Table of Content

20 June 2026, Volume 48 Issue 3

Previous Issue   

Review

APPLICATION OF APPARENT RESISTIVITY OBSERVATION IN EARTHQUAKE PREDICTION

XIE Tao, LI Xin-yan, HAN Ying, YU Chen, LI Xiao-fan

2026, 48(3):  597-650.  DOI: 10.3969/j.issn.0253-4967.20240168

Asbtract ( )   HTML ( )   PDF (22927KB) ( )  

Figures and Tables | References | Related Articles | Metrics

Earthquake prediction based on changes in the resistivity of rock and soil media is a systematic endeavor that integrates field observation, scientific research, and forecasting practice. Apparent resistivity observation, which is widely used in earthquake monitoring and prediction in China, employs the direct-current method to continuously measure temporal changes in resistivity within a fixed subsurface investigation volume. Analytical calculations and numerical analyses of apparent resistivity observations conducted at the surface and underground have been developed by computing the potential distribution of the DC steady-current field. During the establishment of observation stations, electrical sounding should be carried out to invert the electrical structure of subsurface formations, and analytical calculations should be performed to determine the theoretical range of observational values, thereby verifying the correctness of the actual observation array configuration. Both analytical computation and numerical analysis can also be used to investigate the influence of resistivity variations in different parts of the medium on the observations. These methods provide reasonable explanations for three types of annual variation and for the characteristics of diurnal variation in apparent resistivity observations. In anomaly verification, they also enable quantitative evaluation of the interference amplitudes caused by leakage currents and electrically anomalous bodies. In the vertical direction, the one-dimensional sensitivity coefficient distribution reveals the extent to which resistivity changes at different depths affect the observations. The more detailed three-dimensional sensitivity coefficient distribution further reveals the differences in the effects of resistivity changes in different regions of the monitoring area on the observations. Therefore, the sensitivity coefficient distribution characterizes the local features of apparent resistivity observations and allows the influence of resistivity changes in different regions on the observations to be rapidly determined qualitatively. Petrophysical experiments and theoretical resistivity models have revealed the relationship between anisotropic resistivity decreases in water-bearing rock-soil media and the directional propagation of microcracks during compressive stress loading. In shallow subsurface layers, microcrack systems are predominantly aligned with the direction of the maximum horizontal principal compressive stress. During stress loading, the apparent resistivity measured in the direction perpendicular to the maximum principal compressive stress shows the largest variation amplitude, whereas that measured in the parallel direction shows the smallest variation, and that in an oblique direction exhibits an intermediate response. Through analysis of geoelectrical resistivity anisotropy, the orientation of the regional horizontal stress field can be inferred. The relative deformation accumulation around the epicenter before earthquakes has been analyzed using a virtual fault dislocation model. By using regional deformation as an intermediate bridge, a preliminary link has been established between anomalous changes at observation stations and the late-stage seismogenic process of distant earthquakes, and the expected patterns of apparent resistivity anomalies related to seismogenic processes have thus been obtained.

When observation stations are located within compressional enhancement zones during the seismogenic process, apparent resistivity exhibits anomalous decreases, whereas stations located in dilatational domains show anomalous increases. Transitional zones between compression and dilatation typically display weak or negligible anomalous variations. Based on earthquake case studies, persistent apparent resistivity anomalies lasting from several months to approximately two years serve as magnitude indicators for earthquake prediction, providing important constraints on potential epicentral locations and earthquake magnitudes. By contrast, short-impending anomalies characterized by accelerated variations or high-frequency perturbations act as temporal indicators and provide predictive information on earthquake timing. In operational earthquake forecasting, the spatial location and magnitude should first be preliminarily estimated using magnitude indicators, followed by short-term time prediction through continuous monitoring of temporal indicators within the pre-identified risk zone. On this basis, medium- and short-term earthquake predictions have been carried out from the perspective of interpreting field anomalies in terms of their seismic source processes. In recent years, the 2022 M_S6.8 Luding, 2023 M_S6.2 Jishishan, and 2024 M_S7.1 Wushi earthquakes were successfully predicted using the apparent resistivity method. However, the relevant research is scattered across domestic and international literature spanning more than 60 years. This paper therefore reviews and summarizes the main theoretical foundations of earthquake prediction based on apparent resistivity observations, so as to facilitate their understanding and application in geoelectrical earthquake prediction practice.

BEDROCK CHANNEL WIDTH IN ACTIVE OROGENIC BELTS: CONTROLLING FACTORS AND IMPLICATIONS FOR LANDSCAPE EVOLUTION

ZHANG Yi-hui, ZHANG Hui-ping, ZHAO Xu-dong

2026, 48(3):  651-684.  DOI: 10.3969/j.issn.0253-4967.20250002

Asbtract ( )   HTML ( )   PDF (5417KB) ( )  

Figures and Tables | References | Related Articles | Metrics

Fluvial erosion of bedrock channels plays a crucial role in controlling denudation rates in orogenic belts. Dynamic adjustment of bedrock channels in response to perturbations such as tectonic and climatic events controls the mechanisms and pace of regional geomorphic evolution. Fluvial systems generally evolve toward a steady state, adjusting their channel geometry to maintain a stable erosion rate. However, variations in tectonic activity, channel substrate strength, and sediment cover can cause fluvial systems to deviate from this equilibrium state. In response to those disturbances, fluvial systems modify their stream power by altering their geometry. Thus, channel morphology is vital for revealing the processes of geomorphic evolution and their control mechanisms. Stream power incision model is the base for quantitative studies of how fluvial systems adapt to tectonic, climatic, and lithological changes. Channel slope and channel width are the two most commonly used parameters to determine the state of channel stream power. Channel slope influences the rate of stream power dissipation along the downstream direction, while river width dictates the distribution of stream power across the channel bed. A reduction in channel width can concentrate stream energy, thereby promoting downcutting. Previous studies that have used fluvial geomorphology to infer tectonic and climatic information have largely focused on changes of channel slope, often overlooking the critical role of channel width. And the controlling factors and geomorphic indications of channel width, another important parameter of the hydraulic system, have not been fully revealed. Bedrock channel width is a key parameter for revealing the tectonic-erosional interaction and the geomorphologic evolution process, and has gradually become a hot spot in geomorphologic research in recent years. This paper addresses this gap by exploring how variations in channel width in response to external perturbations contribute to geomorphological evolution. In this paper, we provide a systematic review of hydraulic geometry modelling, erosion dynamics theory and global case studies, focusing on the factors controlling the width of bedrock channels(bed erodibility, tectonic deformation, sediment effects, and climate change, etc.). The synthesis of existing studies reveals that: 1)Bed erodibility is the core intrinsic factor controlling channel width, and rock mass strength and joint density regulate channel morphology by influencing the erosion mechanism(abrasion/abrasion): channel in high erosion-resistant bedrock zones tend to have narrower width, and higher steepness; on the other hand, weak or fractured rock layers promote lateral erosion and the formation of wide, shallow river valleys. 2)Tectonic deformation drives channel morphology through changes in uplift rate: rivers in high uplift rate areas concentrate shear forces by contracting width, accelerating downcutting to balance uplift, while rivers in low uplift areas weaken erosive capacity by widening the channel. 3) The dual effect of sediment supply significantly influences the response to width: adequate sediment cover of the streambed inhibits downcutting but enhances lateral erosion, whereas low sediment supply concentrates the energy on basement erosion, creating narrower and deeper channels. 4)Extreme climatic events trigger short-term adjustments in channel width through drastic changes in flood flow and sediment pulse inputs, and in the long term influence morphological evolution through changes in erosion datum. Thus, the current study still has the problems of an unknown multi-factor coupling mechanism and insufficient universality of the quantitative model. In the future, we need to combine high-resolution remote sensing, numerical simulation and chronological data to deepen the application of the width parameter in the geomorphological evolution model, so as to provide a new perspective for the study of tectonic-climatic interactions in orogenic zones.

