“6.1”芦山地震作用下宝兴新华村滑坡动力响应与失稳过程离散元模拟

斜坡浅表层是各类地震地质灾害发育的潜在破坏位置,坡面形态和坡体结构往往造成斜坡动力响应及破坏的复杂化。为探究不稳定斜坡浅表潜在滑动层动力响应特征与失稳过程,以芦山M_s6.1级地震触发的新华村滑坡为例,基于现场调查采用离散元方法建立了二维计算模型,分析了该斜坡潜在滑动层及坡面形态的动力响应特征并对其失稳过程进行了模拟。结果表明：(1)斜坡浅表潜在滑动层具有强烈动力放大效应;(2)微地貌对于潜在不稳定斜坡坡面的放大效应具有明显的影响,浅表潜在滑动层水平向及竖直向加速度在凸出部位的放大效应显著,凹陷部位相较于凸出部位放大效应较低;(3)研究揭示新华村滑坡在微地貌的作用下凸起地形呈现先于凹陷地形遭受破坏,其失稳过程分为震动放大局部震裂-凸出地形破坏-凹陷地形破坏-完全破坏整体下滑-重力堆积5个阶段。该研究结果有助于提升防灾人员对地震诱发潜在不稳定斜坡失稳的认识,为防灾减灾提供理论和数据支撑。     

浑厚山体随高程增加由表及里地震动响应研究

为探究浑厚山体不同高程由表及里地震动响应规律,以冷竹关山体为例,采用离散元软件建立地形与风化介质组合模型,并从底部边界输入汶川地震波信号,研究该山体两侧边坡的内外动力响应规律。结果表明,随着高程的增加,靠大渡河一侧边坡坡体内与坡表的加速度放大系数均表现出先增大后减小的节律性变化,在近坡顶时增大较快并达到最大值;靠近瓦斯沟一侧边坡坡表受地形起伏的影响,加速度放大系数存在凸坡放大、凹坡减小的特征;相同高程,随边坡由表及里深度的增加,加速度放大系数表现为逐步减小,当距坡表150~200 m时放大曲线趋于平缓;随高程的增加,加速度放大系数由表及里的减小速度变缓,且放大曲线收敛平缓的深度增大;随岩体风化程度的增加,岩体介质波速降低,共振效应使得加速度响应增大,与此同时,斜坡地形与介质组合效应使得坡表峰值加速度放大系数在2.0附近。     

四川珙县M_s 5.4级地震斜坡地震动响应特征

四川珙县M_s 5.4级地震是长宁M_s 6.0级地震第三大余震,在震中区诱发了多处滑坡和崩塌,研究坡体不同部位地震动响应特征对地震地质灾害预测和防治意义重大。通过在距离震源近的斜坡不同位置布置监测点,获取地震动参数及相关变化规律,对研究区域的坡体进行稳定性分析和研究。结果表明:地震在南北向的放大效应大于东西向,覆盖层土体最大位移达到了7.42 mm,与地震时南北向晃动强烈吻合;对比分析各个监测点的峰值加速度(PGA)、阿里亚斯(Arias)强度,覆盖层土体比高位基岩的放大效应显著;基岩地震波以低频为主,卓越频率为2～5 Hz,随着高程的增加对高频的地震波有过滤作用,在山顶3~#监测点的阿里亚斯强度比山脚1~#监测点(参考点)放大数倍,地震动能量强;通过该区域的抗震设计反应谱与监测反应谱对比分析,可以得出覆盖层土体和高位基岩地震动强烈,前者的宏观表现为覆盖层土体上老旧房屋毁坏严重,后者为高位基岩坡体滑坡和崩塌。     

