2007).
4.2. Geological and Geomorphological Characteristics of the LPF and WSPF Intersection
To study the geological features, tectonic styles, and evolution of the intersection of the LPF and WSPF, we se-
lected three sites for detailed geomorphological measurements, namely, the Sanguibulong site (Site 1), Tuanjieg-
acha site (Site 2), and Dongwugaigou site (Site 3). The positions of these three sites are shown in Figure1b. The
Sanguibulong site (Site 1) is located in the northeastern part of the LPF. The Tuanjiegacha site (Site 2) is located
in the western part of the WSPF. The Dongwugaigou site (Site 3) is located at the intersection of the NE-trending
LPF and the E-W-trending SPF.
4.2.1. Geological Features of the Sanguibulong Site (Site 1)
The Sanguibulong site (Site 1) is located in the northeastern part of the LPF, ∼20km from the Dongwugaigou
site (intersection point of the LPF and WSPF; Figure1b). The LPF passes through this site and has a strike
of 40°–50° and dips to the SE, which is consistent with the characteristics of the whole LPF, as described by
previous researchers (Dong,2016; Rao etal.,2016). There are two gullies that are ∼100m wide perpendicular
to the fault (Figure3a). The new sediments are transported through the two gullies and deposited to form two
large alluvial fans. The eastern fan is well preserved, while the western alluvial fan features a large trench due to
Figure 10. Paleoearthquake information of the LPF, WSPF, and ESPF (detailed information can be found in AppendixA). (a) Shows the location of the 23
trenches. Among them, 21 trenches are from previous studies, and the Dongwugaigou trench and Delingshan trench are from this study. (b) The occurrence ages of
paleoearthquakes since 25ka are compiled. The Y-axis is the age and the X-axis is the distance to Dongwugaigou.
Tectonics
LIANG ET AL.
10.1029/2021TC006845
13 of 29
artificial sand mining, and three good fault profiles are exposed in the trench wall. The gullies contain four levels
of terraces: T1, T2, T3, and T4, which are 4.4, 13.8, 20.2, and 38m higher than the riverbed, respectively. Two
faulted terraces, P1 and P2, are developed along the fault. P1 and T2 are the same elevations, and P2 and T4 are
the same elevations.
There is a ground fissure on P1, and it is parallel to the fault. The mound left by tree planting in ∼2010 has been
faulted (Figure3c). According to He etal.(2017), these ground fissures are caused by overexploitation of ground-
water, causing the strata to crack along the pre-existing fault. In addition, a fault spring is present at the junction
point of the fault and gully (Figure3d), and water flows out from the spring. The strata in the northwestern,
northern and eastern walls of the trench in the west alluvial fan are all faulted by multiple faults. The northwest-
ern profile has a height of more than 20m, and many nearly parallel faults ruptured the strata into domino style
blocks. The faulted strata dip toward the footwall. The dip angle of these strata is ∼5° in the upper part and ∼17°
in the lower part (Figure3e). In the northern wall of the sand pit, a number of collapse wedges appear near the
main fault and are the result of multiple events (Figure3f). In the eastern wall of the sand pit, multiple faults cut
each other, which may represent different slip events (Figure3g).
4.2.2. Geological Features of the Tuanjiegacha Site (Site 2)
The Tuanjiegacha site (Site 2) is located at the western end of the SPF (Figure1b), where the fault strikes nearly
E-W. The interpretation result of the UAV survey shows that five terraces have developed (Figure4a). The heights
of T1∼T5 from the basin are 5.4, 13.4, 27.8, 52.8, and 78.1m, respectively. T2 and T4 are widely distributed in
the area, corresponding to faulted terraces P1 and P2, respectively. At 1.5km west of Fangrong, a gully cuts T2
(P1), revealing a stepping normal fault profile. According to Rao etal.(2016), the three stepping faults F1, F2,
and F3 are all south-dipping normal faults. The age of 14C samples obtained from Unit 4 was 1,880±30. Rao
etal.(2016) conclude that the last activity of the fault occurred after 1,880±30yr BP.
Figure 11. Paleoearthquake comparison of the LPF, WSPF, and ESPF (detailed information can be found in Table3). (a–g) Shows the rupture segment (red line),
reveal trench (red bar), and calculated magnitude of Eb1–Eb7. (h) Shows the age of Eb1–Eb7. The ages of Eb1 and Eb4–Eb7 are well constrained. (i) Linear regression
by the well-constrained ages, which shows that the recurrence of paleoearthquakes conforms to a periodic model with a period of 1.37±0.11ka.
Tectonics
LIANG ET AL.
10.1029/2021TC006845
14 of 29
Figure 12. Sketch maps show the tectonic evolution model of the intersection of the LPF and WSPF. (a) In the early stage (pre-Quaternary), the pre-Q fault activity
formed a bedrock fault cliff. (b) Fault activity migrated to the basin, forming bedrock-terrace-basin stepping terrain. At the intersection of the LPF and SPF, two left-
stepping small fault segments developed. (c) The fault segments linked together, and the LPF and WSPF formed a large arc-shaped fault. FS, future sediments.
4.2.3. Geological Features of the Dongwugaigou Site (Site 3)
The Dongwugaigou site (Site 3) is located at the intersection of the WSPF and LPF (Figure1b). The NE-striking
LPF and the E-W-striking WSPF intersect here, and the angle between them is ∼145°. Two large gullies have
developed here, and each of them has developed four terraces (Figure5a). The heights of T1–T4 from the basin
are 4.2, 13.3, 24.8, and 46.5m, respectively. T2 and T4 are continuously distributed in the area, corresponding to
faulted terraces P1 and P2, respectively.
We chose to dig a trench in the middle of the two gullies, as shown in Figure5a. Two faulted terraces, the high
terrace and low terrace, have developed at this location. The high terrace is P1, which is continuously distributed
at the front of the mountain range and is ∼15.3m higher than the basin. The low terrace, which is ∼3.3m higher
than the basin, is located in front of the high terrace P1. The Dongwugaigou trench was excavated perpendicular
to the low terrace.
The section of the east wall of the trench exposes four normal faults that strike nearly E-W and dip to the south.
Three paleoearthquake events are revealed (Figures5d and5e). Event 1 was the slip of fault F1, which faulted
layer ⑥ and the layers below with an offset of ∼1.25m and formed large collapse wedge 1 (layer ⑦). Event 2
was the slip of faults F2, F3, and F4. These three faults all faulted layer ⑨, with offsets of 1.6, 1.5, and 0.53m,
and formed collapse wedge 2, collapse wedge 3, and collapse wedge 4 (layer ⑩), respectively. The age of the 14C
sample taken from cover layer ⑪ was 5,450±30 BP, indicating that the time of Event 2 was slightly earlier than
5,450±30 BP. Event 3 was the slip of faults F1 and F2. They both faulted layer ⑭, with offsets of 0.63 and 0.5m,
Tectonics
LIANG ET AL.
10.1029/2021TC006845
15 of 29
Figure 13. Sketch maps show the tectonic evolution model of the intersection of the WSPF and ESPF. (a) In the early stage (pre-Quaternary), the pre-Q fault activity
formed a bedrock fault cliff. The pre-Q WSPF was connected to the pre-Q ESPF with a central bend. (b) Both the WSPF and the ESPF migrated toward the basin,
forming bedrock-terrace-basin stepping terrain. The ESPF propagated northwestward, and the activity of the WSPF decreased northwestward. A triangular relay ramp
formed in the overlapping area. (c) The triangular relay ramp will eventually be destroyed, and the WSPF and ESPF will form a large arc-shaped fault. FS, future
sediments.
Tectonics
LIANG ET AL.
10.1029/2021TC006845
16 of 29
respectively. The age of the 14C sample taken from layer ⑭ was 2,440±30 BP; thus, the time of Event 3 should
be slightly later than 2,440±30 BP. A detailed stratal description of this trench can be found in AppendixB1.
There is a large difference in elevation between the basin and the mountain, reaching 300–400m, and the high-
est elevation of the mountain is 1,410m. The mountain body is composed of late Paleozoic intrusive rocks, and
pre-Quaternary (pre-Q) faults exist between the mountain and the terraces. Figure5a shows that the pre-Q SPF
extends in an E-W direction until it intersects with the pre-Q LPF. However, our trench indicates that the latest ac-
tivity of the fault displaced to the surface and formed the low terrace at the front edge of terrace P1. Figure5a shows
that the WSPF has migrated to the basin since the late Quaternary and gradually transitions stepwise into the LPF.
4.3. Geological and Geomorphological Features at the Intersection Between the Western and Eastern
Sections of the SPF
4.3.1. Geological Features of the Wubulangkou Site (Site 4)
The Wubulangkou site (Site 4) is located at the eastern end of the WSPF, where the ESPF meets the WSPF (Fig-
ure1b). Bedrock, terrace, and basin are present at the surface from north to south. A bedrock fault exists between
the bedrock and the terrace, and an active fault exists between the terrace and the basin (Figure6a). The width
of the terrace is ∼500–800m. The terrace can be roughly divided into P1 and P2, which are ∼13.9 and 30.8m
above the basin, respectively. The upper part of P1 is a sandy gravel deposit, and the lower part is a yellow-green
lacustrine deposit (Figure6b).
Two ground fissures are developed at the front of P1. A trench was excavated perpendicular to the southern ground
fissure (Figures6d–6f). The trench revealed the existence of multiple faults. The faults are mostly south-dipping
normal faults or reverse faults, and their cross-cutting pattern is the result of multiple slip events. Marker layer M1,
which is a brick red gravelly clay layer, crosses the entire trench section and is faulted or folded by multiple faults.