Research paper

APATITE (U-TH)/HE DATING AND THE DETERMINATION OF STANDARD DURANGO

ZHU Shun-lin, YANG Rong

2026, 48(3):  685-703.  DOI: 10.3969/j.issn.0253-4967.20250001

Asbtract ( )   HTML ( )   PDF (2710KB) ( )  

Figures and Tables | References | Related Articles | Metrics

Apatite (U-Th)/He dating, a widely used low-temperature thermochronometer sensitive to shallow crustal processes within the uppermost ~3km of the crust, is critical for constraining tectonic, landscape, and basin thermal evolution. This study aims to validate the accuracy and reliability of the(U-Th)/He dating workflow at Zhejiang University’s thermochronology Laboratory through a systematic analysis of the international standard sample, Durango apatite. The core objective is to ensure the laboratory’s capability to generate precise age and Th/U ratio data consistent with global benchmarks, thereby providing a robust methodological foundation for low-temperature thermochronological studies focusing on near-surface cooling processes(closure temperature <90℃).

To achieve this goal, the laboratory conducted rigorous experimental analyses on a total of 60 Durango apatite grains across three batches. The specific procedures were as follows: First, Durango apatite particles with a 20-micron surface removed were crushed, and apatite fragments were screened under a polarizing microscope. These grains were encapsulated in platinum(Pt)capsules—chosen for their high thermal conductivity and chemical inertness—to facilitate efficient laser heating at 900℃ for 5 minutes and subsequent acid digestion. This single-grain laser heating approach enables complete and rapid degassing of helium from individual apatite crystals, minimizes the risk of cross-contamination between samples, and ensures a higher helium extraction efficiency compared to conventional bulk furnace heating methods. Helium(⁴He)extraction and analysis were performed using an Alphachron mass spectrometer equipped with a diode laser system, a high-vacuum maintenance system(including mechanical, turbo, and ion pumps), and a quadrupole mass spectrometer(QMS). The ⁴He concentration was quantified using the ³He isotope dilution method, with precise corrections for thermal blanks(empty Pt capsules) and line blanks(system background), which were found to be extremely low(average background value of 1.66×10^-12mL, only 1/1 248 to 1/29 280 of the sample signal value). For uranium(U) and thorium(Th) analysis, digested samples were measured by inductively coupled plasma mass spectrometry(ICP-MS). Using a spike solution with known U/Th isotopic ratios(²³⁵U/²³⁸U=838±7,²³⁰Th/²³²Th=10.45±0.5), elemental abundances and Th/U ratios were accurately calculated, with rigorous subtraction of reagent and Pt capsule background values.

The test data showed high consistency with international standards. The measured ages of the 60 Durango grains ranged from 28.48 to 33.20Ma, with an average age of(31.30±2.19)Ma(2σ), closely matching previously reported values such as(31.71±1.1)Ma(Wang et al., 2017)and(31.61±2.7)Ma(Wu et al., 2016). A normal probability plot of the age data revealed a tight distribution(R²=0.980 6) with a slope close to 1, indicating data concentration and high reliability. The Th/U ratios ranged from 17.10 to 26.13, overlapping with global datasets(e.g., 17.23-23.60, 19.2-21.8, 18.0-22.5. Although two grains exhibited slightly higher Th/U ratios(25.76 and 26.13), these values remained within the natural variation range reported in the literature(up to 34.5±3.9), confirming their geological plausibility. This minor Th/U dispersion aligns with the well-documented intra-grain trace element heterogeneity of Durango apatite, confirming our workflow preserves primary sample geochemical signatures without artificial fractionation.

The study concludes that Zhejiang University’s(U-Th)/He dating workflow for single apatite grains is accurate and reliable, with systematic uncertainties effectively controlled through strict background corrections and standardized operations. The extremely low ⁴He background values and precise Th/U measurements validate the laboratory’s capability for high-precision testing. The successful validation of Durango apatite as a reference sample not only demonstrates the laboratory’s contribution to global calibration standard data but also provides an important benchmark for quality control in similar laboratories.

Through testing the Durango apatite standard sample, this study validates the feasibility of the analytical workflow, ensures consistency with international datasets, and promotes the development of low-temperature thermochronology. The capability of Zhejiang University’s (U-Th)/He Chronology Laboratory to generate reliable (U-Th)/He age data positions it as a key participant in global thermochronological research, facilitating the resolution of fundamental questions about thermal and tectonic processes in the field of geosciences. Additionally, this validated standardized workflow supports domestic inter-laboratory comparison, promotes the standardization of (U-Th)/He dating in China, and provides reliable analytical support for the geoscience community.

SEASONAL VARIATIONS OF WAVE VELOCITY AND ITS INFLUENCE FACTORS ON ROCK SLOPE: APPLICA-TION TO THE PUBUGOU ROCK SLOPE IN SOUTHWEST CHINA

LI De-kang, XIE Fan, XIE Jun-ju, DAI Shi-gui, LI Li

2026, 48(3):  704-723.  DOI: 10.3969/j.issn.0253-4967.20250007

Asbtract ( )   HTML ( )   PDF (4217KB) ( )  

Figures and Tables | References | Related Articles | Metrics

Ambient seismic noise interferometry has been recently used to monitor the damage evolution of landslides. The measured velocity changes of slopes at shallow depth play a crucial role in better understanding the mechanical changes of slopes under different external loadings. However, previous studies have primarily focused on the effects of key triggers such as earthquakes and rainfall, or have been limited to analyzing the evolution mechanisms before and after slope instability. In contrast, long-term continuous monitoring of typical rock slopes in southwest China remains scarce, particularly in terms of exploring the quantitative relationships between external factors such as air temperature, rainfall, and seismicity, and seismic velocity changes. This limits a better understanding of how seasonal external factor variations affect slope stability.

Therefore, in this study, we conducted a 2.5-year in situ monitoring campaign of daily seismic velocity variations on the slow-moving Pubugou rock slope in southwestern China, using continuous ambient noise data recorded by two seismic stations deployed from October 1, 2020, to February 26, 2023. To assess the impact of long-term environmental fluctuations on slope stability, collocated meteorological and rainfall sensors were installed to record air temperature and precipitation at a daily sampling rate. Furthermore, we cataloged 27 446 local earthquakes(M_L1.0-6.8) occurring within a 250-km radius of the slope during the study period. Based on these integrated observations, we systematically investigated the relationships between seasonal seismic velocity changes and external forcing factors—specifically, air temperature, rainfall, and regional seismicity. The results show that the seismic velocity changes exhibit seasonal and reversible fluctuations in the range of -1.5% to 2.0% on an annual scale which is well linearly correlated with the changes in the air temperature, while the precipitation induced seismic velocity decreases by -0.3%~-0.6% during the monsoon. We find that the M_L6.8 Luding earthquake that occurred on September 5, 2022 induced a seismic velocity decrease of ~-2.3%, followed by a 23-day recovery phase to the pre-earthquake state by removing the influence of the seasonal air temperature fluctuation using a linear regression model. While the M_L5.6 aftershock occurred on January 26, 2023 induced a velocity decrease of ~-0.4% in the temperature-corrected time series of seismic velocity changes. Furthermore, we propose a seismic velocity-based coupling model on the slope by using the multiple linear regression method to quantify the external loadings, such as the air temperature, rainfall, and seismic activities. The optimal fitting coefficients for the temperature, rainfall-induced pore pressure, and earthquake-induced seismic intensity on the slope are 1.0×10^-3, -3.59×10^-5, and -6.68×10^-4, respectively. These coefficients suggest that the seasonal seismic velocity changes of the slope are positively correlated with the temperature and negatively correlated with the pore pressure and seismic intensity, as we expected. Finally, our results highlight the strong potential of the method to assess the risk of the slope instability by proposing an empirical warning index based on the analysis of a short-term decrease in the seismic velocity changes observed through long-term monitoring.