丘陵地貌区斜坡强震动响应特征——以四川省长宁县为例

前人对高山峡谷地貌区高位斜坡动响应研究较多,但对低山、丘陵地貌区斜坡强震动响应研究很少。并且,研究这类地区顺倾斜坡的地震动响应对揭示长宁M_s 6.0级地震丰富的次生山地灾害动力机理具有一定意义。在四川省长宁县丘陵地貌区典型顺倾斜坡不同部位安置强震监测仪器,捕捉到40多次余震,通过SeismoSignal软件处理典型余震数据,分析峰值加速度、阿里亚斯强度、傅里叶频谱、加速度反应谱,进而采用水平/竖直向谱比法分析局部场地效应,最后对比分析低山、丘陵地貌区与高山峡谷地貌区的斜坡动响应。结果表明:坡顶峰值加速度可达4.171 m·s~(-2),最大位移可达0.008 05 m,当峰值加速度大于1.0 m·s~(-2)时,阿里亚斯强度出现陡增的趋势;局部地形的放大效应使得地质灾害以沿山脊的裂缝及小型崩塌为主,在低频2～5 Hz处,斜坡中上部易发生山体共振。以上研究揭示了砂、页岩顺倾斜坡高位放大效应,为地震山地灾害易发性区划提供地震监测资料。     

四川珙县下软上硬山岭地貌斜坡地震动响应特征

强震中下软上硬坡体同震崩塌发育, 为了揭示这类坡体地震动的响应特征, 在珙县五同村安置了强震监测仪, 对斜坡表面和不同岩性的地震动响应进行监测, 并记录到不同方位、不同震中距的2次地震。研究表明: ①地震动响应规律有极强的方向性和距离性。2次地震相距监测站台的方向和距离不同, 使Ms 4.0级地震的峰值加速度和阿里亚斯强度反而比Ms 3.2级地震小。②0~30 Hz的地震波在低地山岭的高陡临空面附近有放大效应。1#监测点的主频小于3#与5#监测点, 3#监测点的主频最高。5#点的幅值范围为0.018~0.055 m/s-2, 3#点幅值范围为0.036~0.087 m/s-2, 3#点相较于5#点, 其三向最高幅值同比放大了1.58~2.0倍。③泥质砂岩的主频为4.8~8.4 Hz, 灰岩的主频为5.5~21.4 Hz, 不同的岩层共振频率不同, 灰岩对地震波的选频放大效应强于泥质砂岩。④地震波在不同高程的山岭斜坡部位具有选择放大作用, 在一定范围内高程越大地形放大效应越明显。      

珙县低山区不同震源作用下斜坡地震动响应

在宜宾市珙县五同村布设斜坡地震动监测仪器,采集兴文县3.1级、珙县4.5级及长宁县4.6级地震数据。对监测剖面上1^#、2^#监测点采集到的珙县余震地震动响应数据进行滤波和校正处理,对比分析震级、高程、方位等因素对斜坡造成的地震响应后认为：1）当地震波的半波长与山体某些特定部位之间的距离接近时,斜坡地形与地震波的波峰产生耦合作用,地形放大效应显著,达到参考点的2.765倍,此处山体更容易产生较大幅度的摇晃,造成山体地质灾害;2）分析珙县4.5级与长宁县4.6级地震得出,当地震波沿特殊部位（如山脊等）传播时,地震加速度及阿里亚斯强度更加显著;3）分析地震加速度变化得出,在地震荷载作用下,随着震级的增大,斜坡的加速度响应也会增大;4）随着高程和震级的变化,监测点的地震加速度和阿里亚斯强度随之变化,高程和震级对斜坡的地震动响应具有放大效应;5）在高程和微地貌的耦合作用下,斜坡地震加速度也具有明显的放大效应。     

兴文MS5.1地震震裂斜坡地震动响应特征

通过对宜宾珙县震裂斜坡上的3台地震动监测仪器记录到的2022-04-06宜宾兴文M_S5.1地震地震波传播数据进行滤波和基线校正,分析时程曲线、傅里叶频谱和加速度反应谱。结果表明：1)本次地震持时短、幅度小,2^#和3^#监测点S波持时较1^#监测点长2～5 s; 2)傅里叶频谱显示,水平方向低频衰减、中高频放大、频带变宽,垂直方向变化不明显;3)以1^#监测点为参考,2^#和3^#监测点峰值加速度(PGA)放大系数为1.17～3.05,阿里亚斯强度(AI)放大系数在EW和UD向为4.43～7.59,3^#监测点NS向可达18.86;4)对比分析芦山仁家村(非震裂斜坡)监测剖面,认为珙县震裂斜坡存在放大效应。     