In the middle of the trench, due to the movement of two faults with opposite dips, a horst was formed, and the yel-
low-green lacustrine layer was exposed. The lacustrine layer is consistent with the “Jilantai-Hetao megalake” sedi-
ment distributed along the fault. According to previous studies, the top of the lacustrine layer has an age of ∼65ka
(Chen, Fan, Chun, Madsen, Oviatt, Zhao, etal.,2008; Chen, Fan, Chun, Madsen, Oviatt, Zhao, & Yang,2008;
Zhang etal.,2017). The two sand veins in the trench are considered to be sandy soil liquefaction due to intense
shaking of the water-saturated sand layer during the occurrence of a large earthquake. Two-stage collapse wedges
(Unit 9 and Unit 10) were also found in the trench. Unit 10 is cut by faults, indicating at least three events. All these
phenomena indicate that the fault has been seismically active since the middle of the late Pleistocene.
4.3.2. Geological Features of the Shenhuabei Site (Site 5)
The WSPF continues eastward after passing through the Wubulangkou site (Site 4), and the Shenhuabei site
(Site 5) is located ∼2km east of the Wubulangkou site (Figure1b). Three faulted terraces, P1, P2, and P3,
developed along the fault, and they are 5.2, 12.1 and 19m higher than the basin, respectively (Figures7a and
7b). In addition, there is a fluvial fan. The fluvial fan has been displaced by the fault, leaving a clear linear fault
scarp (Figure7d). Section e, which is perpendicular to the fault scarp, shows that the vertical offset of the scarp
is 2.85m (Figure7e).
4.3.3. Geological Features of the Delingshan Site (Site 6)
The Delingshan site (Site 6) is located north of Delingshan town at the northwestern end of the ESPF (Figures1b
and8a). From northeast to southwest, there is bedrock, a wide terrace, and the basin. The width of the terrace
reaches 5km. Two faults and ground fissures exist in the front edge of the terrace. The ground fissures have
faulted man-made farmland ridges and a cement road (Figures8b and 8c), which indicates that the formation
time of ground fissures occurred within recent decades. He etal. (2017) suggested that ground fissures formed
due to the excessive use of groundwater, which caused shrinkage of the layers and movement along an existing
fracture surface. We used the RTK-GPS technique to measure the heights of the two scarps at 8.0 and 13.5m. One
trench excavated perpendicular to the southern ground fissure revealed four faults, F1–F4, all of which appear to
be normal faults that dip to the southwest (Figures8e and8f). Among them, F2 and F3 reach the ground surface,
and F2 forms a ground fissure. The age of the OSL sample from Unit 1 is 1.51±0.20ka, indicating that the latest
activity of faults F2 and F3 occurred later than 1.51±0.20ka BP.
Tectonics
LIANG ET AL.
10.1029/2021TC006845
17 of 29
4.3.4. General Characteristics of the Intersection Between the ESPF and WSPF
Bedrock faults are distributed along the boundary between the piedmont bedrock and the Quaternary terrace.
Their traces are drawn based on field investigations, remote sensing images, and DEM data. The pre-Q WSPF is
oriented nearly E-W at Dahoudian-Wubulangkou and is parallel to the active piedmont fault, and the distance be-
tween them is less than 2km. The pre-Q WSPF passes through Wubulangkou and continues to the ESE. This fault
connects with the NW-striking pre-Q ESPF via a 12-km-wide central bend to the north of Mabozi (Figure9a). In
addition, the pre-Q ESPF has a similar central bend with a width of ∼3km at Wulanhudong (Figure9a).
The latest active faults are distributed along the front edge of the terraces (Figure9b). At the Hongqicun segment
(Dahoudian-Wubulangkou), the active fault at the front of the mountain range strikes nearly E-W and dips to
the south. Bedrock faults are nearly parallel to the active faults. Four to five gully terraces are developed within
a range of <2km, forming a narrow and steep stepping terrain. The Dashetai segment (Mabozi-Tailiang) is
almost the same as the Hongqicun segment. However, in the Kuluebulong segment (Wubalangkou-Wayaotan),
the WSPF passes through Wubulangkou and continues parallel to the bedrock fault, and its activity gradually
weakens. The ESPF continues northwest after passing Mabozi, and the height of the terrace decreases to the
northwest (Figure9d). The area of overlap between the two faults forms a triangular relay ramp. Additionally, a
similar 3-km-wide relay ramp developed near Wulanhudong.
Based on the borehole profile near Shuiquancun from the Hydrogeological Team at the Geological Bureau of the
Inner Mongolia Autonomous Region(1981), combined with observations in the field, a profile of the Shuiquan-
cun borehole is described (Figure9c). The profile reveals three faults, Fa, Fb, and Fc, and a faulted lacustrine
layer UL. The field investigation found that a thick lacustrine profile appeared in Shuiquancun, which is the lacus-
trine strata of the Jilantai-Hetao megalake that are distributed along the fault. According to previous studies, the
top of the lacustrine strata has an age of ∼65ka (Chen, Fan, Chun, Madsen, Oviatt, Zhao, etal.,2008; Chen, Fan,
Chun,Madsen, Oviatt, Zhao, & Yang,2008; Zhang etal., 2017). Zhang etal.(2017) calculated the vertical slip
rate of the fault near Shuiquancun based on the vertical throw of UL across fault Fa (∼61.1m) and obtained a val-
ue of 0.94mm/a. Fc cuts Neogene stratum N but not Quaternary stratum Q, making it a pre-Quaternary fault. Fb
cuts the lower part of the late Pleistocene stratum Q3 but not the upper part of Q3, indicating that it is an early late
Pleistocene fault. Fa cuts Q3 and Holocene stratum Q4, indicating that it is a Holocene fault (Figure9c). We found
in the field survey that terrace T3 is continuously distributed on both sides of the gully toward the mountains. Fb
and Fc do not cut T3. The OSL age of T3 is ∼60–70ka, which is almost the same as the age of the Jilantai-Hetao
megalake. Thus, the geomorphology shows that after the formation of T3 (∼60–70ka), faults Fb and Fc did not
rupture to the surface. Therefore, we can conclude that Fc was active early (before the Quaternary), Fb was active
from the early Quaternary to the early Pleistocene, and Fa was active during the mid-late Pleistocene to Holocene.
Fault activity migrated from Fc to Fb in the early Quaternary to Fa in the mid-late Pleistocene.
5. Discussion
5.1. Paleoearthquake Information of the LPF, WSPF and ESPF
Extensive research on the segmentation and paleoearthquakes of the LPF and SPF has been performed (Chen,
Ran, & Chang,2003, 2003b; Dong et al.,2018; He etal., 2017, 2018; Li etal.,2015; Rao et al.,2016; Yang
etal.,2002,2003). Dong etal.(2018) revealed that there have been at least three seismic events that ruptured the
whole LPF since the Holocene. Therefore, they conclude that the LPF is a nonsegmented active fault that tends
to rupture the entire section during great earthquake events. The SPF is considered to be divided into the Wuji-
ahe segment, Hongqicun segment, Kuluebulong segment, and Dashetai segment, with lengths of 70, 35, 42, and
43km, respectively (Chen, Ran, & Chang,2003; He etal.,2017,2018; Long etal.,2017). We collected data from
23 trenches excavated in this area and compiled the paleoearthquake information in AppendixA and Figure10.
The location of the trenches is shown in Figure10a. In Figure10b, we set the Y-axis as age and the X-axis as the
distance between each trench and Dongwugaigou. Thus, Figure10b shows the paleoseismic information of each
fault segment since 25ka.
Eight trenches were collected for the LPF. Among them, Shabagen, Xibulong, Tuanjie, and Dongsheng-1 are
from Rao etal.(2016), the Bayinwula trench comes from Li etal.(2015), and the Wulanhashao trench, Qingshan
trench, and Dongsheng-2 trench are from Dong etal.(2018). Through comparison, six paleoearthquake events
Tectonics
LIANG ET AL.
10.1029/2021TC006845
18 of 29
(EL1–EL6) occurred at 2.43±0.03–2.3±0.3ka (EL1), 3.06±0.03–4.41±0.03ka (EL2), 5.6±0.6–6.6±0.4ka
(EL3), 6.71±0.03–6.8±0.4ka (EL4), 8.2±0.4–9.81±0.03ka (EL5), and 16.7±1.2–22.2±1.7ka (EL6). Five
of these six events are consistent with those described in Dong etal.(2018). EL3 was ignored in Dong etal.(2018).
However, EL3 can be seen in the Wulanhashao trench, with an occurrence age of 5.6±0.6–6.6±0.4ka, and in
the Dongsheng-2 trench, with an occurrence age of 5.1±0.3–6.8±0.4ka. This event can also be revealed by the
profiles of S2, S9, S3, and Tc2 in Sun etal.(2021) and constrained to 5.6±0.63–6.24±0.71ka. Therefore, we
think that EL3, like EL1, EL4, and EL5, ruptured the entire section of the LPF.
The Dahoudian salient divides the WSPF into the Wujiahe segment and Hongqicun segment (Chen, Ran, &
Yang,2003; He etal.,2018). The Dongwugaigou trench reveals two events (Figure5). Event ED1 occurred after
2.44±0.03ka, and ED2 occurred slightly earlier than 5.45±0.03 ka. As shown in Figure4, Rao etal. (2016)
indicated that the last activity in the Fanrong trench occurred after Unit 4 (1.88± 0.03ka). However, Unit 4 is
actually a fissure filling, which most likely formed after the earthquake. Moreover, faulted Unit 6 was dated at
2.05±0.03ka. Therefore, we constrain the occurrence age of EW1 from 1.88±0.03 to 2.05±0.03ka, which is
consistent with ED1 in the Dongwugaigou trench. EW2 can be constrained from 5.45±0.03 to 5.57±0.45ka
by the Dongwugaigou trench (from this study) and Fendoucun trench (from Wu etal.,1996). EW3 can be con-
strained from 7.67±0.59 to 8.41±0.67ka by the Alagaitu trench (Yang etal.,2003) and Lianfengsidui trench
(from Wu etal.,1996). EW4 is revealed by only the Alagaitu trench, with an occurrence age of 19.92± 1.47–
22.34± 1.76 ka. Therefore, four paleoearthquake events (EW1–EW4) were revealed by the six trenches in the
Wujiahe segment, with occurrence ages of 1.88±0.03–2.05±0.03ka (EW1), 5.45±0.03–5.57±0.45ka (EW2),
7.67±0.59–8.41±0.67ka (EW3), and 19.92±1.47–22.34±1.76ka (EW4). Among them, the occurrence times
of EW1, EW2, and EW3 were well constrained. In the Hongqicun segment, there were only two trenches, the
Heibanhu trench (Wu etal.,1996) and Wubulangkou trench (Yang etal.,2002). The Wubulangkou trench reveals
EH1 and EH2 with occurrence ages of 3.64± 0.29–4.65±0.35ka (EH1) and 5.93± 0.4.88±0.54ka (EH2).