EVIDENCE OF THE HOLOCENE ACTIVITY IN THE NORTHERN SEGMENT OF YANTONGSHAN FAULT

LIU Chao, REN Zhi-kun, LIU Jin-rui, JI Hao-min, WU Zhi-qun, YU Si-han, LEI Qi-yun, DU Peng

2026, 48(3):  724-740.  DOI: 10.3969/j.issn.0253-4967.20250003

Asbtract ( )   HTML ( )   PDF (11340KB) ( )  

Figures and Tables | References | Related Articles | Metrics

The arcuate tectonic belt along the northeastern margin of the Qinghai-Xizang Plateau represents the leading edge of the plateau’s northeastward expansion. Since the Late Cenozoic, this region has accommodated intense tectonic activity, forming four major active fault zones aligned from south to north: the Haiyuan, Tianjingshan, Yantongshan, and Sanguankou-Niushoushan-Luoshan Faults. Situated between the Tianjingshan and Luoshan Faults, the Yantongshan Fault lies at the triple junction of the Qaidam, Alxa, and Ordos blocks.

While previous studies have extensively focused on the larger, more active Tianjingshan and Luoshan Faults, research on the Yantongshan Fault remains sparse. The Tianjingshan fault zone extends for approximately 240km and comprises the northern Tianjingshan Fault and the southern Miaoshan Piedmont Fault. The 1709 Zhongwei M7½earthquake ruptured this zone, which exhibits a late Quaternary left-lateral slip rate of 0.9~1.1mm/a. In contrast, the Luoshan Fault(~50km long)transitioned from thrusting prior to the mid-Late Pleistocene to right-lateral strike-slip motion in the late Late Pleistocene, with an average slip rate of 2.15mm/a.

Unlike its neighbors, the Yantongshan Fault lacks clear geomorphic expression, making it difficult to identify Holocene-active natural fault scarps. Consequently, direct evidence confirming Holocene activity on this fault remains lacking.

The Yantongshan Fault has a total length exceeding 150km and is naturally divided into southern and northern segments by Yaoshan. The northern segment is approximately 90km long, with about 15km concealed beneath the Zhongning Basin and the remaining 75km exposed at the surface. In this study, the northern segment of the Yantongshan Fault was taken as the research object. This study employs methods such as remote sensing interpretation, geological survey, aerial survey of typical geomorphology, trench excavation, and Optically Stimulated Luminescence(OSL)dating. The main findings are as follows: 1) The northern segment of the Yantongshan Fault has linear fault scarps. It is a low-angle reverse fault, dipping to the southwest with a dip angle of about 30°. 2) The geomorphic scarp near Xiaoyushugou in the northern segment of the Yantongshan Fault corresponds to this fault’s trace. During trench excavation, a new fault was identified and has been shown to be active during the Holocene. 3)OSL dating shows that the trench profile records at least two paleo-earthquake events. The latest earthquake event occurred after(7.99±0.42)ka in the Holocene, and at least one paleo-earthquake event occurred after(63.08±3.18)ka in the late Pleistocene. Considering the stratum erosion, the vertical displacements of the two events are at least 1.3m and 1.0m, respectively, and the lower limits of the earthquake magnitudes can reach 6.9 and 6.8. 4) The northern segment of the Yantongshan Fault may still be active at present. Its deep part, together with the Luoshan Fault, intersects the Tianjingshan Fault and ultimately merges with the Haiyuan Fault at the detachment layer. As a secondary fault associated with the Tianjingshan Fault, affected by the northeastward expansion of the Tibetan plateau, there is an abnormal zone of wave velocity and electromagnetism beneath the northern segment of the Yantongshan Fault, which is closely related to moderate-to-strong earthquakes, indicating that it still has seismic risk. Moreover, the longer the energy accumulation time, the higher the current risk. Referring to the average recurrence interval of earthquakes on the Tianjingshan Fault, which is (4 306±212)a, the latest strong earthquake(Event E1) on the Yantongshan Fault exposed by this trench occurred approximately 8 000a later. The elapsed time approaches or even exceeds the recurrence period of paleo-earthquakes, therefore, the seismic risk of its northern segment is urgent.

GRAVITY ISOSTASY AND DEEP SEISMOGENIC ENVIRONMENT IN THE MIDDLE SEGMENT OF THE TANLU FAULT ZONE

WANG Xin, GAO Min, QIAO Ji-hua, XIANG Jian-bin, XU Yu-chen, FAN Ji-di

2026, 48(3):  741-756.  DOI: 10.3969/j.issn.0253-4967.20240159

Asbtract ( )   HTML ( )   PDF (6826KB) ( )  

Figures and Tables | References | Related Articles | Metrics

The Tanlu fault zone(TLFZ)is an ancient intracontinental subduction zone and one of the largest active fault zones in eastern China. The 1668 Tancheng M_S8.5 earthquake occurred within the middle segment of the Tanlu fault zone(MTLFZ), making its deep geophysical structure and the deep seismogenic mechanism of large earthquakes a long-standing focus of research. In this study, high-precision Bouguer gravity data from the MTLFZ and adjacent regions were collected and inverted using the directional derivative method and the Parker-Oldenburg method to determine the distribution of major faults and the Moho depth. The actual crustal thickness was subsequently calculated by combining the Moho depth with digital elevation model(DEM)data, and the isostatic anomaly(Ⅲ)between the calculated crustal thickness and the Airy isostatic crustal thickness was obtained. Based on these results, the deep structural characteristics and deep seismogenic environment of the 1668 Tancheng earthquake were analyzed.

The results show that the Bouguer gravity field can be divided into one fault structural zone and five tectonic blocks, including the MTLFZ(Ⅰ), Luxi Uplift(Ⅱ), Xuhuai Block(Ⅲ), Jiaoliao Block(Ⅳ), Sulu Orogen(Ⅴ), and Lower Yangtze Block(Ⅵ). The MTLFZ is characterized by a NNE-trending high-gravity anomaly zone and a gentle gravity gradient zone, with gradient values generally≤0.6mGal/km. The east-west directional derivative of the gravity field produced the dx map, which reveals nearly NS-, NNE-, and NW-trending faults, including the Sunshidian Fault, Jiaxiang Fault, Tanlu fault zone, Tongzhidian-Sunzu Fault, and Xintai-Mengyin Fault. The north-south directional derivative generated the dy map, which identifies nearly EW-, NEE-, and NWW-trending faults, such as the Fushan Fault, Tiefogou Fault, Wulian-Rizhao Fault, Mengshan Piedmont Fault, and Cangni Fault. The vertical first derivative yielded the dz map, which highlights large-scale fault zones, including the Tanlu fault zone, Haisi Fault, Huaiyin-Xiangshuikou Fault, and Hongze-Goudun Fault. In total, 35 major faults and their intersection relationships were interpreted from the gravity directional derivative field in the study area. Numerous faults on the western side of the MTLFZ intersect the western graben, whereas very few faults intersect the eastern graben, indicating that the eastern graben possesses a relatively rigid structural framework.

The Moho depth gradually decreases from west to east, ranging from 30.6km to 36.0km. The MTLFZ forms a pronounced transition zone, with localized significant uplifts in the Weifang, Tancheng, and Suqian-Jiashan areas. By integrating the Moho depth with topographic elevation data, the actual crustal thickness was calculated and found to gradually increase from east to west. Its spatial distribution pattern is generally consistent with that of the Moho depth, although the amplitude of fluctuations is larger, ranging from 30.9km to 36.7km. The Airy isostatic crustal thickness was calculated to range from 33.99km to 37.08km. According to the gravity isostatic results, the regional crustal isostatic anomaly(Ⅲ) ranges from -1.33km to 3.11km, with relatively small overall variations. In most areas, the anomaly values are close to zero, indicating that the crust is either near isostatic equilibrium or evolving toward a stable state. Pronounced negative III values within the Luxi Block(I<-1.0km in the Tai'an and Jinan areas)suggest ongoing crustal uplift, whereas high positive III values in the eastern coastal region(I>1.1km in the Weifang area and I>1.75km in the Lianyungang area) indicate continued subsidence, driving the crust toward isostatic equilibrium. The isostatic states of the western and eastern grabens of the MTLFZ differ significantly: the western graben is close to isostatic balance, whereas the eastern graben remains in an unbalanced state.

