Loading...

Table of Content

    20 June 2015, Volume 37 Issue 2
    KEY TECHNIQUES AND SEVERAL CASES ANALYSIS IN PALEO- SEISMIC STUDIES IN MAINLAND CHINA(5): NONVISIBILITY, DIEOUT OF FAULT STRANDS AND IDENTIFICATION OF YOUNG PALEOSEISMIC EVENTS
    RAN Yong-kang, WANG Hu, Li Yan-bao
    2015, 37(2):  343-356.  DOI: 10.3969/j.issn.0253-4967.2015.02.001
    Asbtract ( )   PDF (7395KB) ( )  
    References | Related Articles | Metrics

    Nonvisibility(dieout)of fault strands occurs primarily in stratigraphic units associated with young paleoseismic events, which may cause misidentification of the young events and bring more uncertainties for seismic risk assessment. Based on previous related studies, this paper integrates case studies in mainland China to discuss the nonvisibility of fault strands and identification of young paleoseismic events. Nonvisibility of fault strands is prevailing in sandy, soil, silty, loess, and clay-sandy units, and is more possibly associated with strike-slip faults comparing with normal and reverse faults. Case studies on several trenches across surface ruptures produced by the Wenchuan earthquake and others located at different regions suggest that trench siting, excavation, and comprehensive analysis are key technical points to identify young paleoseismic events in the stratigraphic units where nonvisibility of fault strands is prone to occur. Stratigraphic units with more sequences have been suggested to be good sites for trenching to avoid misidentification produced by nonvisibility of fault strands. Multiple trenching is facilitated to lower the influence of local nonvisibility. Assumed extending of upper and lower units, grain sizes, color, and soil horizon are the basic methods to identify nonvisibility. Analysis of microstructures, grain sizes and magnetic susceptibility is one of the future studies related to identification of nonvisibility of fault strands.

    FAULT INTERACTION AND EARTHQUAKE TRIGGERING DURING COSEISMIC PROCESS OF WENCHUAN EARTHQUAKE
    LIU Bo-yan, SHI Bao-ping
    2015, 37(2):  357-373.  DOI: 10.3969/j.issn.0253-4967.2015.02.002
    Asbtract ( )   PDF (5383KB) ( )  
    References | Related Articles | Metrics

    The 2008 MS8.0 Wenchuan earthquake occurring in Wenchuan, Sichuan Province caused heavy casualties and serious economic losses. The northern part of Longmen Shan piedmont fault moved after the main event, with a burst to the surface. Using the model of two faults(Yingxiu-Beichuan and Guanxian-Jiangyou Fault), we discussed the earthquake triggering mechanism and time delay of frictional instability of Longmenshan piedmont fault. Based on the rate- and state-dependent frictional law, we deduced the time delay of new rupture on either condition that shear stress rate is zero or not, and used the results to infer the coseismic process of Wenchuan earthquake. After comparing the time delay of frictional instability of Longmenshan piedmont fault, which was derived from three-component strong ground motion observations of Mianzhu-Qingping and Wenchuan-Wolong stations, we obtained the same results. We found that Yingxiu-Beichuan and Guanxian-Jiangyou Faults did not move together at the same time, but the dynamic seismic waves(stress waves)radiated from Yingxiu-Beichuan Fault triggered the movement of Guanxian-Jiangyou Fault, with a possible time delay of several seconds.

    PRESENT-DAY SLIP RATES AND POTENTIAL EARTHQUAKE RISKS ALONG THE NUJIANG FAULT
    WANG Yan-zhao, WANG Min, SHEN Zheng-kang, SHAO De-sheng, SHI Fa-qi
    2015, 37(2):  374-383.  DOI: 10.3969/j.issn.0253-4967.2015.02.003
    Asbtract ( )   PDF (2030KB) ( )  
    References | Related Articles | Metrics

    Disputes arise during the hydroelectric development along the Nujiang River, among which we focus on the present activity of the Nujiang Fault and its potential earthquake risks. Using velocity profiles derived from an updated GPS dataset in the Nujiang region, slip rates are estimated as 4.7±0.8, 6.4±1.1, and(4.3±0.3)mm/a right-laterally and 3.9±0.6, 5.6±1.1, and(4.5±0.3)mm/a extensionally along the Yabiluo-Liuku, Liuku-Saige, and south Saige segments of the Nujiang Fault, respectively. According to the relationship between accumulated amount of elastic energy and fault slip rate and that between earthquake magnitude and rupture displacement, above estimates and earthquake catalog suggest potentials of earthquakes with magnitudes 6.4, 6.5, and 6.4 or even larger on the Yabiluo-Liuku, Liuku-Saige, south Saige segments, respectively. Provided that the surface trace of the Nujiang Fault is distributed generally along the Nujiang River, our estimates of fault slip rates and seismic potentials should be taken into account in dam siting and determining requirements of seismic fortification for safety.

    A PRELIMINARY DISCUSSION ON THE FOCAL MECHANISM AND SEISMIC TECTONIC FEATURE OF THE 2013 LAIZHOU M4.6 EARTHQUAKE SEQUENCE
    ZHENG Jian-chang, LI Dong-mei, WANG Peng, ZHAO Jin-hua, XU Chang-peng
    2015, 37(2):  384-399.  DOI: 10.3969/j.issn.0253-4967.2015.02.004
    Asbtract ( )   PDF (5271KB) ( )  
    References | Related Articles | Metrics