Obviously, more trenches are needed to reveal the complete paleoearthquake sequence in the Hongqicun segment.
The ESPF has been divided into the Kuluebulong segment and Dashetai segment by Delingshan salient (He
etal.,2018; Long et al., 2017). The Delingshan trench (Figure 4; He etal.,2017), Dongshuiquan trench (He
etal.,2018), and Dongfengcun trench (Yang etal.,2002) are dug in the Kuluebulong segment. These three trench-
es reveal eight earthquake events (EK1–EK8), with occurrence ages of <1.5±0.2ka (EK1), ∼2.32±0.29ka
(EK2), 3.11 ± 0.25–3.56 ± 0.11 ka (EK3), 3.56 ± 0.11–4.53 ± 0.36 ka (EK4), ∼5.77 ± 0.11 ka (EK5),
10.45±0.30.55±0.82ka (EK6), ∼15.03±2.28ka (EK7), and ∼18.73±2.05ka (EK8). The Mabozi trench (He
etal.,2018), Hongmingcun trench (Chen, Ran, & Chang,2003), Wulanhudong 1 trench (He etal.,2018), and
Wulanhudong 2 trench (Chen, Ran, & Chang,2003) are dug at the Dashetai segment. These four trenches reveal
four earthquake events (ED1–ED4), with occurrence ages of <3.59±0.42ka (ED1), 6.08±0.47–8.36±0.64ka
(ED2), 10.41±0.82–12.91±1.01ka (ED3), and 13.92±1.11–15.22±1.22ka (ED4).
We compared the paleoseismic information of each segment in Table3 and Figure11. EK1 (<1.5±0.2ka) in
the Kuluebulong segment is the latest earthquake event. The Delingshan trench, which reveals EK1, is excavat-
ed across the ground fissure. According to He etal.(2017), ground fissures formed due to the excessive use of
groundwater, which caused shrinkage of the layers and movement along an existing fracture surface. There were
no corresponding surface fracture events in the other segments. Therefore, we prefer to interpret that EK1 was not
formed during a great earthquake. However, it may be related to moderate earthquakes in this area.
Many important historical documents in China, such as Han Shu (Volume 27) Wuxingzhi (Anonymous,1986),
Imperial Readings of the Taiping Era (Li,1960), and the general history of China’s disasters (Qin and Han dynas-
ties volume; Yuan,2009), recorded that an M 8 earthquake occurred in the Hetao Basin on November 11, 7 BC
(the second year of Suihe, Han Dynasty). This great earthquake damaged more than 30 cities north of Xi'an city,
killed more than 415 people (Ban,1962), and affected a large area, including Gansu, Ningxia, Shanxi, Inner Mon-
golia, Hebei, and Beijing (Cao,1987). According to the analysis in He etal.(2018), this earthquake occurred on
the Kuluebulong segment of the SPF. However, an M 8.0 earthquake was unlikely to cause only 45km of ground
rupture. Figure10 shows that the latest earthquake event of the LPF is constrained from 2.43±0.03 to 2.3±0.3ka
(EL1) by four trenches. The latest earthquake event of the Wujiahe segment revealed by the Dongwugaigou trench
(Figure5 in this study) and Fanrong (Rao etal.,2016) is constrained from ∼1.88±0.03 to 2.05±0.03ka (EW1).
Moreover, penultimate event EK2 in the Kuluebulong segment occurred slightly <2.39±0.39 ka. Considering the
dating error, the times of EL1, EW1, and EK2 are very close, and they are more likely to be the same event. The
Tectonics
LIANG ET AL.
10.1029/2021TC006845
19 of 29
time of this event can be further restricted to 2.0–2.05ka, which is highly consistent with the 7 BC earthquake in
historical records. This event is revealed by 7 trenches in Figure10: Shabagen (Rao etal.,2016), the Bayinwula
trench (Li etal., 2015), the Wulanhashao trench (Dong etal.,2018), Tuanjie (Rao etal.,2016) in the LPF, the
Dongwugaigou trench (this study), Fanrong (Rao etal., 2016) in the Wujiahe segment, and the Dongshuiquan
(He etal.,2018) trench in the Kuluebulong segment. Therefore, the earthquake that occurred on November 11, 7
BC, faulted the whole LPF, Wujiahe segment of the WSPF and Kuluebulong segment of the ESPF, with a surface
rupture length of 272km (Figure11a; Table3). According to the empirical relationship between the earthquake
moment magnitude (M) and the surface rupture length (L) of active normal faults in regions of extension, the
magnitude of this earthquake was calculated to be M 8.1 (Well & Coppersmith,1994).
According to the same method, we calculated a total of seven earthquakes with surface ruptures of the LPF-SPF
since the Holocene. The age, revealed trenches, rupture segments, rupture length, and calculated magnitude of each
earthquake are shown in Table3 and Figure11. The ages of Eb1 and Eb4–Eb7 are well constrained, while those of
Eb2 and Eb3 are not (Figures11h). We use these well-constrained ages to perform linear regression (Figures11i).
The results show that the coefficient of variation in the linear fitting is 0.08, which means that the paleoearthquake
conforms to the periodic model. The age of Eb2 is close to 3.11ka, and the age of Eb3 is close to 4.41ka. The recur-
rence period of a large earthquake was 1.37±0.11ka. Figures11a–11g also shows that most of the surface ruptures
from great earthquakes in this area are composed of the LPF and part of the WSPF or ESPF. That is, the activities
of the LPF, WSPF, and ESPF are not completely independent. When the rupture crosses the northwest corner or
northeast corner to form a joint rupture in a single large earthquake, there is the potential for an M 8.0 earthquake.
It has been ∼2,000yr since the 7 BC earthquake, which far exceeds the recurrence period of earthquakes in this
area. Therefore, the Hetao Basin currently has structural conditions for the occurrence of an M 7.4–8.1 earthquake.
Nie(2013), based on the scattered tiles in the ancient tombs of the Han Dynasty in Machi, Baotou, and the trenches
excavated in front of Daqingshan Mountain, believed that the epicenter of the earthquake in 7 BC may be in Machi,
Baotou, and the seismogenic fault is the Daqingshan piedmont fault (DPF). Moreover, the DPF is also the seismogen-
ic fault of the 849 AD earthquake (Nie etal.,2010). Therefore, it may be difficult to distinguish whether the scattering
of these tiles and the deformation of the strata in the trenches were caused by the 7 BC earthquake or the 849 AD
earthquake. The intensity lines of most large earthquakes are roughly parallel to the strike of the seismogenic fault,
such as the Wenchuan Mw 7.9 earthquake (Xu etal.,2014). The intensity lines of the 849 AD earthquake are roughly
parallel to the DPF (Nie etal.,2010). However, the seismic damage range lines of the 7 BC earthquake recorded in
the literature are arcs that bulge northwest and are almost parallel to the LPF-SPF (Nie,2013). Combined with the
results of this study, it is more reasonable to regard the LPF-SPF as the seismogenic fault of the 7 BC earthquake.
The depth contours of the bases of the Cenozoic, Quaternary and late Pleistocene are drawn as blue, pink, and
gray lines and are shown in FigureC1. First, the maximum thickness of the deposits in the Linhe Basin since the
Cenozoic has reached 12.2km, and the deposition center is located in Qingshanzhen in the middle of the LPF.
Second, the maximum thickness of sediments since the Quaternary is ∼2,400m, and the deposition center is lo-
cated in Wugaizhen in the northeastern LPF. Third, the maximum deposition thickness since the late Pleistocene
is ∼280m, and the deposition center is near Urad Houqi in the northeastern LPF. The subsidence centers of the
Cenozoic, Quaternary, and late Pleistocene stages of the Linhe Basin are all located in the middle-northern part
of the LPF. At the same time, from the distribution of depth contours, the farther the distance from the deposition
centers is, the smaller the sediment thickness will be. There is just one subsidence center in the LPF at each stage,
and no deposition center formed in the SPF. Therefore, it is reasonable to infer that when large earthquakes occur,
the Linhe Basin may tilt toward the northwest as a whole at depth. Large earthquakes may cause joint rupture of
the LPF and one or more segments of the SPF.
5.2. Tectonic Evolution Model of the Intersection of the LPF and WSPF
The tectonic evolution of faults (or segments) is of great interest to structural geologists (e.g., Duffy etal.,2015;
Fossen & Rotevatn,2016; Frankowicz & McClay,2010; Nixon etal.,2014; Zhang etal.,1991). Many scholars
have discussed the geometric and kinematic relationships of two abutting faults (Fossen & Rotevatn,2016; Giba
etal., 2012; Imber etal., 2004; Long & Imber,2012; Peacock etal.,2016, 2017). Salients, barriers, and relay
ramps are considered to be the most common connected structures (Fonseca,1988; Tsukuda,1991). What are the
relationships between the LPF and WSPF and between the WSPF and ESPF? How did the connection structure
between them evolve?
Tectonics
LIANG ET AL.