The 1668 Tancheng earthquake occurred within the eastern graben of the MTLFZ, where the relatively rigid structural framework, caused by the scarcity of intersecting faults, favors stress and strain accumulation capable of generating major earthquakes. In addition, the crust beneath the Tancheng area within the eastern graben is characterized not only by a pronounced isostatic imbalance(I>1.1km), but also by significant Moho uplift(depth<33km). These conditions together constitute the deep seismogenic environment responsible for the occurrence of the 1668 Tancheng M_S8.5 earthquake.

LATE QUATERNARY ACTIVITY FEATURES OF THE NANLIU SEGMENT OF THE ANQIU-JUXIAN FAULT IN THE TANLU FAULT ZONE

FU Jun-dong, WU Hai-ying, ZHANG Jian-min, LU Zi-lin, XIA Nuan, LI Zhi-heng, WANG Lei, ZOU Qi-feng

2026, 48(3):  757-773.  DOI: 10.3969/j.issn.0253-4967.20240164

Asbtract ( )   HTML ( )   PDF (12862KB) ( )  

Figures and Tables | References | Related Articles | Metrics

The Tanlu fault zone is the largest deep-seated fault zone in eastern China, and its Late Quaternary activity remains a key and challenging issue in geological research. This study focuses on the Nanliu segment of the Anqiu-Juxian fault, one of the most active sections of the Tanlu fault zone, and investigates its Late Quaternary activity characteristics, including shallow fault structure, burial depth of the upper fault tip, stratigraphic displacement, vertical slip rate, and related features.

Because the southern segment of the Anqiu-Juxian Fault is mainly expressed as a blind fault, this study employed an integrated approach combining high-resolution shallow seismic reflection profiling with borehole-constrained geological section analysis. Two high-resolution shallow seismic reflection profiles and two borehole-constrained geological sections were completed, and 23 geochronological samples were collected and analyzed. The resulting shallow structural profiles suggest that the southern segment of the Anqiu-Juxian Fault mainly consists of a western branch fault and an eastern branch fault.

The borehole-constrained section at Songjia village reveals that the eastern branch of the Anqiu-Juxian Fault comprises two branch faults, f₁ and f₂, from west to east. These faults dip eastward at an apparent angle of approximately 80°, exhibit reverse-faulting characteristics, and are arranged in a high-angle stepped pattern. The shallowest burial depth of the upper fault tip of f₁ is approximately 3.85m, and its latest activity is dated to (53.9±2.9) to (58.2±2.3)ka. Variations in displacement among the lower strata indicate multiphase activity of the f₁ fault, with an average vertical slip rate of approximately 0.03mm/a. The upper fault tip of f₂ is buried at a depth of approximately 16.9~18.5m. In the lower part of the borehole core from this section, clear slickensides and densely developed high-angle foliations were observed, indicating dextral-reverse faulting.

The borehole-constrained section at Qiancao Village shows that the western branch of the Anqiu-Juxian Fault consists of five faults, f₁-f₅, from west to east. These faults dip at approximately 80°, display normal-faulting characteristics, and are distributed in a high-angle stepped pattern. All faults cut the overlying Late Pleistocene strata, with a vertical displacement of approximately 7.0m. Among them, f₁ has the shallowest upper fault tip, buried at approximately 9.7m, and its latest activity is dated to (54.8±3.4) to (78.4±3.2)ka. In the lower part of the borehole core from this section, relatively fresh purplish-red fault gouge, slickensides with a dip angle of approximately 40°, and densely developed high-angle schistosity were observed, indicating dextral-normal faulting.

Both the eastern and western branch faults are composed of multiple secondary faults. The occurrence of slickensides, fault gouge, and schistosity in the lower parts of the borehole cores collectively suggests that the fault zone exhibits both reverse and normal components under a dextral transpressional stress regime. The upper fault tips of the eastern branch are shallower than those of the western branch, indicating stronger activity along the eastern branch. Differences in displacement among different lower stratigraphic levels of the eastern branch fault are inferred to result from multiphase fault activity. A large number of liquefiable soil layers, including silty fine sand, medium-fine sand, and medium-coarse sand, are distributed in the borehole-constrained geological sections. However, no evidence of seismic liquefaction, such as sand veins, was identified. This suggests that the surface rupture zone of the 1668 Tancheng earthquake did not extend into the study area.

The 2011 M_W9.0 Tohoku earthquake in Japan significantly modified the stress state of the northeastern and North China sections of the Tanlu fault zone, promoting seismic energy accumulation in the Yishu fault zone. In 2024, the Nanliu segment of the Anqiu-Juxian fault, located in the northern part of the Tanlu fault zone, was affected by a M_W4.7 earthquake and several moderate aftershocks close to magnitude 4 in the Feidong area at the southern end of the Tanlu fault zone. This event represents the largest earthquake along the Tanlu fault zone in recent years. Considering that the Nanliu segment contains multiple normal and reverse faults, has shallow upper fault tips and complex fault structures, has remained strongly active since the late Quaternary, and has repeatedly been identified as a national seismic risk zone, this segment is considered to pose a certain level of seismic hazard. Further monitoring and investigation of this area are therefore recommended.

The results of this study provide fundamental data for major engineering site selection, urban planning and construction, and earthquake disaster prevention. They also offer reference value for understanding geodynamic issues related to the structural evolution of the Tanlu fault zone.

DEEP STRUCTURAL CHARACTERISTIC OF MAJOR FAULTS IN THE JIN-JI-MENG BOUNDARY AREA BASED ON BOUGUER GRAVITY DATA

LUO Xiang-fei, LI Zhong-liang, WANG Ze-yuan, YU Bo, JI Ji-fa, HAO Peng-fei, LIU Dong-yang, HE Wei-min

2026, 48(3):  774-796.  DOI: 10.3969/j.issn.0253-4967.20240078

Asbtract ( )   HTML ( )   PDF (7899KB) ( )  

Figures and Tables | References | Related Articles | Metrics

This paper applies multi-scale wavelet decomposition to high-precision Bouguer gravity anomalies in the study area, yielding Bouguer gravity anomaly fields of different scales within the same spatial domain. We investigate the relationships between these multi-scale Bouguer gravity anomaly fields and fault structures. By integrating previous geological and geophysical exploration results, this study further examines the major faults and deep structural characteristics of the study area. The main findings are as follows:

(1) As the wavelet detail order increases from low to high, the closed contours of the Bouguer gravity anomaly field evolve from small to large, and from scattered and complex to continuous and simple. The 1st- and 2nd-order detail fields mainly reflect geological structures and density inhomogeneities in the upper crust. The 3rd-order detail field mainly reflects density differences of geological bodies in the middle and upper crust, whereas the 4th-order detail field reflects density differences of geological bodies in the lower crust. The strikes of the gravity anomaly axes are predominantly NNE and NE, consistent with the regional structural trend. Gravity anomalies change sharply at block boundaries. All four lower-crustal faults are located along block boundaries, indicating that deep large faults at these boundaries control the distribution of gravity anomalies.

(2) Based on the wavelet detail fields of gravity anomalies at different orders and the fault analysis results, a total of 21 faults were identified in the study area, including 13 upper-crustal faults, 4 middle-crustal faults, and 4 lower-crustal faults. However, owing to factors such as fault scale, crustal structure, and data resolution, the Bouguer gravity anomaly field could not resolve all faults. To verify the reliability of the results, seismic reflection profiles and previous regional tectonic studies were compiled to compare the fault cutting depths interpreted from the gravity field with those derived from seismic profiles. Except for the Tianzhen-Yanggao Fault and the Sangganhe Fault, which were not identified in the gravity field, the fault cutting depths inferred from the gravity data show good agreement with those from the seismic profiles for the remaining faults, thereby supporting the credibility and reliability of gravity-based fault identification.