    An MS4.6 earthquake occurred at noon on Nov. 23, 2013 at Laizhou, Shandong Province, China. This earthquake is the largest event since the Sept. 20, 1995 Cangshan MS5.2 earthquake in Shandong area, and shook the whole Shandong Peninsula. The local area has low seismic activities, only one ML3 earthquake sequence was recorded from 1970 to 2012. But since 2012, small shocks break out every now and then, up to the recent MS4.6 sequence.We investigate the faulting process of the 2012—2014 Laizhou M4.6 earthquake sequence by combining relocated hypocenters and focal mechanisms. CAP method and additional bootstrap technique are employed to stably invert the moment tensor solution and to estimate its uncertainties. The average faulting parameters are: A. strike=239.6°, dip=75.0°, rake=174.4°; B. strike=331.1°, dip=84.6°, rake=15.0°, and error range of P, T axes is about 20°。We use HASH method to solve the focal mechanism solutions for 12 small events(ML≥3.0)in the sequence, and adopt double difference method(HypoDD)to analyze precisely the aftershock distribution.Relocation images show that, except 3 small shocks away from the swarm, the concentrated area of Laizhou sequence presents a NE-oriented major axis, and the sources distribution indicates a NW dipping fault, with a dip angle about 70°, which is in accord with the solutions for small events retrieved by HASH method.Finally, a discussion on the structural features of seismic tectonic and faulting process is made by using of all the results and relative geological data, and several opinions are concluded as follows:(1) There was an ordered rupture process at the earlier stage. At the very beginning(Jan 1, 2012 ML 3.2), rupture spread towards northeast. After the MS4.6 mainshock, rupture of the aftershocks became disordered, and sources distribution became more stochastic.(2) Small events before the mainshock scattered around the main rupture area; the occurrence of MS4.6 event filled up the gap.(3) Strike-slipping is the dominant faulting type in the earlier stage of the sequence. Two foreshocks right before the mainshock display some thrust component. This maybe implicates the strengthening of regional stress relative to the mainshock. The focal mechanism variation of small aftershocks indicates stress field's adjustment at deep source area after the mainshock.(4) Slipping vectors of the fault are in accord with accurate location results, which reveals the dynamics of faulting process.(5) The seismotectonic characters of Laizhou earthquake sequence revealed by this paper are consistent with other regional geology data. Focal mechanisms conform to the orientation of regional maximum horizontal principal compressive stress. This implies that Laizhou earthquake sequence occurred under the regional stress field, and has relationship with the relative motion between tectonic blocks.

    IDENTIFY PALEO-EARTHQUAKES USING QUANTITATIVE MORPHOLOGY OF BEDROCK FAULT SURFACE——A CASE STUDY ON THE HUOSHAN PIEDMONT FAULT
    HE Hong-lin, WEI Zhan-yu, BI Li-si, XU Yue-ren
    2015, 37(2):  400-412.  DOI: 10.3969/j.issn.0253-4967.2015.02.005
    Asbtract ( )   PDF (6424KB) ( )  
    References | Related Articles | Metrics

    The quantitative analysis of morphologic characteristics of bedrock fault surface is a useful approach to study faulting history and identify paleo-earthquake. It is an effective complement to trenching technique, especially to identify paleo-earthquakes in a bedrock area where the trenching technique cannot be applied. In this paper, we calculate the 2D fractal dimension of three bedrock fault surfaces on Huoshan piedmont fault in Shanxi graben, China using the isotropic empirical variogram. Taking average fractal dimensions of every horizontal tape and plotting them along the vertical axis, we find the fractal dimension presents pronounced segmentation in vertical direction. This step change of the average fractal dimensions demonstrates obvious segmentation of the fault surface morphology. Then, the segmentation of fault surface morphology, showing different exposure duration of each segment, is caused by periodic faulting earthquake, but not continuous erosion. Therefore, taking best normal fitting of average fractal dimensions of each segment as a characteristic value to describe the surface morphology of the fault surface segment, the characteristic value can be used to estimate the exposure duration of the fault surface segment and then the occurrence time of the faulting earthquake that made the segment exposed. The width of each fault surface segment can also be regarded as an approximate vertical coseismic displacement. Based on the segmentation of quantitative morphology of the three fault surfaces on the Huoshan piedmont fault, we identify three faulting earthquake events. Combined with trenching results reported by previous researches, we attempt to fit an empirical relationship between the exposure time and the morphological characteristic value on the fault. The co-seismic vertical displacement of a characteristic earthquake on the Huoshan piedmont fault is estimated to be 3.5m(3~4m), the average width of all middle fault surface segments. Moreover, the small gap of average 0.5~1m width between two adjacent segments, where fractal value increases gradually with the increased fault surface height, is inferred to be caused by erosion between two faulting earthquakes.

    THE SEISMOGENIC STRUCTURE OF THE M8.0 PINGLUO EARTHQUAKE IN 1739
    LEI Qi-yun, CHAI Chi-zhang, DU Peng, YU Jing-xing, WANG Yin, XIE Xiao-feng
    2015, 37(2):  413-429.  DOI: 10.3969/j.issn.0253-4967.2015.02.006
    Asbtract ( )   PDF (10823KB) ( )  
    References | Related Articles | Metrics

    The 1739 M8.0 Pingluo earthquake is the largest destructive earthquake occurring on the Yinchuan plain in history. However, there are different understandings about the seismogenic structure of this earthquake. In this paper, we re-evaluate the seismogenic structure of the 1739 M8.0 Pingluo earthquake after our investigation and detailed measurement of the seismic dislocations on the Great Wall and the surrounding tableland, and also the latest results of trenching, drilling, and shallow seismic exploration are considered as well. The results show that the latest rupture event of the Helanshan piedmont fault occurred after 600~700a BP, the Great Wall built in Ming Dynasty about 500 years ago was faulted by Helanshan piedmont fault. Although the distribution of Yinchuan buried fault coincides much with the distribution of the meizoseismal area, the fault's northward extending stopped at Yaofu town, and its Holocene active segment is less than 36km in length. The latest surface rupture occurred shortly before 3400a BP. The 1739 Pingluo earthquake did not rupture the ground surface along the Yinchuan buried fault. The presence of growth strata and the non-synchronous deformation of strata near the fault demonstrate that Yinchuan buried fault did not rupture at all or there was rupture but absorbed by the loose layers in the 1739 Pingluo earthquake. Therefore, the Helanshan piedmont fault is the seismogenic structure of the 1739 M8 Pingluo earthquake, rather than the Yinchuan buried fault, and there is no synchronous rupture between two faults. The difference of location between the seismogenic structures and the meizoseismal area of the Pingluo M8 earthquake may be caused by the factors, such as fault dip, groundwater depth, basin structure, loose formations, the degree of residents gathering, so on. The phenomenon that the meizoseismal area shifts to the center of the basin of earthquake generated by faulting of a listric fault on the boundary of the basin should be paid more attention to in seismic fortification in similar areas.