基于射线理论的垂直P波入射过程滑坡滑带应力放大特征及成因分析

汶川地震触发的最大滑坡——大光包滑坡是受控于斜坡先期层间构造错动带(软弱层)的大型地震滑坡, 该软弱层不但遭受了强烈的历史构造碎裂, 还在地震中产生了大面积新生震裂, 其成因及对大光包滑坡启动的贡献一直是国内外研究焦点。以大光包滑坡为地质原型, 将层间构造错动带概化为含软弱层单元地质体模型, 基于地震波射线理论, 建立了垂直P波入射过程考虑软弱层顶底界面波场转换和时间延迟的动力响应理论模型, 开展了垂直振动作用下含软弱层单元地质体振动台模型试验, 获得了软弱层应力放大特征及受地震强度和频率的影响规律, 揭示振动波在软弱层顶底界面的波场转换和能量分配产生了振幅衰减, 以及因软弱层与上下硬层介质属性差异造成的时间延迟, 共同导致应力分异和叠加, 促使软弱层应力放大; 从而认为, 强震过程中大光包滑坡先期层间构造错动带应力放大导致了带内岩体碎裂, 降低了滑带岩体抗剪强度, 从而促使强震过程中滑坡快速启动。      

基于流固耦合理论的巴东黄土坡滑坡地震响应研究

为研究三峡蓄水对湖北巴东黄土坡滑坡地震稳定性的影响,基于流固耦合理论建立二维有限差分数值模型。在计算无水(蓄水前)及有水(175 m实验性蓄水)条件下滑坡体场地地震动、应力响应及孔隙水压力差异特征的基础上,进一步给出滑坡体安全系数时程,探讨在蓄水与水库地震共同作用下的滑坡体稳定性及成因机理。结果表明：1)无水条件下滑坡体上硬层与下硬层的地震动峰值加速度放大系数约为1.33及3.31,具有显著的场地放大效应及软弱夹层消能减震作用;2)瞬时地震荷载下,由于滑坡体上下硬层之间非协调性变形与非快速排水的共同作用,软弱层超孔隙水压力快速形成,并呈现出阶段性积累上升趋势,伴有瞬时放大特征。     

基于流固耦合理论的巴东黄土坡滑坡地震响应研究北大核心CSCD

为研究三峡蓄水对湖北巴东黄土坡滑坡地震稳定性的影响,基于流固耦合理论建立二维有限差分数值模型。在计算无水(蓄水前)及有水(175 m实验性蓄水)条件下滑坡体场地地震动、应力响应及孔隙水压力差异特征的基础上,进一步给出滑坡体安全系数时程,探讨在蓄水与水库地震共同作用下的滑坡体稳定性及成因机理。结果表明:1)无水条件下滑坡体上硬层与下硬层的地震动峰值加速度放大系数约为1.33及3.31,具有显著的场地放大效应及软弱夹层消能减震作用;2)瞬时地震荷载下,由于滑坡体上下硬层之间非协调性变形与非快速排水的共同作用,软弱层超孔隙水压力快速形成,并呈现出阶段性积累上升趋势,伴有瞬时放大特征。     

Seismic response of the Lengzhuguan slope during Kangding Ms5.8 earthquake

In order to investigate the role of the amplification of peak ground acceleration(PGA) in seismic landslide formation mechanisms and study how earthquake waves interact with rock structures, a few strong-motion seismometers are installed at various locations on both sides of the Lengzhuguan gully. Five strong-motion seismometers were triggered at different depths in a tunnel at the same altitude during the Kangding Ms 5.8 earthquake on November 25 th, 2014. The data reveal that the horizontal peak acceleration(PGA_H) at each site decreased with increasing site depths. The PGAH at the deepest monitoring site(99 m from the tunnel entrance) was approximately half that of the outermost site. The amplitude of the acceleration response spectrum was also attenuated from the entrance inwards, the dynamic magnification factor(β) of the standard acceleration spectrum was less than 3.5, and rate of change was the same as that for the amplitude acceleration response. The Fourier spectra of each monitoring site also decreased from the outside inwards, and the components of the Fourier spectra were more complex at the surface.     