(3) The Moho depth ranges from 29.0 to 42.5km and generally deepens from southeast to northwest. In the Taihang Mountains, the Moho rapidly shoals and the crust becomes thinner. Crustal thickness varies markedly among different blocks: it is the greatest in the Yinshan-Yanshan fault block, followed by the Ordos fault block and the Shanxi rift zone. The overall trend of Bouguer gravity anomalies broadly reflects variations in crustal thickness. The Moho discontinuity zone represents not only a zone of crustal thickness variation, but also a deep large-fault zone and a boundary between fault blocks.

(4) The four deep large faults identified in the study area—the Kouquan Fault(F₇), the northern margin fault of the Yanfan Basin(F₁₅), the northern piedmont fault of the Liulengshan Mountains(F₁₆), and the Yanhecheng-Zijingguan Fault(F₁₉)—are all located at block boundaries. Moreover, except for the northern piedmont fault of the Liulengshan Mountains, the other three faults also coincide with zones of marked Moho variation. Thus, the Moho discontinuity zone is both a crustal thickness transition zone and a deep large-fault zone, while also serving as a boundary between fault blocks.

(5) All three earthquakes with magnitudes of 7 or greater in the study area occurred on or near active fault zones. Earthquake occurrence requires the accumulation and release of stress and strain energy. Because fault zones are relatively weak regions within the crust, they are the primary sites of stress concentration and release, and their activity directly generates earthquakes. In addition, two of these earthquakes occurred where the Moho interface changes significantly. Abrupt changes in the Moho may trigger deep stress perturbations, which couple with fault activity to induce seismic events. In other words, beneath locally uplifted or subsided segments of the Moho surface controlled by faults lies a deep tectonic setting favorable for major earthquakes.

(6) Owing to the influence of the grid size and spatial extent of the Bouguer gravity anomaly data, the downward cutting depth of faults inferred from the power spectrum does not represent the true burial depth and shows some deviation from the actual fault depth. It therefore reflects only the approximate downward extent of the faults.

Through a comprehensive analysis and systematic review of the spatial distribution of major faults in the study area, combined with investigation of the Moho surface, deep large faults, and historical strong earthquakes, this paper improves our understanding of the seismogenic environment and deep structural characteristics of the region. These results are of practical significance for understanding earthquake mechanisms and constraining potential earthquake locations, and can provide a basis for future earthquake prevention and disaster mitigation.

STUDY ON THE TRIGGERING FACTOR OF THE LOESS LIQUE-FACTION INDUCED BY THE JISHISHAN EARTHQUAKE

WANG Xiu-ying, ZHAO Guo-cun, FAN Xi-wei, GAO Peng, ZHANG Guo-hong, MA Zhi-xia, CHEN Xu-geng

2026, 48(3):  797-813.  DOI: 10.3969/j.issn.0253-4967.20240141

Asbtract ( )   HTML ( )   PDF (3082KB) ( )  

Figures and Tables | References | Related Articles | Metrics

An Extremely severe liquefaction and mudflow disaster was triggered by the Jishishan M6.2 earthquake that occurred on December 18, 2023, resulting in 20 fatalities. Many subsequent studies have been conducted on the liquefaction and mud-flow disaster triggered by this earthquake. However, those studies mainly focus on the issues of geological conditions, water content, dynamics of the mudflow, and the disaster chains triggered by this earthquake, and few studies are focusing on the issues related to earthquake triggering factors.

The EEW(Earthquake Early Warning) Networks in China was completed just before this earthquake, and as a result, the densely distributed early warning stations in the epicenter area recorded the strong motion vibration process, providing a large number of actual observations for loess liquefaction studies using strong motion data. Therefore, factors potentially affecting liquefaction are analyzed using the strong motion data obtained under different site conditions in the vicinity of the epicenter of the Jishishan Earthquake.

Two groups of collocated strong motion data, obtained from adjacent station pairs located in the loess covering regions, one group consisting of acceleration observations from on-ground soil site and rock site and the other consisting of acceleration observations from on-ground soil site and underground soil site, are used to carry out the analysis from the perspective of amplitude, duration, and frequency characteristics. Some important understandings are obtained as follows.

The surficial loess layer shows a significant amplification effect on the strong motion amplitude, and the PGA amplification factor can even exceed 2. Due to this amplification effect, the seismic energy released in the surficial layer is several times greater than that of bedrock or underground soil layers.

Due to the amplification effect of the surficial loess layer, the bracketed duration of strong motion vibration is prolonged for the on-ground site observations, with an average duration being 1.3-1.4 times longer than that of the bedrock site or underground soil site.

The results of spectral ratios show a significant amplification effect for seismic waves under 5Hz, with the most significant amplification effect being 1-3Hz. This frequency is consistent with the average predominant frequency of 2Hz obtained in previous studies in the liquefaction and mudflow covering area, suggesting that the amplification effect should be due to the resonance of the seismic wave with the surficial loess layer.

The vertical component of strong motion must play an important role in the liquefaction process near the epicenter as it can reduce a structure’s shear stress. The maximum PGA of this earthquake reaches 949Gal, obtained from the station closest to the liquefaction site.

The surficial loess layer near the liquefaction point is under saturated status due to continuous irrigation just before this earthquake, which liquefied rapidly and evolved into a mudflow disaster under the joint dramatic action of both the horizontal and vertical strong motion vibration amplified by the surficial loess layer.

Fewer liquefaction case studies use actual observations, especially for loess liquefaction. The results obtained in this paper help understand the earthquake triggering factor of loess liquefaction, and also have reference significance for using strong motion data to study geological hazards induced by earthquakes.

DISTRIBUTION CHARACTERISTICS OF SHEAR WAVE VELOCITY AND ITS EMPIRICAL RELATIONSHIP WITH DEPTH IN TYPICAL SOIL LAYERS IN SHANGHAI AREA

YAN Zhao-lun, YE Ying-chen, LI Zong-chao, PENG Xiao-bo, ZHOU Zheng-hua, LI Xiao-jun

2026, 48(3):  814-829.  DOI: 10.3969/j.issn.0253-4967.20240165

Asbtract ( )   HTML ( )   PDF (7087KB) ( )  

Figures and Tables | References | Related Articles | Metrics

To provide a reference for predicting shear-wave velocity in the Shanghai area, this study compiled a large amount of borehole data from the region, classified and organized the major soil layers and their spatial distributions, analyzed the relationship between shear-wave velocity and depth, and established depth-dependent relationships and extrapolation models for different soil types.

After excluding records with abnormal waveform signals and low signal-to-noise ratios from the shear-wave velocity tests, 318 valid boreholes were retained, comprising a total of 30, 846 test points distributed within a depth range of 3~100m. The measured shear-wave velocities generally range from 100 to 500m/s. The data distribution is relatively concentrated and exhibits clear regularity, with shear-wave velocity showing a positive correlation with depth. At depths of approximately 50m or shallower, the variation of shear-wave velocity with depth displays a pronounced nonlinear trend. At depths of approximately 15m or shallower, shear-wave velocity increases relatively slowly with depth, after which the rate of increase gradually accelerates. Around 20m, the increase in shear-wave velocity with depth reaches its maximum rate and then gradually slows down. From approximately 50 to 100m, the increase in shear-wave velocity becomes more gradual and generally follows a linear trend.