    TYPICAL CASE ANALYSIS ON APPLICATION OF MULTI-METHOD DETECTION TECHNIQUE TO ACTIVE FAULT EXPLORATION IN SUQIAN CITY
    CAO Jun, RAN Yong-kang, XU Han-gang, LI Yan-bao, ZHANG Peng, MA Xing-quan, LI Li-mei
    2015, 37(2):  430-439.  DOI: 10.3969/j.issn.0253-4967.2015.02.007
    Asbtract ( )   PDF (6737KB) ( )  
    References | Related Articles | Metrics

    Active fault is one of potential geohazards in cities. Locating and dating buried active faults in urban areas have been a difficult issue in active fault exploration. In this paper, we take the detection of the buried active fault performed at Hehuan Road in the north of Suqian city as an example. We preliminarily mapped the fault through field investigation and shallow seismic reflection survey technique. Furthermore, based on the principle of doubling section method, we conducted multiple drilling to constrain the upper faulted point which is located in a range of 5m in horizon and 4.4~6.1m in depth. Finally, we determined the exact location and latest activity of the fault by trenching. Obviously, good results have been acquired on the accurate location and activity of the Suqian segment of Anqiu-Juxian Fault using multi-level and multi-means detection method. Besides, we observed from the detection at the Hehuan Road site that at least four paleoseismic events occurred during the past 80000 yrs, and the result indicates that the latest faulting event on the fault is younger than(5.9±0.3)ka BP and the buried active fault at the Hehuan Road is a Holocene active fault. The result of buried active fault detection at the Hehuan Road site provides quantitative parameters for evaluation of seismic hazards and planning the width of safety distance in Suqian City.

    LATE QUATERNARY ACTIVITY OF ZHUMA FAULTON THE NORTH SEGMENT OF DALIANGSHAN FAULT ZONE
    SUN Hao-yue, HE Hong-lin, WEI Zhan-yu, GAO Wei
    2015, 37(2):  440-454.  DOI: 10.3969/j.issn.0253-4967.2015.02.008
    Asbtract ( )   PDF (12430KB) ( )  
    References | Related Articles | Metrics

    Daliangshan Fault Zone (DFZ) constitutes a significant part of the eastern boundary of Sichuan-Yunnan Active Block (SYAB). Studying the activity and slip rate of this fault zone is not only of great significance in understanding the movement of tectonic blocks and crustal deformation at the southeastern margin of Tibetan plateau, but also valuable in seismic hazard assessment and mid- and long-term forecasting of earthquake in west Sichuan. Zhuma Fault is the east branch of northern segment of DFZ which consists of six branch faults. Based on the detailed field investigations and through the accurate RTK (GPS) surveying and dating of the displaced landforms, we find that Zhuma Fault has been active since Holocene with a dominant left-lateral movement pattern and constrain its slip rate to be 1.5~3.1mm/a. Furthermore, a trench was excavated which reveals two paleoearthquakes occurring within(50.3±5.7)~30ka BP and 30~(17.4±1.2)ka BP, respectively from the stratigraphic evidence and OSL dating data. Although the slip rate on the Zhuma Fault is a little smaller than that on the southern segment of DFZ, we suggest uniform slip rates on the DFZ in consideration of the existence of another branch faults on the northern segment. The similar slip rate on DFZ to those on Anninghe Fault Zone (AFZ) and Zemuhe Fault Zone (ZFZ) implies that DFZ possesses a comparable partitioning component of displacement of Xianshuihe-Xiaojiang Fault System (XXFS) to AFZ and ZFZ. Further, the sum of slip rates on central segment of XXFS shows a good agreement with that on northern or southern segment. Thus, it is suggested that the DFZ not only patches the gap generated by the deviation of the strikes of AFZ and ZFZ from the average strike of XXFS, thus, making it a perfect small arc on earth, but also covers the deficiency in displacement and slip rate between central segment and northern or southern segment to maintain the XXFS to be harmonious. Moreover, according to the sedimentary characteristics and dating data, it is revealed that the alluvial-proluvial fans along the Zhuma Fault are formed by the glacial melt water in the last deglaciation after the Younger Dryas cooling event and such landforms could be widely developed in this region.

    NEW CHRONOLOGICAL CONSTRAINT ON THE CO-SEISMIC SURFACE RUPTURE SEGMENTS ASSOCIATED WITH THE LITANG FAULT
    ZHOU Chun-jing, WU Zhong-hai, ZHANG Ke-qi, LI Jia-cun, JIANG Yao, TIAN Ting-ting, LIU Yan-hui, HUANG Xiao-jin
    2015, 37(2):  455-467.  DOI: 10.3969/j.issn.0253-4967.2015.02.009
    Asbtract ( )   PDF (5296KB) ( )  
    References | Related Articles | Metrics

    Detailed mapping shows that there are two segments of co-seismic surface ruptures on the Litang-Dewu left-lateral strike-slip fault. The north segment is about 25km long, with a strike about 135°NE. The maximum horizontal left-lateral displacement on the north one is~1.8m and located at the high floodplains on the north side of the Wuliang River near Cun'ge village, offsetting the linear ridges that were left behind by human activity. The south segment is about 41km, striking generally about 146°NE. The maximum horizontal left-lateral displacement is located at the piedmont near the north side of the Rongjia mountain pass and the river floodplain scarp here is offset about 3.2m. There is a surface rupture gap about 11km between these two co-seismic surface rupture segments. The distribution of the co-seismic surface ruptures acquired by detailed mapping in the field survey, the earthquake event revealed by the trench, the AMS-14C dating result, the historical records of earthquakes at least since AD 1729 in the study area and the visiting on the local people, show consistently that the northern co-seismic surface rupture segment is most possibly produced by the 1729 Litang earthquake. The 1948 Litang earthquake was only responsible for the southern surface rupture segment. However, if only according to the 2 sigma calendar calibrated results of 14C dating, it cannot be excluded the possibility that the north segment maybe was produced by some older large earthquake occurring at some time during the AD 1420 to AD 1690. The moment magnitude(MW)of the 1729 earthquake is about 6.7 and that of the 1948 earthquake is about 7.0 calculated from the empirical relations between the earthquake magnitude and the rupture length.

    RESEARCH ON TECTONIC STRESS OF THE NORTHEAST SEGMENT OF HANCHENG FAULT ZONE
    LI Zi-hong, LI Bin, LIU Hong-fu, YAN Xiao-bing, HU Gui-rang
    2015, 37(2):  468-481.  DOI: 10.3969/j.issn.0253-4967.2015.02.010
    Asbtract ( )   PDF (4028KB) ( )  
    References | Related Articles | Metrics

    Knowledge about regional stress field is a significant basis for better understanding the tectonic activity of faults. This study reports the magnetic fabric investigation performed at sites along the northeastern section of the Hancheng Fault zone after we finished the 1 : 50,000 field mapping of active fualts in this region. Samples were collected at selected sites at Shaojialing, Zhubeizhuang and Shangyukou. Our results show that magnetic fabric, derived from anisotropy of magnetic susceptibility, reveals oblate susceptibility ellipsoids that are slightly modified compared to the inferred original depositional fabric. A dominant distribution of Kmax along the NW-SE direction indicates that the northeastern section of the Hancheng Fault zone is subjected to horizontal extensional stress along this direction. A weakly NW-SE directional distribution of Kmin is interpreted to reflect the action of horizontal compressive stress. This NW-SE compressive stress at the Shaojialing site appears to be somewhat stronger than that at the Shangyukou and Zhubeizhuang sites. Nevertheless, magnetic fabric properties are located in the oblate area in the PJ-T and Flinn diagrams. This may reflect inhomogeneous distributions of tectonic stresses along the Hancheng Fault zone, indicating that even within a strongly extensional stress-field, prolate magnetic fabric may be difficult to develop. This work may provide basic evidence for further studies on the activity of the Hancheng Fault zone.