10.1029/2021TC006845
20 of 29
The terraces formed at different times can invert the tectonic evolution process well. We classified and compared
the geomorphic surfaces at different sites (Table4). From the comparison results, at the intersection of the LPF
and WSPF, the heights of T1, T2/P1, and T3 are basically the same as those at Sanguibulong, Dongwugaigou, and
Tuanjiegacha, which are 4.2–5.4, 13.4–13.8, and 20.2–24.8m, respectively. We can infer that since 66ka (aban-
donment age of T3), although the fault shows stepping fault segments at the Dongwugaigou site, it has the same
throw rates as the faults at the other two sites. However, at the intersection of the WSPF and ESPF, P1 has a much
lower height at the Shenhuabei site (5.2m) than at the Wubulangkou site and Delingshan site (13.5–13.9m). In
addition, we can see from Figure9c that Fb (referring to WSPF) does not offset the top of the lacustrine stratum,
which indicates that Fb has not been active here since 65ka. In summary, after the WSPF passes the Wubulang-
kou site, the activity decreases rapidly to the east. However, at the Delingshan site, the height of P1 is consistent
with that at the Wubulangkou site, indicating that the faults at these two points have the same uplift rates. There-
fore, in the Kuluebulong section where the faults overlap, the activity of the WSPF is gradually decreasing, and
it is still active only near Shenhuabei. The majority of fault activity occurs along the ESPF. From Figure9d, the
topographic profile of the terrace front edge indicates that the ESPF gradually increases in elevation from west to
east and now reaches Delingshan without being connected to the WSPF.
The northwestern margin of the Ordos Block is generally an extensional environment with extension oriented
to the NW to NNW (Deng etal.,1999; Fan etal.,2003; Qu etal.,2017; Zhao etal.,2016). Inversions based on
existing focal mechanism solutions, hydraulic fracturing, borehole collapse, and fault slip data from this area re-
vealed that the extension direction in the area is ∼319°–139° (He etal.,2017,2020; K. Liang etal.,2018,2019;
Zhang etal.,2017). The strike directions of the LPF, WSPF, and ESPF are 45°, 90°, and 120°, respectively. The
acute angles between the strike directions of these three faults (45°, 90°, and 120°) and the extension direction
(319°–139°) are 86°, 49°, and 19°, respectively. Therefore, the strike direction of the LPF is nearly perpendicular
to the extension direction, and the LPF behaves mainly as a normal fault. However, the angles between the strike
directions of the WSPF and ESPF and the extension direction are small, which may produce horizontal strike-slip
components and cause the WSPF and ESPF to act as sinistral transtensional strike-slip faults (Zhang etal.,2017).
The tectonic evolution process at the intersection of the LPF and WSPF in Dongwugaigou is shown by a sketch
map in Figure12. The field investigation found that the bedrock fault plane developed at the contact surface
between the bedrock and the Cenozoic sediments and cut the N and E strata and was covered by Q strata (Fig-
ure9c), indicating that the fault represents a pre-Quaternary fault. Therefore, we infer that in the early stage of
fault activity, pre-Q fault activity caused mountain uplift and basin subsidence, forming a bedrock fault cliff and
resulting in sedimentation in the basin (Figure12a). Under the continuous action of extensive horizontal stress,
accommodation space tends to form near the fault plane, and a new, steeper fault will form. Since then, fault
activity has occurred on the new fault plane, and the old fault plane has been abandoned. Fault activity gradually
migrated to the side of the basin.
The Sanguibulong site (Site 1) shows that the LPF has a strike of ∼40°–50° (Figure3), and the Tuanjiegacha site
(Site 2) shows that the WSPF has a strike of nearly 90° (Figure4). The Dongwugaigou site (Site 3) shows two
left-stepping small fault segments developed at the intersection of the LPF and the WSPF at present (Figures5
and12b). Therefore, the fault segment is a manifestation of the growth stage of the fault formation. With the
continuation of extensional faulting, the fault segments at the intersection will link together to form an arc fault.
At that time, the LPF and SPF will connect into one large normal fault (Figure12c).
Table4 shows that the heights of T2 and T1 are the same at the Sanguibulong site (Site 1), Dongwugaigou site
(Site 2), and Tuanjiecun site (Site 3), which indicates that the small fault segments have been growing since at
least 24ka. The Dongwugaigou trench was excavated on the eastern small fault segment (Figure5a). The results
of the trench show that during the latest major seismic event (7 BC) and the penultimate major seismic event
(Figures5e, 5f,10b, and 11a), the small fault segment ruptured. This result indicates that although these small
fault segments are not connected to one larger fault, they rupture with the LPF and SPF during a large earthquake.
5.3. Tectonic Evolution Model of the Intersection of the WSPF and the ESPF
The tectonic evolution process at the intersection of the WSPF and the ESPF is shown by sketch maps in Figure13.
The towering bedrock fault cliffs and bedrock fault planes observed in the field in front of the mountain indicate
Tectonics
LIANG ET AL.
10.1029/2021TC006845
21 of 29
that in the early stage (pre-Quaternary), the fault slipped along the bedrock fault planes between the bedrock and
Cenozoic sediments (Figure13a). After passing the Wubulangkou site (Site 4), the pre-Q WSPF continued to extend
eastward, gradually deflecting to the southeast to Mabuzi and connecting to the northwest to the pre-Q ESPF via a
central bend (Walker etal.,2010). According to the model proposed by Mildon etal.(2016), at this time, the slip-
ping vectors of the WSPF, ESPF, and the central bend were consistent; that is, the hanging wall of the fault slipped
down overall. The continuous sliding of the fault caused both the ESPF and the WSPF to migrate toward the basin.
The ESPF gradually propagated straight northwest from Mabuzi, and the activity of the WSPF in the Kuluebulong
segment gradually decreased and ceased activity northwestward (Figure13b). The angle between the NW-striking
ESPF and extension direction is only 19°. This result indicates that the ESPF may have a sinistral strike-slip compo-
nent. Perhaps this horizontal motion component caused it to quickly propagate northwestward after passing Mabuzi
and maintain very good linearity. The present geomorphological observations show that the WSPF is still active at
the Shenhuabei site (Site 5, Figure7), and the height of terrace P1 at the Shenhuabei site is obviously lower than that
at the Wubulangkou site (Site 4, Figure6; Table4). The Dongshuiquan trench revealed that the latest large earth-
quake (7 BC) caused a surface rupture of the Kuluebulong segment (Figure10; He etal.,2018). Figure8a shows
that the fault propagated to the vicinity of Delingshan. The topographic profile of the Kuluebulong segment shows
that this area is a triangle with a length of ∼40km and a maximum width of ∼12km (Figure9d). That is, the inter-
section of the WSPF and the ESPF forms a triangular relay ramp (Conneally etal.,2014; Fossen & Rotevatn,2016).
The continued activity of the ESPF will cause the fault to propagate northwestward until it intersects the WSPF.
At that time, the WSPF and ESPF connect to form a large arc-shaped fault (Figure13c). After that, the part of the
WSPF east of the intersection will be abandoned and cease to be active, and the triangular relay ramp formed earlier
will also be abandoned (Long & Imber,2012; Peacock & Sanderson,1994). This result indicates that the newly
generated fault will shortcut the pre-existing faults. This situation is somewhat similar to how the newly generated
Daliangshan fault zone shortcuts the Xianshuihe-Xiaojiang fault system on the central segment (He etal.,2008).
6. Conclusion
A detailed investigation of the geological and geomorphological characteristics and paleoseismic information
collected from the northern boundary faults of the Linhe Basin (LPF, WSPF, and ESPF) yielded the following
conclusions:
1. Twenty-three trenches were collected to compile the paleoseismic sequence of each segment of the LPF,
WSPF, and ESPF. Seven of these trenches revealed historical large earthquake events in 7 BC. The length
of the rupture reached 272 km, which was estimated to be M 8.1. The trenches revealed seven paleoseis-
mic events since the Holocene in the Hetao Basin. These events obey the periodic model with a period of
1.37±0.11ka. Since the elapsed time of the latest event (2.0ka) has exceeded the recurrence period of the
earthquake, the area is currently at risk for an M 7.4–8.0 earthquake.
2. Two levels of fault terraces (P1 and P2) and four levels of flood terraces (T1–T4) are present in the study area.
Flood terrace T2 with fault terrace P1 and flood terrace T4 with fault terrace P2 form two joint geomorphic
surfaces in the area. The consideration of the genesis mechanisms and distribution characteristics of geomor-
phic surfaces is an important factor for comparing the surface topography along faults.
3. The NE-striking LPF and E-W-striking WSPF are connected by two left-stepping small fault segments. Al-
though these small fault segments are not connected to one larger fault, they might rupture together with the
LPF and WSPF during large earthquakes.
4. The E-W-striking WSPF and NW-striking ESPF are connected by a large triangular relay ramp. According to
the tectonic evolution trend of the fault in the Kuluebulong segment, the newly generated fault will shortcut
the pre-existing faults.
Appendix A
The faults and the segments the trenches belong to, the peloeseismics they revealed and their locations are offerd.
The corresponding information is plotted in Figure10.
Tectonics
LIANG ET AL.