In the Shanghai area, strata within 100m depth consist of unconsolidated Quaternary deposits ranging from the Holocene to the Middle Pleistocene, mainly composed of silty clay, clay, silt, and sand. Based on the lithology, consistency, and density of the soil layers, 46 soil layers were classified and simplified stepwise from subcategories to broader categories. According to the distribution of field-measured shear-wave velocity errors, soil types were merged into a single category when the difference in shear-wave velocity distribution between two soil types was within ±5%. Statistical analysis indicates that the differences in measured shear-wave velocity are relatively small between mucky silty clay and silty clay, between clayey silt and sandy silt, and among silty sand, fine sand, and medium-to-coarse sand, with average deviations of less than 5%. These soil types can therefore be grouped into silty clay, silty soil, and sandy soil, respectively. However, the deviations among fluid-plastic clay, soft-plastic clay, and stiff-plastic clay, as well as those between silty soil and sandy soil and among silty soil, sandy soil, and clay, are relatively large, with average deviations exceeding 5%; therefore, these soil types are not suitable for merging. Overall, shear-wave velocity follows the order of stiff-plastic clay>soft-plastic clay>fluid-plastic clay, sandy soil>silty soil, and silty sand>clay. Accordingly, the typical soil layers within 100m depth in the Shanghai area were grouped into five categories: Silty clay, soft-plastic clay, stiff-plastic clay, silty soil, and sandy soil.

Linear, quadratic polynomial, and power-function models were used to fit the depth-velocity relationships for the five soil types, and the goodness of fit was evaluated using the coefficient of determination, R². The quadratic polynomial and power-function models showed the best fitting performance, whereas the linear model performed relatively poorly. Compared with the power-function model, the quadratic polynomial model performed less satisfactorily in fitting deep soil layers, particularly for soft-plastic clay and silty soil, where substantial deviations from the mean values occurred at greater depths and unreasonable trends were observed. In contrast, the power-function model demonstrated better overall fitting performance, effectively constraining the fitted shear-wave velocities in deep soil layers and producing a relatively stable increasing trend, thereby showing advantages for velocity extrapolation. By extrapolating the shear-wave velocity characteristics of soil layers deeper than 100m using the power-function model and comparing the results with those reported by the Shanghai Earthquake Agency, it was found that the deep shear-wave velocities extrapolated from measured data at depths shallower than 100m increase gradually with depth, while the rate of increase progressively decreases. The variation trend obtained from the model extrapolation is generally consistent with the fitting results of the Shanghai Earthquake Agency, although the extrapolated values in this study are approximately 70m/s higher overall. However, considering the measured velocity data from deep boreholes, which exhibit substantial variability, the extrapolation curve obtained in this study generally falls within the range of the measured data distribution, and its trend is consistent with the measured data. This indicates that extrapolating shear-wave velocities for soil layers deeper than 100m using measured data from similar soil layers at depths shallower than 100m is relatively reliable. The empirical fitting formula for shear-wave velocity in soil layers deeper than 100m obtained in this study is V_S=-3.2+76.2×H^0.382, which can be used as a reference in practical applications.

In summary, this study conducted a detailed investigation of soil-layer distributions and the variation of shear-wave velocity with depth in the Shanghai area. A power-function model was used to fit the relationship between shear-wave velocity and depth for five typical soil layers, and shear-wave velocities were extrapolated for depths shallower than 3m and deeper than 100m. The depth-dependent shear-wave velocity relationships obtained in this study can provide a reference for predicting shear-wave velocities in the Shanghai region.

THREE DIMENSIONAL S-WAVE VELOCITY STRUCTURE AND SEISMICITY CHARACTERISTICS IN THE RONGXIAN-WEIYUAN-ZIZHONG AREA OF SOUTHERN SICHUAN

JIANG Ning-bo, LI Da-hu, ZUO Hong, WANG Chao-liang, GAO Mi, YI Gui-xi, CHEN Xue-fen

2026, 48(3):  830-852.  DOI: 10.3969/j.issn.0253-4967.20240123

Asbtract ( )   HTML ( )   PDF (13845KB) ( )  

Figures and Tables | References | Related Articles | Metrics

Since 2014, seismic activity along the southern margin of the Sichuan Basin has shown a significant increase in both frequency and magnitude. In particular, the Rongxian-Weiyuan-Zizhong region in southern Sichuan has experienced five destructive earthquakes with magnitudes of M_S≥4.7, as well as numerous events with M_S≥4.0, in recent years. However, the unique geological characteristics of this region, including its compartmentalized fold structures and the absence of Late Cenozoic sedimentary deposits, have hindered understanding of the seismogenic mechanisms and spatiotemporal evolution of earthquakes in the area. This limited understanding further constrains the accurate assessment of future seismic trends and associated seismic hazards. In response to the increasing occurrence of destructive earthquakes, constructing a detailed deep structural model of southern Sichuan is essential for identifying the geometry and properties of seismogenic faults, investigating earthquake generation mechanisms, and evaluating regional seismic risk.

To investigate the deep tectonic background and seismogenic processes of earthquakes in this region and to better assess seismic activity patterns and potential hazards, a dense short-period seismic array consisting of 75 stations with an average spacing of approximately 5km was deployed across the Rongxian-Weiyuan-Zizhong area. Continuous waveform data recorded by the array were processed using the ambient noise cross-correlation method to obtain empirical Green’s functions of surface waves between station pairs. Fundamental-mode Rayleigh-wave phase velocity dispersion curves within the period range of 1-8 s were extracted. Subsequently, the Fast Marching Surface-wave Tomography(FMST)method and CPS3.30 software were employed to invert the dispersion data and construct a high-resolution three-dimensional S-wave velocity model down to a depth of 6km. For microearthquake detection and location, the LOC-FLOW workflow was adopted. Specifically, a Lightweight Phase Picking Network(LPPN)was used for seismic phase identification, the Rapid Earthquake Association and Location(REAL)method was applied for phase association, and hypoinverse together with hypoDD was employed for precise earthquake relocation. In total, 1, 255 seismic events were identified, significantly exceeding the number reported in the regional seismic catalog during the same period. The dense seismic array provided excellent monitoring capability, enabling comprehensive analysis of the three-dimensional S-wave velocity structure, seismicity distribution, and deep tectonic environment of the study area.

The results show that most earthquakes recorded by the dense array in the Rongxian-Weiyuan-Zizhong region have magnitudes below M_L3.0, with a completeness magnitude of approximately M_L0.6. Their spatiotemporal distribution exhibits strong clustering characteristics, forming a ring-like pattern surrounding the urban area of Weiyuan. Most earthquakes occurred within the sedimentary cover overlying the crystalline basement, with focal depths primarily concentrated between 1 and 5km. The shallow three-dimensional S-wave velocity structure exhibits pronounced lateral heterogeneity. Within the upper 2km, S-wave velocities in most areas range from 2.2 to 2.6km/s, whereas some areas near the Lijiachang fold exhibit velocities exceeding 2.8km/s. The velocity structure generally increases with depth. At depths between 3 and 6km, velocities within the northwestern anticlinal structures exceed 2.8km/s, whereas the southeastern sedimentary cover exhibits velocities ranging from 2.6 to 2.8km/s. The spatial distribution of velocity anomalies correlates strongly with regional topography, with low-velocity anomalies corresponding to the southeastern sedimentary basin and high-velocity anomalies associated with the Lijiachang fold in the northwest.

Significant lateral variations in shallow subsurface properties are observed between the Rongxian and Weiyuan-Zizhong seismic zones. The epicenters of the Rongxian earthquake sequence, the Weiyuan M_S5.4 earthquake, and the Zizhong M_S5.2 earthquake are mainly concentrated near the boundaries between high- and low-velocity anomalies. The spatial distribution of the Rongxian-Weiyuan-Zizhong earthquake sequences is closely related to the velocity structure of the crustal medium, indicating that seismicity is strongly controlled by heterogeneous subsurface structures. A pronounced low-velocity anomaly is observed at depths of 3~5km on the southeastern side of the Rongxian seismic source region, suggesting the possible presence of deep fluids. It is therefore inferred that earthquakes in this region are jointly controlled by regional tectonic stress and fluid-induced perturbations, which may reactivate pre-existing buried faults.