    CHARACTERISTIC SLIP BEHAVIOR ANALYSIS OF THE WESTERN SEGMENT OF XIANGSHAN-TIANJINGSHAN FAULT ZONE SINCE LATE QUATERNARY
    LI Xin-nan, LI Chuan-you
    2015, 37(2):  482-495.  DOI: 10.3969/j.issn.0253-4967.2015.02.011
    Asbtract ( )   PDF (7367KB) ( )  
    References | Related Articles | Metrics

    The Xiangshan-Tianjingshan Fault zone is an integral part of the northeastern Tibet plateau fault system, and the 1709 earthquake of M7 1/2 happened on the eastern segment of this fault. But there remains a fresh surface rupture produced by the latest earthquake and a lot of gullies left-laterally dislocated synchronously along the western segment of the Xiangshan-Tianjingshan Fault zone, which has no historical records. To determine the western segment's seismicity since the Late Quaternary, we measured 240 horizontal offsets of the gullies or ridges and 62 vertical offsets, combining with the field geologic investigation and satellite remote sensing decipherment. Characteristics of distribution of the horizontal and vertical offsets are obtained by projecting the measured displacements along the stretch of the fault. Through probability density simulation and frequency statistical analysis for the horizontal offsets, the results show that there are obvious grouping character and multiple relationships. The six groups of horizontal offsets may represent 6 paleoearthquakes, with a similar event sequence in the trench excavated on the western segment. The coseismic offset of the latest earthquake is 3m, and the cumulative offsets produced by other older earthquakes is 6m, 9m, 12m, 16m and 20m, respectively, and each earthquake has a similar coseismic offset. Therefore, we suppose that the activity on the western segment of the Xiangshan-Tianjingshan Fault zone obeys a characteristic slip model since Late Quaternary.

    ISOSTATIC GRAVITY ANOMALY, STRUCTURAL FEATURES AND SEISMICITY OF NORTH CHINA CRATON
    LI Zhe-jun, XU Ru-gang, ZHANG Wei-jing, ZHANG Yi, LI Hui, YANG Guang-liang, LONG Jian-feng
    2015, 37(2):  496-509.  DOI: 10.3969/j.issn.0253-4967.2015.02.012
    Asbtract ( )   PDF (5175KB) ( )  
    References | Related Articles | Metrics

    Isostatic gravity anomaly is considered a sign of the isostatic state of the crust, and studies show that the isostatic state of crust is closely connected with the structural features and seismicity in many areas. In order to investigate the isostatic state of crust and to understand its relation to structural features and seismicity in North China Craton, a new isostatic residual gravity map of North China Craton has been computed using recently released earth gravitational model and digital terrain models. Free-air gravity anomalies of North China Craton have been prepared using the gravity data set of Earth Gravitational Model 2008(EGM2008). EGM2008 data set is believed to be reliable and studies show that EGM2008 free-air gravity anomalies have a general accuracy of 10.5mGal(1mGal=10-3cm/s2)in China. The topographic-isostatic corrections were computed based on an Airy-Heiskanen model of local compensation using a strict algorithm based on digital elevation model(DEM), the average crust thickness of the study area was derived from CRUST2.0, and the topographic and bathymetric data sets were derived from digital elevation model ASTER GDEM 2009(1arc second resolution)and ETOPO1(1arc minute resolution)respectively. Topographic-isostatic corrections were then added to the free-air gravity anomalies to determine the isostatic gravity anomalies of North China Craton with a gridding resolution of 5arc minutes. According to the results of calculations, distribution of isostatic gravity anomalies and its relations to structural features and seismicity of North China Craton were discussed. The results indicate that the spatial distribution of isostatic gravity anomalies is remarkably uneven in North China Craton, and isostatic gravity anomalies are very different between different fault blocks. Isostatic gravity anomalies of North China Craton are mainly controlled by neo-tectonic movements, and are significantly influenced both by lateral variations in crust density and deep structures. The close relation between isostatic gravity anomalies and neo-tectonic movements may imply that there are crustal features that are not compensated regionally and isostatic disequilibrium in North China Craton. The results also indicate that there are some connections between the distributions of isostatic gravity anomalies and seismicity in North China Craton, earthquakes tend to occur around areas with remarkable high or low isostatic gravity anomalies and at transition zones between positive and negative gravity anomalies, and we suggest that special attention should be paid to areas with similar isostatic gravity anomaly characteristics when performing seismic hazard analysis.

    NUMERICAL SIMULATION OF TEMPORAL-SPATIAL DISTRIBUTION OF PORE-FLUID PRESSURE INDUCED BY ZIPINGPU RESERVOIR IMPOUNDMENT
    YANG Yu, YANG Xiao-song, DUAN Qing-bao
    2015, 37(2):  510-523.  DOI: 10.3969/j.issn.0253-4967.2015.02.013
    Asbtract ( )   PDF (3960KB) ( )  
    References | Related Articles | Metrics

    The depth dependence of permeabilities of the fault rocks collected from the ruptures of 2008 Wenchuan earthquake and their host rocks were experimentally determined. On the basis of the measurement data, the temporal-spatial distributions of pore-fluid pressure induced by Zipingpu reservoir impoundment were numerically simulated. Modeled results indicate that the pore-fluid pressure is closely related to the patterns of the permeabilities. In other words, the results with the depth dependent permeabilities are strikingly different from those with the constant permeabilities. Existence of a fault hydro-dynamically connecting to the bottom of reservoir has a significant influence on the simulation of fluid infiltration owing to higher permeability in fault zones. The wider the fault zone is, the more obvious of fluid infiltrating along the fault. If fault width ranges from 100m to 300m, the pore-fluid pressure in the hypocenter region is about 0.1~0.15MPa, which is comparable with the normal stress caused by the reservoir impoundment on fault zone near the hypocenter of 2008 Wenchuan earthquake. This simulation provides an important constraint on the studies associated with the potential relationship between Zipingpu reservoir impoundment and Wenchuan earthquake.