10.1029/2021TC006845
22 of 29
Fault Segmentation Trench name Events Age (ka) Limitation Distance Location
LPF Shabagen (Rao etal.,2016) E5a 4.41±0.03 ↑ Older Lim. −112.48
3.06±0.03 ↓ Younger Lim. −112.48
E5b 2.43±0.03 ↑ Older Lim. −112.48
Bayinwula (Li etal.,2015) EB1 2.0±0.4 ↓ Younger Lim. −93.6 40°44′24.00″N,
106°28′48.00″E
2.7±0.4 ↑ Older Lim. −93.6
Wulanhashao (Dong etal.,2018) EW1 2.3±0.3 ↓ Younger Lim. −87.87 40°47′31.4″N,
10632′51″E
2.5±0.2 ↑ Older Lim. −87.87
EW2 5.6±0.6 ↓ Younger Lim. −87.87
6.6±0.4 ↑ Older Lim. −87.87
EW3 6.6±0.4 ↓ Younger Lim. −87.87
15.26±0.06 ↑ Older Lim. −87.87
Xibulong (Rao etal.,2016) E1a 6.71±0.03 ↓ Younger Lim. −62.49
7.27±0.03 ↑ Older Lim. −62.49
Qingshan (Dong etal.,2018) EQ1 6.6±0.2 ↓ Younger Lim. −56.74 405733.90N,
1064855.90E
6.9±0.3 ↑ Older Lim. −56.74
Tuanjie (Rao etal.,2016) E3a 2.64±0.03 ↑ Older Lim. −48.35
Dongsheng-1 (Rao etal.,2016) E2a 9.81±0.03 ↑ Older Lim. −37.52
E2c 4.79±0.03 ↓ Younger Lim. −37.52
Dongsheng-2 (Dong etal.,2018) ED1 5.1±0.3 ↑ Younger Lim. −33.53 41447.50N,
107211.00E
6.8±0.4 ↑ Older Lim. −33.53
ED2 8.2±0.4 ↓ Younger Lim. −33.53
10.0±1.6 ↑ Older Lim. −33.53
ED3 16.7±1.2 ↓ Younger Lim. −33.53
22.2±1.7 ↑ Older Lim. −33.53
WSPF Wujiahe Seg. Dongwugaigou (This study,
Figure5)
ED1 2.44±0.03 ↑ Older Lim. 0 41167.63N,
1072122.16E
ED2 5.45±0.03 ↓→ Slightly Younger Lim. 0
Site 4, Fanrong (Rao etal.,2016) E4a 1.88±0.03 ↓ Younger Lim. 6.44
2.05±0.03 ↑ Older Lim. 6.44
Alagaitu (Yang etal.,2003) EA1 23.28±1.75 ↓ Younger Lim. 40.38
27.69±2.24 ↑ Older Lim. 40.38
EA2 19.92±1.47 ↓ Younger Lim. 40.38
23.28±1.75 ↑ Older Lim. 40.38
EA3 7.67±0.59 ↓ Younger Lim. 40.38
22.34±1.76 ↑ Older Lim. 40.38
Fendoucun (Wu etal.,1996) EF1 2.81±0.22 ↓ Younger Lim. 52.93
5.57±0.45 ↑ Older Lim. 52.93
EF2 17.93±1.43 ↑ Older Lim. 52.93
Wujiahe (Wu etal.,1996) EW1 7.35±0.59 ↑ Older Lim. 61.18
Lianfengsidui (Wu etal.,1996) EL1 8.41±0.67 ↑ Older Lim. 64.39 41172.36N, 10879.00E
Hongqicun Seg. Heibanhu (Wu etal.,1996) EH1 12.87±1.03 ↓ Younger Lim. 92.84
Wubulangkou (Yang etal.,2002) EW1 7.09±0.55 ↓ Younger Lim. 97.7
EW2 7.09±0.55 ↑ Older Lim. 97.7
EW3 5.93±0.47 ↓ Younger Lim. 97.7
6.88±0.54 ↑ Older Lim. 97.7
EW4 4.65±0.35 ↑ Older Lim. 97.7
EW5 3.64±0.29 ↓ Younger Lim. 97.7
Table A1
Paleoseismic Information of the LPF, WSPF, and ESPF From 23 Trenches
Tectonics
LIANG ET AL.
10.1029/2021TC006845
23 of 29
Table A1
Continued
Fault Segmentation Trench name Events Age (ka) Limitation Distance Location
ESPF Kuluebulong Seg. Delingshan (This study, Figure8) ED1 1.51±0.2 ↑ Older Lim. 108.45 411541.32N,
1083722.08E
ED2 22.25±3.05 ↓ Younger Lim. 108.45
38.37±4.98 ↑ Older Lim. 108.45
Dongshuiquan (He etal.,2018) ED1 19.01±2.19 ↓ Younger Lim. 117.32 411220.69N,
1084649.15E
ED2 18.73±2.05 ↑→ Slightly Older Lim. 117.32
ED3 15.03±2.28 ↑→ Slightly Older Lim. 117.32
ED4 10.45±0.32 ↑→ Older Lim. 117.32
ED5 5.86±0.69 ↑→ Slightly Older Lim. 117.32
ED6 2.32±0.29 ↑→ Older Lim. 117.32
Dongfengcun (Yang etal.,2002) ED1 5.77±0.11 ↓ Younger Lim. 126.68
10.55±0.82 ↑ Older Lim. 126.68
ED2 5.77±0.11 ↑→ Slightly Older Lim. 126.68
ED3 3.56±0.11 ↓ Younger Lim. 126.68
4.53±0.36 ↑ Older Lim. 126.68
ED4 3.11±0.25 ↓ Younger Lim. 126.68
3.56±0.11 ↑ Older Lim. 126.68
Dashetai Seg. Mabozi (He etal.,2017) EM1 29.07±3.85 ↑ Older Lim. 143.03 41652.39N,
1085912.53E
EM2 9.83±1.17 ↑ Older Lim. 143.03
EM3 3.9±0.46 ↑ Older Lim. 143.03
Hongmingcun (Chen, Ran, &
Chang,2003)
EH1 28.23±2.23 ↓ Younger Lim. 146.55 410605.0N,
1090046.3E
35.15±2.74 ↑ Older Lim. 146.55
EH2 17.77±1.4 ↓ Younger Lim. 146.55
28.23±2.23 ↑ Older Lim. 146.55
EH3 13.07±1.03 ↓ Younger Lim. 146.55
17.77±1.4 ↑ Older Lim. 146.55
EH4 4.51±0.35 ↓ Younger Lim. 146.55
10.37±0.8 ↑ Older Lim. 146.55
Wulanhudong 1 (He etal.,2018) EW1 9.38±1.24 ↑ Older Lim. 170.3 40582.29N,
109152.21E
EW2 3.59±0.42 ↑ Older Lim. 170.3
Wulanhudong 2 (Chen, Ran, &
Chang,2003)
EW1 19.12±1.47 ↓ Younger Lim. 171.32 40°5753.07N,
1091519.4E
31.14±2.46 ↑ Older Lim. 171.32
EW2 13.92±1.11 ↓ Younger Lim. 171.32
15.22±1.22 ↑ Older Lim. 171.32
EW3 10.41±0.82 ↓ Younger Lim. 171.32
12.91±1.01 ↑ Older Lim. 171.32
EW4 6.08±0.47 ↓ Younger Lim. 171.32
8.36±0.64 ↑ Older Lim. 171.32
Note. In the limitation column, “↑ Older Lim.” refers to the sample age being older than the age of the event and “↓ Younger Lim.” refers to the sample age being
younger than the age of the event.
Tectonics
LIANG ET AL.
10.1029/2021TC006845
24 of 29
Appendix B
Stratigraphic descriptions of the trenches.
1. Detailed stratum description of the east wall of the Dongwugaigou trench (Figures5d and5e): ① The dark red-yel-
low silt layer has a high soil content and a small amount of fine silt. There are white calcium streaks in the layer.
② Light gray-white river facies with a gravel-bearing coarse sand layer, and the gravel content is 15%. ③ Light
gray-white medium sand layer with fine gravel, and the gravel content is 5%. ④ Dark red-yellow clay layer, with a
small amount of calcium, massive unbedding, and no gravel. ⑤ Brick red gravel layer: most of the gravel is granite
fragments, the gravel is poorly sorted and rounded, and the maximum particle size is 10cm. The vertical drop on
both sides of fault F1 at the top of this layer is 2m. The thickness of this layer on the footwall of F1 is 1.2m, and the
thickness on the hanging wall is significantly increased, <2.3m. ⑥ Brick red gravel layer, where the diameter and
content of gravel are obviously larger than those in layer ⑤. The top boundary of this layer is clear, and the vertical
drop on both sides of F1 is 1.88m. ⑦ Light yellow-gray medium-fine sand layer, containing soil blocks and fine
sand blocks; this layer is a wedge-shaped body with a maximum thickness of 1.2m near fault F1. This layer was
simultaneously broken by faults F2, F3, and F4. ⑧ Light gray-white fluvial facies in a gravel-bearing coarse sand
layer, with obvious bedding development. ⑨ The upper part of this layer is a white calcareous clay layer with high
calcium content, tight cementation, no bedding, and no gravel. The lower part of this layer is a light yellow-gray
medium-fine sand layer with a small amount of clay and fine gravels. This layer is faulted by F1, F2, F3, and F4,
with offsets of 0.63, 2.1, 1.5, and 0.53m, respectively. ⑩ Light yellow-gray gravel-bearing medium-coarse sand
layer. This layer is composed of three collapsed wedges on the hanging walls of F2, F3, and F4. The cementation is
loose and collapses easily. This layer may have formed during the same earthquake event. According to the dating
results of the DWGG-14C-01 sample, which was taken from overlying layer ⑪, the time of the earthquake was
slightly earlier than 5,450±30 BP. ⑪ Gray-yellow sandy soil layer, with poor bedding, and a developed lenticular
sandy gravel layer. ⑫ Light yellow-gray fine sand layer with medium-coarse sand blocks. ⑬ Dark yellow-gray
gravel-bearing clay layer. ⑭ Light yellow-gray loose sand layer. The offsets of the bottom face on both sides of F1
and F2 are 0.63 and 0.5m, respectively. The age of the DWGG-14C-2 sample obtained in this layer is 2,440±30
BP, indicating that the time of the last event was later than 2,440±30 BP.
2. Stratigraphic description of the east wall of the Wubulangkou trench (Figures6e and6f). Unit 1: gray-black sur-
face soil that contains plant roots. Unit 2: gray-yellow silty clay layer, with very fine particles and without gravel.