ANALYSIS OF TECTONIC DEFORMATION CHARACTERISTICS OF THE DAQINGSHAN FRONTAL FAULT ZONE USING HIGH-RESOLUTION REMOTE SENSING IMAGERY

QU Hao-xin, GAO Li-xin, BAO Ren-kai, LI Dong, ZHANG Hao

2026, 48(3):  853-869.  DOI: 10.3969/j.issn.0253-4967.20240172

Asbtract ( )   HTML ( )   PDF (18705KB) ( )  

Figures and Tables | References | Related Articles | Metrics

The Daqingshan Frontal Fault is a large-scale, highly active, basin-controlling normal fault located along the northern margin of the Hetao Basin. Since the Cenozoic, its tectonic activity has played a crucial role in the uplift of the Daqingshan Mountains and in the evolution of the Hohhot-Baotou Basin. This study aims to systematically characterise the fault zone’s geometric configuration, segmentation, and tectonic deformation characteristics using high-resolution remote sensing data, and to explore its deformation mechanisms and the background of tectonic evolution. Based on stereo imagery from the Gaofen-7 satellite, combined with existing geological maps, DEM data, and field investigation, we conducted a systematic tectono-geomorphic interpretation of the Daqingshan frontal fault zone. Through image enhancement, digital terrain analysis, and 3D geomorphic modeling, typical landform units such as river terraces, alluvial platforms, and planation surfaces were identified and extracted. On this basis, multiple profiles were constructed nearly perpendicular and sub-parallel to the fault strike to derive quantitative tectonic parameters, such as vertical uplift and tilt angles of different fault blocks, thereby enabling analysis of the spatial distribution and segmentation of fault activity. The results show that tectonic deformation along the Daqingshan Frontal fault zone exhibits significant spatial heterogeneity, with clear vertical differential uplift and prominent segmental activity along strike.

The main findings are as follows: 1)According to the spatial distribution and deformation characteristics of planation and terrace surfaces, the tectonic landforms can be classified into two main types: terrace-dominated and planation-dominated zones. The central segment shows the greatest uplift and the most pronounced tilting(2°), indicating the strongest tectonic activity, whereas the western and eastern segments exhibit relatively weaker deformation. 2)Profiles nearly perpendicular to the fault strike reveal an overall stepped uplift pattern across the fault zone. Fault blocks are primarily displaced by normal faults, typically forming two levels of structural blocks. Uplift magnitudes generally decrease stepwise from the mountain crest toward the piedmont. In the western and central segments, planation surfaces on upper blocks are mostly southward-tilted, while in the eastern segment, they show a northward dip. 3)Longitudinal profiles along the fault zone demonstrate well-defined stratified geomorphic structures, allowing subdivision into multiple segments, including DQS-A, DQS-B, DQS-CD, and DQS-E. Compared with previous segmentation schemes, the spatial positions and geometric morphologies exhibit high consistency, thereby verifying the effectiveness and reliability of remote sensing methods in fault segmentation research. However, in areas with complex lithology or intense late-stage geomorphic modification, further processing is needed to improve the accuracy of interpretation. 4)Measured relative heights and absolute elevations of third-order terraces indicate that the central segment of the fault zone has experienced the strongest tectonic activity and highest cumulative deformation since the Late Pleistocene. This is highly consistent with the distributions of tilting angles and maximum uplift heights of major planation surfaces, further confirming the reliability of the remote-sensing interpretation. These results demonstrate that piedmont landforms controlled by fault activity can intuitively reflect the spatial distribution of displacement along the fault zone.

Through integrated analysis of high-resolution remote sensing data and multi-source geomorphic information, this study systematically reveals the strong spatial heterogeneity and distinct structural segmentation characteristics of the Daqingshan frontal fault zone. The central segment is the most tectonically active, reflecting both stress concentration and significant surface uplift. The opposite tilting directions of planation surfaces in the eastern and western segments suggest multiphase deformation and overprinting, highlighting the complex spatiotemporal evolution of the fault zone.

PCA-BASED DAMAGED BUILDING CHANGE DETECTION MODEL USING UAV IMAGE TEXTURE FEATURES: CASE STUDIES OF THE 2023 GANSU JISHISHAN M_S6.2 AND 2021 YUNNAN YANGBI M_S6.4 EARTHQUAKES

DU Hao-guo, ZUO Xiao-qing, LIN Xu-chuan, LU Yong-kun, DU Hao-biao, CHEN Yong-sheng, LI Ji-chao, ZHANG Fang-hao, HE Shi-fang, DENG Shu-rong, ZHAO Zheng-xian, XU Jun-zu, BAI Xian-fu, ZHANG Yuan-shuo

2026, 48(3):  870-891.  DOI: 10.3969/j.issn.0253-4967.20240144

Asbtract ( )   HTML ( )   PDF (11465KB) ( )  

Figures and Tables | References | Related Articles | Metrics

Rapid and accurate localization of individual buildings and identification of earthquake-induced damage are essential for post-earthquake loss assessment and the efficient allocation of rescue resources. To overcome the limitations of conventional seismic damage detection methods, including strong dependence on training samples, blurred building boundaries, and interference from non-building objects, this study proposes a damaged-building change detection model based on the fusion of texture features(TFs) from unmanned aerial vehicle(UAV)imagery and principal component analysis(PCA). By integrating pre-earthquake satellite imagery with post-earthquake UAV data, the proposed model substantially improves the accuracy and efficiency of seismic damage identification through multi-feature fusion and dimensionality reduction. Case studies of the 2023 Jishishan M_S6.2 earthquake in Gansu Province and the 2021 Yangbi M_S6.4 earthquake in Yunnan Province were conducted to systematically validate the applicability and technical advantages of the model.

A normalized digital surface model(nDSM)was generated from UAV-derived digital surface model(DSM)data using triangulated irregular network(TIN)iterative filtering and point-cloud thinning, effectively removing vegetation and terrain effects and enabling precise extraction of individual building outlines. Building targets in pre- and post-earthquake single-phase(SP)images were segmented using region of interest(ROI)technology, while high-precision image registration was performed to ensure spatiotemporal consistency. For change detection, a multi-feature fusion framework was developed, including three models: texture-feature PCA change detection(PCA+TF+ROI+CD), texture-feature intensity-difference change detection(TF+DC+ROI+CD), and intensity-difference change detection(DC+ROI+CD). These models were designed to address the limitations of traditional single-phase machine-learning methods, such as maximum likelihood(ML) and Mahalanobis distance(MD), and were compared with ML+SP, ROI+ML+SP, MD+SP, and ROI+MD+SP methods.

Experimental results show that the PCA+TF+ROI+CD model achieved an overall accuracy(OA) of 89.4% and a Kappa coefficient of 0.85 in the Yangbi earthquake case, substantially outperforming conventional single-phase methods, including ML with an OA of 78.2% and MD with an OA of 75.6%. The model also achieved post-earthquake image matching accuracies above 88% in both case studies, with values of 88%±4% in Gansu and 91%±3% in Yunnan. Full processing of the disaster-affected areas was completed within 7h, satisfying the 24-h emergency response requirement. By integrating nDSM and ROI techniques, the model reduced the false-detection rate by 30% and improved boundary clarity by 20%. Moreover, the combination of large-scale pre-earthquake satellite imagery and high-resolution post-earthquake UAV data enables multi-scale analysis, ranging from regional damage assessment to individual building-level inspection. In the Gansu case, for example, multi-texture PCA fusion incorporating contrast, dissimilarity, and mean features suppressed redundancy caused by inter-feature correlations, reduced speckle noise, and enabled visualization of damage distribution at both pixel and building scales.

Nevertheless, the model performance remains sensitive to image quality, with accuracy decreasing under cloud cover, shadows, or low-resolution conditions. In complex urban environments, blurred boundaries between buildings and adjacent objects may require supplementary LiDAR data. In addition, limitations in computational efficiency for large-scale data processing may restrict real-time emergency applications. Future work should focus on multi-source data fusion, including optical, radar, and LiDAR data, to improve robustness under adverse weather and complex terrain conditions; transfer learning and adaptive algorithms to enhance model generalization; and optimized parallel computing architectures to improve processing efficiency in operational disaster response. Overall, this study provides an efficient and transferable solution for post-earthquake building damage detection, with both methodological innovation and practical value, while further improvements are needed in multi-source data compatibility, real-time processing, and adaptability to complex scenarios.