    SHAKEMAP OF THE YUTIAN, XINJIANG M7.3 EARTHQUAKE ON 12 FEB 2014
    CHEN Kun, YU Yan-xiang, GAO Meng-tan, KANG Chuan-chuan
    2015, 37(2):  524-528.  DOI: 10.3969/j.issn.0253-4967.2015.02.014
    Asbtract ( )   PDF (1381KB) ( )  
    References | Related Articles | Metrics

    Based on geological structure in the epicenter area, focal mechanism solutions of this earthquake, the ground motion attenuation in western China, data of seismic stations and local site effect, Shakemap was acquired after the Yutian, Xinjiang earthquake on 12 February 2014. The results show that maximum shaking intensity may reach Ⅸ for Yutian earthquake, and there are no towns located in the area with shaking intensity Ⅷ and Ⅸ. Area with the shaking intensity Ⅶ is about 15 000km2. Shaking intensity of Aqiang and Yeyike town may reach Ⅶ. Area with shaking intensity Ⅵ is about 43 000km2, which includes Minfeng, Yutian County, Bositan, Nuer, Arele, Yingbage, Niya, Salewuzeke town, etc. In villages and towns between Yutian and Cele County, which are near the boundary of intensity Ⅵ area, earthquake damage may aggravate, because shear wave velocity of the soil in this area is relatively low, and seismic ground motion is peculiarly prone to be amplified. Amplitude limiting phenomenon occurred in the Qiemo seismic station, which is located in the direction of rupture, instead of in Hetian seismic station, to which the epicentral distance is less than that to Qiemo. This may be due to the directivity effect of this earthquake, which increased the level of shaking in this region.

    IMPROVEMENT OF EMPIRICAL GREEN'S FUNCTION METHOD USING A DYNAMIC CORNER FREQUENCY
    XIA Chen, ZHAO Bo-ming
    2015, 37(2):  529-540.  DOI: 10.3969/j.issn.0253-4967.2015.02.015
    Asbtract ( )   PDF (4099KB) ( )  
    References | Related Articles | Metrics

    In traditional empirical Green's function method, the difference between corner frequencies of large and small events is seldom considered. Since the simulation is performed by the superposition of the small events records, the results would have same corner frequencies as the small events, which contradict the fact that the corner frequency decreases with the moment magnitude. The results by traditional empirical Green's function method underestimate the low-frequency content when n2 small events are added. Assuming the corner frequency of a sub-fault is inversely proportional to the square root of the numbers of ruptured sub-faults, we construct a new function to apply the dynamic corner frequency to empirical Green's function method based on the ω2 source model. In the constructed function, the corner frequency of each sub-fault changes with the rupture propagation. We also develop a schematic diagram to show how the improved empirical Green's function method works.
    To validate the empirical Green's function method with dynamic corner frequency, we simulate the near fault strong ground motion of Wenchuan earthquake. Three aftershocks of Wenchuan earthquake are used as Green's functions of the asperities based on their magnitude and locations. Then ground motions of Wenchuan earthquake are synthesized by traditional empirical Green's function method and the empirical Green's function method with the dynamic corner frequency respectively. Comparison between the simulated results and observed data indicates that the traditional empirical Green's function method underestimates the low-frequency content of ground motions. While the empirical Green's function method with the dynamic corner frequency boosts the low-frequency motions without changing the high-frequency content. The results from the empirical Green's function method with the dynamic corner frequency show good agreement with the observed records after the low-frequency contents are improved, which proves that the dynamic corner frequency is valid to eliminate the underestimation due to the difference between the corner frequencies of large events and small ones.
    This study also proves that the empirical Green's function method with the dynamic corner frequency is a powerful tool to synthesize the near fault strong ground motions from MW7.9 earthquake.

    THE SEISMOGENIC ENVIRONMENT OF THE 2013 MINXIAN-ZHANGXIAN MS6.6 EARTHQUAKE BASED ON THE DEEP ELECTRICAL STRUCTURE
    ZHAO Ling-qiang, ZHAN Yan, ZHAO Guo-ze, CHEN Xiao-bin, YANG Hao, JIANG Feng
    2015, 37(2):  541-554.  DOI: 10.3969/j.issn.0253-4967.2015.02.016
    Asbtract ( )   PDF (5536KB) ( )  
    References | Related Articles | Metrics

    On July 22, 2013, an MS6.6 earthquake occurred at the junction of Minxian and Zhangxian. After the earthquake, magnetotelluric(MT)measurement was carried out at 45 sites along the NE-oriented profile across the West Qinling orogen(the west segment)and the earthquake area. Remote reference, "robust", and phase tensor decomposition techniques were used to process the MT data, and the NLCG two-dimensional inversion method was adopted to get the deep electrical structures. The deep electrical structure images indicate that there exists an inverted trapezoidal high-resistivity layer in the West Qinling orogenic belt(west segment)at the depth from the surface to about 20km deep, which is shallow in the northeast and southwest and deep in the middle. Under the high-resistivity layer is a low-resistivity layer, and they conjoin each other. There is a low-resistivity layer in the Songpan-Ganzi block(north part)at the southwest side of West Qinling orogenic belt(west segment)under the depth of 20km in the lower crust, which is shallow in the northeast and deep in the southwest, and the Longxi Basin at its northeast has a stable layered structure, suggesting that West Qinling orogenic belt(west segment)is being subject to the northward extrusion of the Songpan-Ganzi block and southward resistance of the Longxi Basin. The East Kunlun Fault(Tazang segment)faulted the low-resistivity layer in the lower crust of Songpan-Ganzi block. The Diebu-Bailongjiang Fault and Guangaishan-Dieshan Fault zone extend to a shallow depth and merge into the East Kunlun Fault(Tazang segment)in the deep part. The characteristic of low-resistivity of the media in the deep-seated structures in the East Kunlun Fault(Tazang segment)is the underlying cause for the gradual decrease of horizontal slip rate and gradual increase of vertical movement of the Tazang segment. The West Qinling Fault is a main geoelectric boundary zone, which extends through the Moho; Lintan-Tanchang Fault zone behaves as a low-resistivity layer with a certain width, which extends into the low-resistivity layer in the mid to lower crust. The source region of Minxian-Zhangxian MS6.6 earthquake locates in the core of inverted "trapezoid" of the low-resistivity layer in the West Qinling orogenic belt(west segment), that is, in the contact area between the high to low resistivity layers, and also in the low-resistivity fractured zone near the Lintan-Tanchang Fault. The interaction of southwest-northeast pushing from Songpan-Ganzi block and resistance of Longxi Basin block at its northeast is external dynamics of the Minxian-Zhangxian MS6.6 earthquake, and the high- and low-resistivity medium property and their contact relation in the seismic source region of the earthquake are the internal factor to generate this earthquake.