Unit 3: brick red gravel layer, gravel is subangular, gravel diameter 2–3cm, well sorted. Unit 4: red calcareous
clay layer, tightly cemented, with high gravel content at the bottom. Unit 5: light yellow-gray coarse sand layer
with a small amount of fine gravel. Unit 6: calcareous sandy gravel layer; the calcareous layer is slab-like, tightly
cemented, and intermittently distributed. Unit 7: light gray fluvial sand-gravel layer, poor sorting, gravel shows
directional arrangement. Unit 8: gray-yellow medium-fine sand layer with a small amount of fine gravel bands.
This layer is faulted by multiple faults arranged in a domino shape. Unit 9: gray-white gravel collapse wedge,
messy accumulation, large gravel with high content. Unit 10: a large collapse wedge, mainly composed of coarse
sand and gravel. There are some clumps of clay and fine sand. Calcareous nodules develop along the fault planes
of F7 and F8. The collapsed wedge was broken into two parts by a reverse fault. Unit 11: yellow-green lacustrine
facies strata, mainly composed of clay and silt fine sand, which is consistent with the widely distributed Jilan-
tai-Hetao megalake sediment. Many small faults developed in this layer.
3. Stratigraphic description of the Delingshan trench (Figures 8e and 8f; He etal.,2017). Unit 1: red-brown
topsoil that contains plant roots. Unit 2: white calcareous sand layer. Unit 3: light-red sand-gravel layer, hori-
zontal bedding, poorly rounded, poorly sorted, with a white calcareous clay layer. Unit 4: gray-white gravel
layer, horizontal bedding, poorly rounded, poorly sorted, with a maximum gravel size of 5cm. Unit 5: gray-
white fine sand layer with horizontal bedding, 6-cm-thick gravel layer in the middle, poorly rounded, poorly
sorted. Unit 6: light-red silty-fine sand layer, unstratified. Unit 7: coarse sand-gravel layer with cross bedding,
generally rounded, poorly sorted, with a maximum gravel size of 4cm. Unit 8: gray-black coarse sand layer
with inclined bedding, full depth unknown. Unit 9: gray-white fine sand layer, unstratified.
Appendix C
(a) The maximum deposition thickness since the late Pleistocene is approximately 280 m, and the deposition
center is near Urad Houqi in the northeastern LPF. (b) The maximum thickness of sediments since the Quaternary
is approximately 2,400 m, and the deposition center is located in Wugaizhen in the northeastern LPF. (c) The
Tectonics
LIANG ET AL.
10.1029/2021TC006845
26 of 29
maximum thickness of the deposits in the Linhe Basin since the Cenozoic has reached 12.2 km, and the deposi-
tion center is located in Qingshanzhen in the middle of the LPF.
Data Availability Statement
The topography data (SRTM data) are from the website (http://srtm.csi.cgiar.org/srtmdata/). Google image data
are from (http://earth.google.com/). We thank the anonymous reviewers and the editor for their helpful comments
on the manuscript.
References
Anonymous (1986). Han Shu (Vol. 27, p. 507). Shanhai Ancient Literature Press. (in Chinese).
Ban, G. (1962). Han Shu (Vol. 27, pp. 1454–1455). Zhonghua Book Company. (in Chinese).
Burbank, D. W., & Anderson, R. S. (2011). Tectonic geomorphology. Blackwell Science.
Cao, Z. X. (1987). Historical earthquake references of Beijing area (p. 1). Forbidden City Press (in Chinese).
Chen, F., Fan, Y., Chun, X., Madsen, D. B., Oviatt, C. G., Zhao, H., etal. (2008a). Preliminary research on Megalake Jilantai-Hetao in the arid
areas of China during the late quaternary. Science Bulletin, 53(11), 1725–1739. https://doi.org/10.1007/s11434-008-0227-3
Chen, F., Fan, Y., Chun, X., Madsen, D. B., Oviatt, C. G., Zhao, H., & Yang, L. P. (2008b). Preliminary study on the formation mechanism of the
“Jilantai-Hetao” Megalake and the lake evolution history in the Hetao region. Quaternary Sciences, 28(5), 866–873.
Chen, L. C., Ran, Y. K., & Chang, Z. P. (2003a). Characteristics of late Quaternar y faulting and paleoseismic events on the east of Delingshan
segment of the Sertengshan piedmont fault. Dizhen Dizhi, 25, 555–565.
Chen, L. C., Ran, Y. K., & Yang, X. (2003b). Late Quaternary activity and segmentation model of the Sertengshan piedmont fault. Earthquake
Research in China, 19, 255–265.
Cheng, S., Li, C., Yang, G., & Ren, D. (2006). The denudational-surface sequence and controls in the landscape development in the Langshan
Mountains-Seerteng Mountains, Inner Mongolia. Quaternary Sciences, 26(1), 99–107.
Commins, D., Gupta, S., & Cartwright, J. (2005). Deformed streams reveal growth and linkage of a normal fault array in the Canyonlands graben,
Utah. Geology, 33(8), 645–648. https://doi.org/10.1130/g21433ar.1
Conneally, J., Childs, C., & Walsh, J. J. (2014). Contrasting origins of breached relay zone geometries. Journal of Structural Geology, 58, 59–68.
https://doi.org/10.1016/j.jsg.2013.10.010
Cowie, P. A., Attal, M., Tucker, G. E., Whittaker, A. C., Naylor, M., Ganas, A., & Roberts, G. P. (2006). Investigating the surface pro-
cess response to fault interaction and linkage using a numerical modelling approach. Basin Research, 18(3), 231–266. https://doi.
org/10.1111/j.1365-2117.2006.00298.x
Crone, A. J. (1987). Introduction to directions in paleoseismology (Open-File Report 87-683). US Geological Survey Open File Report.
Deng, Q. D. (2007). Active tectonics map of China. Seismological Press.
Deng, Q. D., Chen, L. C., & Ran, Y. (2004). Quantitative studies and applications of active tectonics. Earth Science Frontiers, 11(4), 383–392.
Deng, Q. D., Cheng, S., Min, W., Yang, G., & Reng, D. (1999). Discussion on Cenozoic tectonics and dynamics of Ordos block. Journal of
Geomechanics, 5, 13–21.
Deng, Q. D., & Xu, X. W. (1994). Studies on the surface rupture zone of 1303 Hongdong earthquake of M = 8.0 and paleoearthquakes of Huoshan
fault in Shanxi province. Earthquake Research in China, 8(2), 231–245.
Dong, S. (2016). Late Quaternary tectonic activity and paleoseismology along the Langshan range-front fault (PhD thesis). Institute of Geology;
China Earthquake Administration.
Dong, S., Zhang, P., Zheng, W., Yu, Z., Lei, Q., Yang, H., etal. (2018). Paleoseismic observations along the Langshan range-front fault, Hetao
Basin, China: Tectonic and seismic implications. Tectonophysics, 730, 63–80. https://doi.org/10.1016/j.tecto.2018.02.012
Duffy, O. B., Bell, R. E., Jackson, C. A. L., Gawthorpe, R. L., & Whipp, P. S. (2015). Fault growth and interactions in a multiphase rift fault
network: Horda Platform, Norwegian North Sea. Journal of Structural Geology, 80, 99–119. https://doi.org/10.1016/j.jsg.2015.08.015
Fan, J. X., Ma, J., & Diao, G. L. (2003). Contemporary tectonic stress field around the Ordos fault block inferred from earthquake focal mecha-
nisms. Seismology and Geology, 25, 88–99.
Feng, X., Ma, J., Zhou, Y., England, P., Parsons, B., Rizza, M. A., & Walker, R. T. (2020). Geomorphology and paleoseismology of the Wein-
an fault, Shaanxi, Central China, and the source of the 1556 Huaxian earthquake. Journal of Geophysical Research: Solid Earth, 125,
e2019JB017848.
Fonseca, J. (1988). The Sou Hills: A barrier to faulting in the central Nevada seismic belt. Journal of Geophysical Research: Solid Earth, 93(B1),
475–489. https://doi.org/10.1029/jb093ib01p00475
Fossen, H., & Rotevatn, A. (2016). Fault linkage and relay structures in extensional settings—A review. Earth-Science Reviews, 154, 14–28.
https://doi.org/10.1016/j.earscirev.2015.11.014
Frankowicz, E., & McClay, K. R. (2010). Extensional fault segmentation and linkages, Bonaparte Basin, outer North West Shelf, Australia. AAPG
Bulletin, 94(7), 977–1010. https://doi.org/10.1306/01051009120
Gaudemer, Y., Tapponnier, P., Meyer, B., Peltzer, G., Guo, S., Chen, Z., etal. (1995). Partitioning of crustal slip between linked, active faults in
the eastern Qilian Shan, and evidence for a major seismic gap, the ‘Tianzhu gap’, on the western Haiyuan Fault, Gansu (China). Geophysical
Journal International, 120, 599–645.
Giba, M., Walsh, J. J., & Nicol, A. (2012). Segmentation and growth of an obliquely reactivated normal fault. Journal of Structural Geology, 39,
253–267. https://doi.org/10.1016/j.jsg.2012.01.004
He, H., Yasutaka, I., & He, Y. (2008). Newly-generated Daliangshan fault zone-shortcutting on the central segment of Xianshuihe-Xiaojiang fault
system. Science in China Series D:Earth Sciences, 38(5), 564–574.
He, X., Zhang, X., He, Z., Jia, L., Zhao, Q., Fu, L., & Ye, P. (2015). Development features of the Late Quaternary debris f low and their tectonic
significance in Langshan area, Inner Mongolia. Geological Bulletin of China, 34(9), 1735–1748.
He, Z., Ma, B., Hao, Y., Zhao, J., & Wang, J. (2020). Sur face rupture geomorphology and vertical slip rates constrained by terraces along the
Wulashan piedmont fault in the Hetao Basin, China. Geomorphology, 358, 107–116. https://doi.org/10.1016/j.geomorph.2020.107116
Acknowledgments
This work was supported by the National
Key Research and Development Program
of China (2018YFC1504100), research
grants from the National Institute of
Natural Hazards, Ministry of Emergency
Management of China (ZDJ2019-28), and
research grants from the National Natural
Science Foundation of China (Nos.