RESEARCH ON AUTOMATIC EXTRACTION METHOD OF BUILDINGS IN RURAL AREAS BASED ON UAV IMAGES

LI Xiao-yang, HAN Zhen-hui, LI Xiao-hui, XIE Heng-yi

2026, 48(3):  892-906.  DOI: 10.3969/j.issn.0253-4967.20240171

Asbtract ( )   HTML ( )   PDF (10994KB) ( )  

Figures and Tables | References | Related Articles | Metrics

The pre-assessment of earthquake disaster losses is an important component in preparing for earthquake emergencies and reducing disaster losses, and it is also an important basis for rapid assessment after the earthquake. The data of building height, area, distribution position, and structure type are the basis of pre-evaluation work. To enable automatic extraction of building data, this paper focuses on housing in rural areas and adopts a method based on low-altitude UAV photogrammetry to continuously improve the accuracy of building data extraction, thereby providing important basic data support for the pre-assessment of earthquake disaster losses.

Because of the large number of houses in the pre-assessment area, it is impossible to complete the field investigation one by one. First, the visible light remote sensing image of the study area is obtained by UAV, and then the accurate extraction of building data is completed according to the steps of data preprocessing, difference analysis, automatic extraction of building information, and result verification. In data preprocessing, the digital surface model, elevation model and orthophoto data of the study area are generated through data analysis, parameter setting, spatial three-dimensional encryption, stitching and other steps. In the process of building data extraction, the extraction method based on elevation information is compared with the supervised classification method in terms of extraction efficiency and accuracy. The extraction method based on elevation information is based on the idea of making a difference between DSM and DEM. By making mean and standard deviation of pixels in different ranges, the height difference range is determined, and then the misclassified areas are eliminated according to the threshold. The error and noise problems are reprocessed to extract more accurate building information. In order to verify the accuracy of data extraction, in addition to superimposing and comparing the extracted images with DOM images, a field sampling survey was conducted to further verify the extracted results. In the proportion of housing structure types, the housing structure types are further interpreted by using the above housing extraction results and combining them with the field sampling survey, which is used for the output of pre-evaluation results.

In the supervised classification method, building information is mainly obtained by identifying the building features in the DOM images. The results show that this method can extract buildings accurately, and the extracted building area is 94 051.3m². However, there is a wrong extraction for cement roads with similar brightness and features to the roof, and there is a phenomenon of missing extraction for some old roof houses. In addition, the height information of buildings in this method can only be obtained by visual interpretation of inclined images in the same position, and then the number of buildings can be judged according to the interpreted height, and then the building area can be calculated. Based on the method of height difference extraction, in the difference analysis, the average and standard deviation of pixels are counted according to the height difference greater than 2 meters, and the range of height difference is determined to be 2.7 meters to 12 meters. Then, taking the house area as the reference value, the threshold value is set to 4.8m² to eliminate the misclassified area. and the problems of error and noise are reprocessed, and finally, the house area was 80 566.8m². The results show that this method is not only more accurate in extracting information, but also more convenient in obtaining the height, area and distribution position of houses. By comparing and analyzing the remote sensing images obtained by UAV with the results of the field investigation, the accuracy of house extraction is evaluated. The results show that the extraction accuracy of the method based on elevation information on the ground buildings is over 90%. Compared with the supervised classification method, there are relatively few misclassified areas, which can identify the building information more accurately. The extracted house information can be used to further interpret the proportion of houses with different structural types in the study area and improve the efficiency of interpretation.

In this study, rural areas are selected as the research area. To obtain accurate building data information in the research area, orthographic and oblique remote sensing images of the research area are collected by unmanned aerial vehicles, and two automatic extraction methods are compared and analyzed, which verifies the accuracy of the extraction methods. Compared with the supervised classification method, the extraction method based on elevation information adopted in this paper has higher reliability, higher extraction accuracy, and better extraction effect when the ground object type is complex. The overall extraction accuracy is not affected by the ground object type, and the building information in the research area can be extracted quickly and accurately. The extraction results can be used to interpret the area and proportion of buildings with different structural types, and can supplement and improve the spatial distribution database of buildings in the study area. This method realizes the automatic extraction of building information in the study area, improves the efficiency and accuracy of the pre-assessment work, and has been applied in the pre-assessment of earthquake disaster losses in rural areas of several districts and counties.

RESEARCH ON THE INFLUENCE MECHANISM OF DISASTER SELF-RESCUE AND MUTUAL-RESCUE ATTITUDES AMONG MEGACITY RESIDENTS

LI Yi-hang, ZHANG Zhi-hao, YUAN Qing-lu, DONG Yan-di, LIU Xue-tao, ZHANG Rui-cheng

2026, 48(3):  907-920.  DOI: 10.3969/j.issn.0253-4967.20250124

Asbtract ( )   HTML ( )   PDF (1144KB) ( )  

Figures and Tables | References | Related Articles | Metrics

Public self-rescue and mutual rescue are effective actions to reduce disaster losses and are an important part of social disaster response capabilities. To investigate the influencing mechanisms of disaster self-rescue and mutual rescue attitudes among residents in megacities, this study is based on survey data collected from July to October 2024 regarding earthquake risk perception in Chaoyang District, Beijing. An ordered logistic regression model was used to conduct an empirical analysis of 92 810 valid questionnaires, focusing on the driving effects of core variables—such as trust in government disaster prevention capabilities and personal disaster risk perception—on public attitudes.

The survey data reveal a contradiction of “high willingness, low skills.” Over 46% of respondents expressed confidence in their ability to protect themselves during an earthquake, and 65.32% agreed they could provide timely assistance to others. However, only 48.98% could correctly identify appropriate emergency measures, and 52.57%had never participated in disaster-related training or drills.

The empirical results show that trust in government and perceived impact are two core drivers. For each one-level increase in trust, the probability of positive self-rescue and other-rescue attitudes increases by 23.2% and 16.0%, respectively. Similarly, a one-level increase in perceived family impact increases these probabilities by 5.1% and 11.2%.

Further analysis indicates that attitudes vary significantly by gender, age, and disaster experience. Males’ self-rescue confidence relies more on trust and impact perception, while females are more sensitive to psychological panic. Trust is also a more significant factor for those aged 35-50 and individuals with little disaster experience.

Robustness tests using OLS and Ordered Probit models confirmed the reliability of these conclusions. Recommendations include enhancing government transparency to build trust, improving routine risk communication, and providing targeted skill training to bridge the gap between willingness and capability.

Conclusions and Recommendations: First, trust in the government’s disaster prevention capabilities is a key external driver of public self-rescue and mutual rescue. The government should recognize the positive correlation between disaster information disclosure and individual risk perception, continuously improve decision-making efficiency and operational standards, and effectively enhance public trust through greater transparency in disaster prevention actions. Second, individual risk perception is the primary motivation for self-rescue and mutual rescue. The government should prioritize routine risk awareness education to ensure the public understands local disaster risks, and focus on optimizing risk communication mechanisms so that risk perception can be effectively translated into actionable will and capability. Third, a practical contradiction exists between “high willingness and low skills” in public response, stemming from insufficient coverage and efficacy of current training. The government needs to provide targeted knowledge and skills training for different demographic groups(e.g., gender, age, disaster experience), strengthen public capability-building, and transform subjective willingness into tangible disaster reduction outcomes. This will ultimately establish a virtuous cycle where trusted government capacity enhances public willingness, which in turn improves public competence and overall societal disaster resilience.

The results reveal the key mechanisms and group heterogeneity of public disaster response attitudes, providing an empirical basis for formulating optimized policies to enhance societal disaster prevention, mitigation, and response capabilities.