    CHARACTERISTICS OF GEOLOGICAL HAZARDS IN YILIANG EARTHQUAKE AND STATISTICAL ANALYSES OF THEIR SPATIAL DISTRIBUTION
    FENG Xin-ke, LI Zhi-qiang, LI Xiao-li, WANG Ming-zhen
    2015, 37(2):  555-564.  DOI: 10.3969/j.issn.0253-4967.2015.02.017
    Asbtract ( )   PDF (4418KB) ( )  
    References | Related Articles | Metrics

    On September 7, 2012 at 11: 19 and 12: 16(Beijing time), two catastrophic earthquakes with MS5.7 and MS5.6 struck the junction area of Yiliang County, Yunnan Province and Weining County, Guizhou Province, China. The research result indicates that the September 7, 2012 Yiliang earthquake triggered a mass of secondary geological disasters. The spatial distribution of the Yiliang earthquake triggered geological disasters is presented in this paper, by field check and using visual interpretation of remote sensing imageries. The statistical analyses results of their spatial distribution indicate that there are about 213 geological hazards throughout an area of about 0.67km2, consisting of collapse, landslide, rolling stones and mud-rock flow etc. We analyzed five characteristic parameters of these geological disasters triggered by Yiliang earthquake using GIS spatial analyses, such as seismic intensity, the formation stratum, slope angle, distances to river system and distances to road, and investigated in depth the control effect of these influence factors on the earthquake-induced geological hazards. This study reveals that: 1)most of the earthquake-induced geological hazards occurred in the Carboniferous strata along the Luoze River; 2)seismic region with intensity Ⅷ within the study area is the high incidence area of geological disasters, and the geological hazards have positive correlation with seismic intensity; 3)the hazards are mostly distributed in the slope range between 20°~50°; 4)geological disasters are densely distributed in the distance of 500m to river system and road, and they have negative correlation with the distance.

    DISCUSSION ON GENESIS OF INDUCED EARTHQUAKE BASED ON FOCAL MECHANISM IN XIANGJIABA RESERVOIR REGION
    FENG Xiang-dong, YUE Xiu-xia, WANG Yue-feng, WANG Xiao-shan, DIAO Gui-ling, ZHANG Hong-zhi, CHENG Wan-zheng, LI Yue, FENG Zhi-ren
    2015, 37(2):  565-575.  DOI: 10.3969/j.issn.0253-4967.2015.02.018
    Asbtract ( )   PDF (3160KB) ( )  
    References | Related Articles | Metrics

    Xiangjiaba Reservoir is currently China's third largest reservoir and began impounding water at the end of 2012. After the impoundment, the water level rose to 71m, while seismic activity near the dam was not significantly increased. At the end of June 2013, the reservoir began impounding water again, the water level continued to rise and flooded the tail region of the reservoir. In the reservoir area and the Xiluodu reservoir area in the upstream, a reservoir seismic network including 35 seismic stations was set up which can roundly record earthquakes in this area. According to the records of the reservoir seismic network from September 2007 to June 2013, only 38 earthquakes with ML≥1.0 occurred, 0.66 times a month on average, while in July-September 2013, 186 earthquakes with ML≥1.0 occurred, with an average of 62 events a month, nearly 100 times of that in the past. So, most of the earthquakes are induced earthquakes. At the same time 553 earthquakes with ML≤1.0 were also recorded in this area. A large number of small earthquakes occurring in the strong earthquake background area have caused a big stir. The source location of these earthquakes are rechecked based on 3D velocity model, 94% of the rechecked focal depth is less than 5km. Based on observations of the reservoir seismic network and vertical P- and S-wave's maximum amplitude ratio method, we inversed 9 focal mechanisms before the impoundment and 69 focal mechanisms after the impoundment in the tail region of the reservoir. Using these focal mechanisms, the stress field of the northern part and southern part of the study area is calculated in order to analyze the characteristic and cause of the induced earthquakes. The results indicate that most of the 69 focal mechanisms are strike-slip type, there is more transitional type, and less normal type and thrust type. The focal mechanisms spatial orientation is complex, fracture types are diverse, which may indicate that the stress state is uneven and the control of regional stress field to small earthquakes is weak. The stress field in the south and north is quite different and not consistent with regional stress field. The north shows compressive stress state while the south shows a state of weak extension. The Yaziba Fault, which passes through the tail region of reservoir, is an active fault, but does not control the induced seismicity, which may indicate that the reservoir storage inhibits the reverse fault activity. Carbonate rocks, limestone and karst cave are developed in the tail region. Analysis believes that reservoir water flows into the caves, penetrates into cracks and joints, leading to increased pore pressure, reducing the frictional strength and fracture strength and increasing reservoir water load which cause elastic deformation, so, it is believed that the combined action of all the above factors is the cause for the induced earthquakes.

    STUDY OF TEMPORAL SEISMIC VELOCITY CHANGES FROM AMBIENT SEISMIC NOISE AND ITS APPLICATION TO CHANGBAISHAN VOLCANO MONITORING
    LIU Guo-ming
    2015, 37(2):  576-587.  DOI: 10.3969/j.issn.0253-4967.2015.02.019
    Asbtract ( )   PDF (4423KB) ( )  
    References | Related Articles | Metrics

    This paper presented the current status of seismic noise interferometry in reconstructing the seismic velocity and discussed the four phases of data processing procedure from single-station data preparation to quality control. Each phase of the processing procedure has been discussed in detail with salient examples using seismic data acquired in Changbaishan volcano region. In the end, we used continuous recordings in Jilin, China to track the temporal changes of seismic velocity with the method introduced in this paper. We filtered the data between the frequency band 0.02 and 0.1Hz, where the Rayleigh waves are most sensitive to the shear wave velocities in the crust and uppermost mantle. The long-term seismic variations present strong seasonal influence which may be caused by precipitation. From 2001 to 2007, the amplitude of the relative seismic velocity change increased sharply and the highest variation reached 2%, far beyond the background level. During this time, a turbulence of the magma beneath this volcano occurred. So we conclude that this anomaly of velocity was caused by the turbulence of magma.