41872227 and 41602221). The OSL sam-
ples were dated by Dr. Junxiang Zhao at
the Key Laboratory of Crustal Dynamics,
China Earthquake Administration. The
high-resolution structure-from-motion
DEMs used in this paper are included in
the Open Science Framework (https://osf.
io/7g49m/).
Tectonics
LIANG ET AL.
10.1029/2021TC006845
27 of 29
He, Z., Ma, B., Long, J., Wang, J., & Zhang, H. (2018). New progress in paleoearthquake studies of the East Sertengshan piedmont fault, Inner
Mongolia, China. Journal of Earth Science, 29(2), 441–451. https://doi.org/10.1007/s12583-017-0937-z
He, Z., Ma, B., Long, J., Zhang, H., Liang, K., & Jiang, D. (2017). Recent ground fissures in the Hetao Basin, Inner Mongolia, China. Geomor-
phology, 295, 102–114. https://doi.org/10.1016/j.geomorph.2017.07.008
He, Z., Zhang, X., Jia, L., Wu, F., Zhou, Y., Bao, S., et al. (2014). Genesis of piedmont terraces and its neotectonic movement significance in
Langshan mountain area, Inner Mongolia. Geoscience, 28(1), 98–108.
Hopkins, M. C., & Dawers, N. H. (2015). Changes in bedrock channel morphology driven by displacement rate increase during normal fault
interaction and linkage. Basin Research, 27(1), 43–59. https://doi.org/10.1111/bre.12072
Hou, J. J., Han, M. K., Chai, B. L., & Han, H. Y. (1998). Geomorphological observations of active faults in the epicentral region of the Huaxian
large earthquake in 1556 in Shaanxi Province, China. Journal of Structural Geology, 20(5), 549–557.
Hydrogeological Team, Geological Bureau of Inner Mongolia Autonomous Region. (1981). Hydrogeological and soil chemical profile of Hetao
Plain, Bameng, Inner Mongolia. China Geological Data Administration.
Imber, J., Tuckwell, G. W., Childs, C., Walsh, J. J., Manzocchi, T., Heath, A. E., etal. (2004). Three-dimensional distinct element modelling of
relay growth and breaching along normal faults. Journal of Structural Geology, 26(10), 1897–1911. https://doi.org/10.1016/j.jsg.2004.02.010
Jia, L., Zhang, X., He, Z., He, X., Wu, F., Zhou, Y., etal. (2015). Late quaternary climatic and tectonic mechanisms driving river terrace devel-
opment in an area of mountain uplift: A case study in the Langshan area, Inner Mongolia, northern China. Geomorphology, 234, 109–121.
https://doi.org/10.1016/j.geomorph.2014.12.043
Jia, L., Zhang, X., Ye, P., Zhao, X., He, Z., He, X., etal. (2016). Development of the alluvial and lacustrine terraces on the norther n margin of the
Hetao Basin, Inner Mongolia, China: Implications for the evolution of the yellow river in the Hetao area since the late Pleistocene. Geomor-
phology, 263, 87–98. https://doi.org/10.1016/j.geomorph.2016.03.034
Jiang, W. (2002). Paleo-earthquake event and co-seismic vertical deformation recognition along the Langshan-Seertengshan Pediment Fault,
Inner Mongolia. Collections of Crustal Tectonics and Crustal Stress, 15, 45–52. (in Chinese with English abstract).
Jiang, W., Deng, Q., Xu, X., & Xie, X. (2004). Suface rupture zone of the 1303 Hongdong M8 earthquake, Shanxi province. Acta Seismology
Sinica, 26(4), 355–362.
Johnson, K., Nissen, E., Saripalli, S., Arrowsmith, J. R., McGarey, P., Scharer, K., etal. (2014). Rapid mapping of ultrafine fault zone topography
with structure from motion. Geosphere, 10(5), 969–986. https://doi.org/10.1130/ges01017.1
Kirby, E., & Whipple, K. X. (2012). Expression of active tectonics in erosional landscapes. Journal of Structural Geology, 44, 54–75. https://doi.
org/10.1016/j.jsg.2012.07.009
Lei, Q., Chai, C., Du, P. Y. , J., Wang, Y., & Xie, X. (2015). The seismogenic structure of the M8.0 Pingluo earthquake in 1739. Seismology and
Geology, 37(2), 413–429.
Li, F. (1960). Imperial readings of the Taiping era. Zhonghua Book Company. (in Chinese).
Li, K., & Nie, Z. (1987). Formation and evolution of Hetao fault depression and seismic activity.
Li, Y., Ran, Y., Chen, L. C., Wu, F., & Lei, S. X. (2015). The latest surface rupture events on the major active faults and great historical earth-
quakes in Hetao fault-depression zone. Seismology and Geology, 37(1), 110–125. https://doi.org/10.3969/j.issn.0253-4967.2015.01.009
Liang, K., Ma, B., Li, D., Tian, Q., Sun, C., He, Z., etal. (2019). Quaternary activity of the Zhuozishan West piedmont fault provides insight into
the structural development of the Wuhai Basin and Northwestern Ordos Block, China. Tectonophysics, 754, 56–72. https://doi.org/10.1016/j.
tecto.2019.02.004
Liang, K., Sun, C., Ma, B., Tian, Q., Li, D., Li, D., & He, Z. (2018). Investigation of the Yellow River buried fault in the Wuhai Basin, Northwest-
ern Ordos Block, China, using deep/shallow seismic reflection and drilling techniques. Journal of Asian Earth Sciences, 163, 54–69. https://
doi.org/10.1016/j.jseaes.2018.05.025
Liang, M., Ran, Y., Wang, H., Li, Y., & Gao, S. (2018). A possible tectonic response between the Qingchuan Fault and the Beichuan-Yingxiu
Fault of the Longmen Shan Fault zone? Evidence from geologic observations by paleoseismic trenching and radiocarbon dating. Tectonics,
37. https://doi.org/10.1029/2017TC004949
Liu, X., Zhang, X., Ye, P., He, Z., Fu, L., Jia, L., & He, X. (2016). Extraction of tectonic geomorphology parameters based in DEM and analysis
of difference on tectonic activity about Langshan Mountain, Inner Mongolia. Journal of Geomechanics, 22(1), 152–161.
Long, J. J., He, Z., Zhang, H., & Ma, B. (2017). Characteristics of structural geomorphology and segmentation of Sertengshan piedmont fault
from Dahoudian to Wayaotan. Geoscience, 31(1), 71–80.
Long, J. J., & Imber, J. (2012). Strain compatibility and fault linkage in relay zones on normal faults. Journal of Structural Geology, 36, 16–26.
https://doi.org/10.1016/j.jsg.2011.12.013
Maddy, D., Bridgland, D., & Westaway, R. (2001). Uplift-driven valley incision and climate-controlled river terrace development in the Thames
Valley, UK. Quaternar y International, 79(1), 23–36. https://doi.org/10.1016/s1040-6182(00)00120-8
Maddy, D., Bridgland, D. R., & Green, C. P. (2000). Crustal uplift in southern England: Evidence from the river terrace records. Geomorphology,
33(3–4), 167–181. https://doi.org/10.1016/s0169-555x(99)00120-8
Middleton, T. A., Walker, R. T., Parsons, B., Lei, Q., Zhou, Y., & Ren, Z. (2016). A major, intraplate, normal-faulting earthquake: The 1739
Yinchuan event in northern China. Journal of Geophysical Research: Solid Earth, 121, 293–320.
Mildon, Z. K., Roberts, G. P., Faure Walker, J. P., Wedmore, L. N. J., & McCaffrey, K. J. W. (2016). Active normal faulting during the 1997 seis-
mic sequence in Colfiorito, Umbria: Did slip propagate to the surface? Journal of Structural Geology, 91, 102–113. https://doi.org/10.1016/j.
jsg.2016.08.011
Nie, Z. (2013). Preliminary investigation on the historical M8 earthquake occurred in 7 BC at Baotou, Inner Mongolia. Acta Seismologica Sinica,
35(4), 584–603.
Nie, Z., Wu, W., & Ma, B. (2010). Surface rupture of the A.D. 849 earthquake occurred to the east of Baotou city, China, and discussion on its
parameters. Acta Seismologica Sinica, 32(1), 94–107.
Nixon, C. W., Sanderson, D. J., Dee, S. J., Bull, J. M., Humphreys, R. J., & Swanson, M. H. (2014). Fault interactions and reactivation within a
normal-fault network at Milne Point, Alaska. AAPG Bulletin, 98(10), 2081–2107. https://doi.org/10.1306/04301413177
Northrup, C., Royden, L., & Burchfiel, B. (1995). Motion of the Pacific plate relative to Eurasia and its potential relation to Cenozoic extension
along the eastern margin of Eurasia. Geology, 23, 719–722.
Ou, Q., Kulikova, G., Yu, J., Elliott, A., Parsons, B., & Walker, R. (2020). Magnitude of the 1920 Haiyuan earthquake reestimated using seismo-
logical and geomorphological methods. Journal of Geophysical Research: Solid Earth, 125. https://doi.org/10.1029/2019JB019244
Peacock, D. C. P., Nixon, C. W., Rotevatn, A., Sanderson, D. J., & Zuluaga, L. F. (2016). Glossary of fault and other fracture networks. Journal
of Structural Geology, 92, 12–29. https://doi.org/10.1016/j.jsg.2016.09.008
Tectonics
LIANG ET AL.
10.1029/2021TC006845
28 of 29
Peacock, D. C. P., Nixon, C. W., Rotevatn, A., Sanderson, D. J., & Zuluaga, L. F. (2017). Interacting faults. Journal of Structural Geology, 97,
1–22. https://doi.org/10.1016/j.jsg.2017.02.008
Peacock, D. C. P., & Sanderson, D. J. (1994). Geometry and development of relay ramps in normal fault systems. AAPG Bulletin, 78(2), 147–165.
https://doi.org/10.1306/bdff9046-1718-11d7-8645000102c1865d
Pezzopane, S. K., & Weldon, R. J. (1993). Tectonic role of active faulting in central Oregon. Tectonics, 12(5), 1140–1169. https://doi.
org/10.1029/92tc02950
Qu, W., Lu, Z., Zhang, M., Zhang, Q., Wang, Q., Zhu, W., & Qu, F. (2017). Crustal strain fields in the surrounding areas of the Ordos Block, cen-
tral China, estimated by the least-squares collocation technique. Journal of Geodynamics, 106, 1–11. https://doi.org/10.1016/j.jog.2017.01.005
Ran, Y., Chen, L., Yang, X., & Han, Z. (2003). Recurrence characteristics of late-quaternary strong ear thquakes on the major active faults along
the northern border of Ordos block. Science in China Series D: Earth Sciences, 46(2), 189–200.
Ran, Y., Chen, W. S., Xu, X. W., Chen, L. C., Wang, H., & Li, Y. B. (2015). Late Quaternary paleoseismic behavior and rupture segmentation
of the Yingxiu-Beichuan fault along the Longmen Shan fault zone, China. Tectonics, 33, 2218–2232. https://doi.org/10.1002/2014TC003649
Ran, Y., Wang, H., Li, Y., & Chen, L. C. (2012). Key techniques and several cases analysis in paleoseismic studies in mainland China (1): Trench-
ing sites, layouts and paleoseismic indicators on active strike-slip faults. Seismology and Geology, 34, 197–210. https://doi.org/10.3969/j.
issn.0253-4967.2012.02.001
Rao, G., Chen, P., Hu, J., Yu, Y., & Qiu, J. (2016). Timing of Holocene paleo-earthquakes along the Langshan piedmont fault in the western
Hetao Graben, North China: Implications for seismic risk. Tectonophysics, 677–678, 115–124. https://doi.org/10.1016/j.tecto.2016.03.035
Snavely, N., Seitz, S. M., & Szeliski, R. (2008). Modeling the world from internet photo collections. International Journal of Computer Vision,
80(2), 189–210. https://doi.org/10.1007/s11263-007-0107-3
Starkel, L. (2003). Climatically controlled terraces in uplifting mountain areas. Quaternary Science Reviews, 22(20), 2189–2198. https://doi.
org/10.1016/s0277-3791(03)00148-3
State Seismological Bureau (SSB). (1988). Active fault system around Ordos massif. Seismological Press.
Sun, A., Hu, X., & Niu, S. (1990). Geological characteristics of active structures in the Langshan area, Inner Mongolia . Journal of Hebei Institute
of Geology, 13(1), 27–34.
Sun, C., Li, D., & Li, L. (2021). The paleoearthquakes of the Langshan piedmont fault, north China since late Pleistocene. Natural Hazards
Research, 1(2021), 48–62. https://doi.org/10.1016/j.nhres.2021.06.005
Sun, J. (2005). Long-term fluvial archives in the Fen Wei Graben, central China, and their bearing on the tectonic history of the India Asia
collision system during the Quaternary. Quaternary Science Reviews, 24(10–11), 1279–1286. https://doi.org/10.1016/j.quascirev.2004.08.018
Tapponnier, P., Peltzer, G., & Armijo, R. (1986). On the mechanics of the collision between India and Asia. Collision Tectonics: Special Publi-
cation of the Geological Society of London, 19, 115–157.
Tapponnier, P., Peltzer, G., Le Dain, A., Armijo, R., & Cobbold, P. (1982). Propagating extrusion tectonics in Asia: New insights from simple
experiments with plasticine. Geology, 10, 611–616.
Tsukuda, E. (1991). Geometrical barriers on faults and their rupturing processes. Journal of Geography, 100(3), 417–428. https://doi.org/10.5026/
jgeography.100.3_417
Uyeda, S., & Kanamori, H. (1979). Back-arc opening and the mode of subduction. Journal of Geophysical Research, 84, 1049–1061.
Uysal, M., Toprak, A. S., & Polat, N. (2015). DEM generation with UAV Photogrammetry and accuracy analysis in Sahitler hill. Measurement,
73, 539–543. https://doi.org/10.1016/j.measurement.2015.06.010
Walker, J. P. F., Roberts, G. P., Sammonds, P. R., & Cowie, P. (2010). Comparison of earthquake strains over 102and 104year timescales: In-
sights into variability in the seismic cycle in the central Apennines, Italy. Journal of Geophysical Research, 115(B10), B10418. https://doi.
org/10.1029/2009jb006462
Wallace, R. E. (1981). Active faults, paleoseismology, and earthquake hazards in the western United States. In D. W. Simpson, & P. G. Richards
(Eds.), Earthquake prediction-an international review (pp. 209–216). American Geophysical Union, Maurice Ewing Series.
Wang, L., Dong, R., Zhang, Y., Wang, X., & Guo, W. (1984). Cenozoic structures and earthquakes in the Hetao area: Implications for the forma-
tion of the Cenozoic rifted basins around the Ordos Block. North China Earthquake Sciences, 2(4), 8–16.
Well, D. L., & Coppersmith, K. J. (1994). New empirical relationships among magnitude, rupture length, rupture width, rupture area, and surface
displacement. Bulletin of the Society of America, 84(4), 974–1002.
Wen, X. (2014). Structure of source regions of the 1979 Ms 6.0 Wuyuan earthquake and the 1996 Ms 6.4 Baotou earthquake in Inner Mongolia,
China. Seismology and Geology, 36(3), 586–597.
Whittaker, A. C., & Walker, A. S. (2015). Geomorphic constraints on fault throw rates and linkage times: Examples from the Northern Gulf of
Evia, Greece. Journal of Geophysical Research: Earth Surface, 120(1), 137–158. https://doi.org/10.1002/2014jf003318
Wu, W., Nie, Z., & Xu, G. (1996). Research on active faults in the western segment of Seertengshan piedmont fault. Active Fault Research, 5,
113–124.
Xu, C., Xu, X., Yao, X., & Dai, F. (2014). Three (nearly) complete inventories of landslides triggered by the May 12, 2008 Wenchuan Mw 7.9
earthquake of China and their spatial distribution statistical analysis. Landslides, 11(3), 441–461. https://doi.org/10.1007/s10346-013-0404-6
Yang, H., Yang, X., Cunningham, D., Hu, Z., Huang, X., Huang, W., et al. (2020). A regionally evolving transpressional duplex along the
northern margin of the Altyn Tagh Fault: New kinematic and timing constraints from the Sanweishan and Nanjieshan, China. Tectonics, 39,
e2019TC005749. https://doi.org/10.1029/2019TC005749
Yang, L. P. (2008). Reconstruction of Paloe-Megalake ‘Jilantai-Hetao’ based on remote sensing and DEM (PhD thesis). Lanzhou University.
Yang, X.-P., Ran, Y.-K., Hu, B., & Guo, W.-S. (2002). Active fault and paleoearthquakes of the piedmont fault (Wujumengkou-Dongfeng Village)
for Seerteng Mountains, Inner Mongolia. Earthquake Research in China, 18(2), 127–140.
Yang, X.-P., Ran, Y.-K., Hu, B., & Guo, W.-S. (2003). Paleoseismic activity on Wujiahe segment of Serteng piedmont fault, Inner Mongolia. Acta
Seismologica Sinica, 16(1), 67–78. https://doi.org/10.1007/s11589-003-0008-7
Yeats, R. S., & Prentice, C. S. (1996). Introduction to special section: Paleoseismology. Journal of Geophysical Research: Solid Earth, 101(B3),
5847–5853. https://doi.org/10.1029/95jb03134
Yuan, Z. L. (2009). The general history of China’s disaster (Qin and Han dynasties volume) (p. 82). Zhengzhou University Press. (in Chinese
with English Abstract).
Zhang, H., He, Z., Ma, B., Long, J., Liang, K., & Wang, J. (2017). The vertical slip rate of the Sertengshan piedmont fault, Inner Mongolia, China.
Journal of Asian Earth Sciences, 143, 95–108. https://doi.org/10.1016/j.jseaes.2017.04.014
Zhang, P., Molnar, P., & Xu, X. W. (2007). Late Quaternary and present-day rates of slip along the Altyn Tagh Fault, northern margin of the
Tibetan Plateau. Tectonics, 26, TC5010. https://doi.org/10.1029/2006TC002014
Tectonics
LIANG ET AL.
10.1029/2021TC006845
29 of 29
Zhang, P., Slemmons, D. B., & Mao, F. (1991). Geometric pattern, rupture termination and fault segmentation of the Dixie Valley—Pleasant Val-
ley active normal fault system, Nevada, U.S.A. Journal of Structural Geology, 13(2), 165–176. https://doi.org/10.1016/0191-8141(91)90064-p
Zhang, S., Ren, J., & Nie, G. (2007). Paragenesia of Quaternary pediments and river terraces on the north piedmont of Wutai Mountains. Chinese
Science Bulletin, 52(4), 521–530. https://doi.org/10.1007/s11434-007-0081-8
Zhang, W., Jiao, D., Zhang, P., Molnar, P., Burchfield, B. C., Deng, Q., etal. (1987). Displacement along the Haiyuan Fault associated with
the great 1920 Haiyuan, China, earthquake. Bulletin of the Seismological Society of America, 77(1), 117–131. https://doi.org/10.1785/
BSSA0770010117
Zhao, W., Hou, G., & Hari, K. R. (2016). Two episodes of structural fractures and their stress field modeling in the Ordos Block, northern China.
Journal of Geodynamics, 97, 7–21. https://doi.org/10.1016/j.jog.2016.02.005