    GEO-ONTOLOGY MODELING FOR EARTHQUAKE EMERGENCY DECISION-MAKING KNOWLEDGE
    XU Jing-hai, NIE Gao-zhong
    2015, 37(2):  588-597.  DOI: 10.3969/j.issn.0253-4967.2015.02.020
    Asbtract ( )   PDF (3333KB) ( )  
    References | Related Articles | Metrics

    China is a country prone to serious earthquake disaster. After an earthquake, earthquake emergency and rescue are very important for the disaster relief, which is also one of three earthquake disaster mitigation jobs led by China Earthquake Administration. In earthquake emergency fields, earthquake preplans and GIS-based earthquake emergency command systems are the main researches and work. How to increase the intelligence and pertinence of the systems and preplans is an important and difficult issue in this area. Earthquake emergency decision-making knowledge provides a possible solution method. The modeling of earthquake emergency decision-making knowledge is the foundation for its use.
    We analyze the semantic need of earthquake emergency decision-making knowledge modeling. From the perspective of geospatial knowledge modeling, the earthquake emergency knowledge modeling primitives are put forward, which are adapted from geospatial knowledge modeling primitives. Using geo-ontology as the foundation, the earthquake emergency knowledge modeling primitives include: abstract geospatial modeling primitives, geographic modeling primitives and earthquake emergency field modeling primitives.
    A framework model for the earthquake emergency knowledge is proposed and according to the knowledge granularity the framework is divided into the earthquake emergency basic knowledge level, earthquake emergency rule knowledge level and earthquake emergency procedural knowledge level. Then the modeling of the earthquake emergency rule knowledge is discussed, which is composed of rule conditions and rule actions. Meanwhile, the modeling of the earthquake emergency procedural knowledge is introduced based on workflow method and consists of work nodes, control nodes and data nodes.
    Finally, Suzhou intelligent earthquake emergency decision-making assisting system is developed, in which several earthquake emergency decision-making knowledge are used. The protégé and ArcEngine software packages are used to realize the earthquake emergency knowledge modeling and application. Through the application system, the usage of framework model is demonstrated.

    NONLINEAR FINITE-ELEMENT SIMULATION OF CONJUGATE FAULTS SYSTEM AND ASSOCIATED EARTHQUAKE SWARM
    GUO Ting-ting, XU Xi-wei, XING Hui-lin, Yu Gui-hua
    2015, 37(2):  598-612.  DOI: 10.3969/j.issn.0253-4967.2015.02.021
    Asbtract ( )   PDF (9722KB) ( )  
    References | Related Articles | Metrics

    In this paper, adopting nonlinear finite-element method for faults with frictional contact, single fault and conjugate faults are calculated and compared respectively. And conjugate faults system is analyzed about its structural significance, combined with preparation and occurrence of strong earthquakes in the capital circle region. Study shows that seismogenic process of typical conjugated fault system from lock to unlock is well explained by Coulomb friction criteria, and unstable events can take place alternately along conjugate faults. The slip behavior of seismogenic tectonic model of great earthquake composed of two conjugate faults is verified. The numerical simulation as well as the analysis and discussion on the results provide scientific basis for earthquake forecasting and monitoring in areas where conjugate faults have developed.

    GEOMORPHIC FEATURES AND REMOTE SENSING RESEARCH OF FAULT ACTIVITY ALONG THE SOUTHEASTERN PIEDMONT FAULTS OF NYENCHEN TONGLHA MOUNTAINS
    JIAO Qi-song, ZHANG Jing-fa, JIANG Wen-liang
    2015, 37(2):  613-626.  DOI: 10.3969/j.issn.0253-4967.2015.02.022
    Asbtract ( )   PDF (10370KB) ( )  
    References | Related Articles | Metrics

    Based on ALOS, ETM+images and field works, combining with the existing research results of the study area, using information enhancement and image fusion methods, we extracted the texture, color and water-bearing features and studied the spatial distribution and development of the southeastern piedmont faults of the Nyenchen Tonglha Mountains. Moreover, SL index and Hack profile were used to analyze and compare the regional tectonic activity. The results show that the main faults obviously present a three-stage distribution on remote sensing images. Fault movement has produced different surface topography, such as fault scarp, fault facet and surface rupture zone. Small pull-apart basin, rift lakes and swamps were found in the basin. Their distribution and development are obviously controlled by faults. Geomorphic evidences interpreted from images generally indicate the fault movement property as normal faulting with strike-slip component. Major rivers cross the southeastern piedmont faults of the Nyenchen Tonglha Mountains from northwest to southeast and flow into Dangxiong-Yangbajain rift basins. The rivers with length bigger than four kilometers are selected to calculate the tectonic geomorphology parameters. The Hack profiles of rivers present obvious convex uplift that represents strong tectonic differential uplifting. Rivers had no time to make adjustments in the process of development and the tectonic movement produced convex and concave shape on the river section traces. The area where standard stream length-gradient index is abnormal indicates strong tectonic movement. This abnormal changes not only verify the impact on river profile caused by fault movement, but also improve the fault location accuracy when interpreted combining with these abnormal features. The average SL/K value in this area tends to increase from F1 to F3. From the point of historical earthquakes distribution, a large amount of small earthquakes occurred mainly on F3 and seldom on F1 and F2. This trend is similar to SL/K value change. It indicates that the fault activity increases accordingly from F1 to F3. The standard length-gradient index K represents the river erosion ability, which increases from F1 to F3. This feature shows that normal fault movement is strong on F3 and tectonic uplift has a significant impact on river erosion. Movement on F1 and F2 show strong strike-slip and weak normal faulting, whereas normal faulting is stronger on F3. Dislocation of rivers is more evident on the remote sensing image. The southeastern piedmont faults of Nyenchen Tonglha Mountains and Dangxiong-Yangbajain rift basins are important conversion and absorption zones in the central Tibetan plateau, where the seismic activity is still high and more attention should be paid.

    THE APPLICATION EXPERIMENT OF THREE-DIMENSIONAL SEISMIC REFLECTION METHOD IN THE DETECTION OF ACTIVE FAULTS: A CASE FROM LUHUATAI FAULT
    FENG Shao-ying, LIU Bao-jin, ZHAO Cheng-bin, HE Yin-Juan, TAN Ya-li, JIA Yan-xia
    2015, 37(2):  627-635.  DOI: 10.3969/j.issn.0253-4967.2015.02.023
    Asbtract ( )   PDF (4050KB) ( )  
    References | Related Articles | Metrics

    To test the effect of three-dimensional seismic reflection methods used in active fault survey, we have done a three-dimensional shallow seismic reflection exploration experiment around the Luhuatai Fault in the west of Yinchuan Basin. The experiment uses swath geometry of 8 lines and 10 shots. Every two adjacent swaths overlap 3 survey lines, thus 5 swaths and 28 survey lines are laid in total. The ground sampling grid is 5m×20m and the CMP grid is 2.5m×5m. The data volume that reflects the three-dimensional spatial structure of the Luhuatai Fault is obtained.In data processing, we select the suitable three-dimensional seismic data process modules. The main processes are composed of raw data input, three-dimensional geometry defining and checking, anomalous trace edit and first arrival mute, spherical divergence compensation, surgical filtering to eliminate surface waves, surface-consistent amplitude compensation, surface-consistent deconvolution, velocity analysis and residual static correction(twice iteration), DMO and the third time velocity analysis, final stacking, three-dimensional post-stack de-noising and horizontally interpolating, one-pass 3-D migration.
    3-D seismic data interpretation uses the way of human-computer interaction. Through a variety of methods such as multi-line profiles contrasting, time slicing, three-dimensional visualization, and 3-D coherence cube technology, the reflection horizons are discerned and tracked, and the three-dimensional data volume reflecting the spatial variation of strata and faults is obtained. The results after fine processing and synthetical interpretation show that the Luhuatai Fault consists of two normal faults that incline to each other. The major fault inclines to SE, and the minor fault inclines to NW. The distance between them gradually increases from north to south. In addition, the minor fault merges into the major fault at the depth of approximately 780~800m. The up-breakpoint of the major fault has a tendency of deepening from north to south. The up-breakpoint depth is about 25~30m in the northern part of experimental area, and about 35~40m in the southern part of experimental area.
    The experimental results show that the three-dimensional seismic data has the advantages of large volume of data, information-rich, high accuracy of migration, and high precision of tomography. It can reflect the three-dimensional spatial distribution of strata and faults in different aspects, and it is beneficial for the imaging of complex structures and faults.

    ACCURACY ANALYSIS OF TERRAIN POINT CLOUD ACQUIRED BY "STRUCTURE FROM MOTION" USING AERIAL PHOTOS
    WEI Zhan-yu, Arrowsmith Ramon, HE Hong-lin, GAO Wei
    2015, 37(2):  636-648.  DOI: 10.3969/j.issn.0253-4967.2015.02.024
    Asbtract ( )   PDF (5527KB) ( )  
    References | Related Articles | Metrics

    The need to acquire high-quality digital topographic data is evident throughout geoscience research. The use of these data elevates the research level of geosciences. Airborne and terrestrial light detection and ranging(LiDAR)are currently the most prevalent techniques for generating such data, but the high costs and complex post processing of these laser-based techniques restrict their availability. In the past few years, a new stereoscopic photogrammetry mapping method called Structure from Motion(SfM)has been applied in geoscience, in which the 3D digital topography is reconstructed using feature matching algorithms from overlapping photographs of multiple viewpoints. SfM only needs a series of overlapping images with no special requirements about the camera positions, orientations and lens parameters, making it possible to use images collected from an affordable SfM platform to rapidly generate high-quality 3D digital topography. This paper summarizes the basic principles and the SfM workflow, and shows that SfM is a low-cost, effective tool for geoscience applications compared to LiDAR. We use a series of digital aerial photos with~70% overlap collected at one-thousand-meter height to produce a textured(color)SfM point cloud with point density of 25.5/m2. Such a high density point cloud allows us to generate a DEM with grid size of 0.2m. Compared with LiDAR point cloud, statistical analysis shows that 58.3% of LiDAR points deviate vertically from the closed SfM point by <0.1m and 88.3% by <0.2m. There is different SfM accuracy in different landforms. The SfM accuracy is higher in low dips and subdued landforms than in steep landforms. In consideration of relative vertical error of 0.12m in LiDAR data, SfM has a higher measuring accuracy compared with LiDAR.

    A TENTATIVE TEST ON MODERATELY STRONG EARTHQUAKE PREDICTION IN CHINA BASED ON THERMAL ANOMALY INFORMATION AND BP NEURAL NETWORK
    SONG Dong-mei, SHI Hong-tao, SHAN Xin-jian, LIU Xue-mei, CUI Jian-yong, SHEN Chen, QU Chun-yan, SHAO Hong-mei, WANG Yi-bo, ZANG Lin, CHEN Wei-min, KONG Jian
    2015, 37(2):  649-660.  DOI: 10.3969/j.issn.0253-4967.2015.02.025
    Asbtract ( )   PDF (4301KB) ( )  
    References | Related Articles | Metrics

    Earthquake prediction is one of the key areas of earthquake research. Thermal infrared abnormity, which is the abnormally increased land surface temperature, is universal before earthquake and has complex nonlinear relation with the three elements of earthquake. Combining the advantage of neural network, this paper provides a method for earthquake prediction by taking thermal anomaly as information source and constructing a neural network to carry out the test. Based on the MODIS data which has synthesis of eight days with 1km resolution, taking a 10°×10° rectangle, whose center is the epicenter, as research area, and a two-month time before earthquake as the time range, we used RST algorithm to extract thermal anomaly information before earthquake. Considering the time-space relationship between thermal anomaly information and the fault zone, thinking carefully about the information of the neural network input neurons, we constructed BP neural network and used 100 earthquake cases with magnitudes larger than 5, and 70 random samples without earthquake in the research region for training and simulating. According to the statistical analysis, the prediction accuracy is 80%, missing prediction rate is 20%, and false prediction rate is 13.3%. Prediction accuracy of magnitude with error within magnitudes of 2 is 69%, prediction accuracy of earthquake origin time with error within 30 days is 87.5%, and prediction accuracy of epicenter location with error within 3°is 81.2%. The result shows that BP neural network-based earthquake prediction is feasible by using thermal infrared abnormal precursor extracted by RST method. However, in this experiment, the determination of the start time of thermal abnormity, the origin point and range of research area are based on the known epicenter location and time. In fact, the result depicts a non-linear relationship between earthquake and thermal abnormity, and the accuracy of prediction reflects the correlation degree. Therefore, the prediction accuracy of future earthquake may be not as large as our result. For future earthquake prediction, accurate selection of research area and neuron number of hidden layer in neural network has great influence on prediction result.