The starting mechanism and movement process of the coseismic rockslide: A case study of the Laoyingyan rockslide induced by the “5.12” Wenchuan earthquake

The"5.12"Wenchuan earthquake in 2008 triggered a large number of co-seismic landslides.The rear boundary or cracks of co-seismic landslide are generally located at the steep free surface of thin or thick mountains.Dynamic process of this kind of landslides could be divided into two parts:the seismic dynamic response of the slope and the movement process of rock mass.Taking the Laoyingyan rockslide as an example,the amplification effect was studied by single-degree-of-freedom system analysis method.Besides,the dynamic process of landslide under seismic loading was simulated by the finite difference method(FDM)and discrete element method(DEM).The amplification coefficient of the rockslide to seismic wave is 1.25.The results show that the critical sliding surface of the Laoyingyan rockslide was formed at the 23 th seconds under the action of seismic wave.At the same time,tension failure occurred at the rear edge of the sliding mass and shear failure occurred at the front edge.The maximum displacement was 0.81 m and the initial velocity was 2.78 m/s.During the initiation process of the rockslide,the rock mass firstly broke down along the joints which are along the dip of the rock stratum,and then collapsed bodily along the secondary structural planes.In the process of movement,the maximum velocity of rock mass was 38.24 m/s.After that,the rock mass underwent multiple collisions,including contact,deceleration to 0 and speed recovery after rebound.Finally,due to the constant loss of energy,the rocks stopped and accumulated loosely at the foot of the slope.The longest distance of movement was about 494 m.Besides,the smaller the damping ratio,the farther the rock mass moved.Compared with the results without considering the amplification factor,the movement distance of landslide by considering the amplification factor was more accurate.The study of the Laoyingyan rockslide is helpful to strengthen our field identification of potential co-seismic rockslides.At the same time,understanding its movement and accumulation process can help us better predict the hazard scope of the co-seismic rockslides,and provide a reference for the design of treatment projects.     

Dynamic process analysis of the Niumiangou landslide triggered by the Wenchuan earthquake using the finite difference method and a modified discontinuous deformation analysis

The Niumiangou landslide was the largest landslide triggered by the 2008 Wenchuan earthquake, which was significantly affected by the amplification effect of seismic acceleration. The ringshear experiments indicated that the materials in the source area of the Niumiangou landslide were subjected to friction degradation under a big shear displacement, which may result in rapid movement of the landslide. In order to better understand the landslide movement and study the effect of the friction degradation on movement mechanisms, the dynamic process of Niumiangou landslide was simulated with a new numerical method, which combines the finite difference method(FDM) and the discontinuous deformation analysis(DDA). First, the FDM was used to study the initiation time, amplification effect and velocity of the landslide. Afterwards, these initiation velocities were applied to the blocks in the DDA model by corresponding coordination in the FDM model. A displacementdependent friction model of the sliding surface was incorporated into DDA code to further understand the kinetic behavior of the landslide. The results show that the displacement-dependent friction strongly decreases the friction coefficient of sliding surface under a big displacement, which can obviously promote the run-out and velocity of landslide. The model output well matches the topographic map formed by the landslide. This implies that the proposed model can be applied to the simulation of earthquake-induced landslides with amplification effect, and the friction degradation model is important to clarify the movement mechanism of high-speed and long-distance landslides.     

Slope seismic response monitoring on the aftershocks of the Wenchuan earthquake in the Mianzhu section

Previous investigations have shown that the seismic response of slopes during the Wenchuan earthquake was highly variable.The present study tries to give an answer to the question:Which are the main factors affecting the seismic response degree of slopes? With the support of the China Geological Survey Bureau,we set 3 monitoring sections in Jiulong slope,Mianzhu city,China with the aim to record the site response of the slope during the aftershocks of the Wenchuan earthquake.After the Wenchuan earthquake,which happened on 12 May 2008,30 aftershocks have been recorded in these monitoring points.We analyzed 11 records,with magnitudes ranging from ML = 4.6 to ML = 3.1.The amplification factors of the horizontal compound PGA and 3D compound PGA have been determined for the 3 points at different elevations on the slope.Results showed that the dynamic response of the slope on the earthquake was controlled by factors such as topography and the thickness of the Quaternary overburden.     

Analysis of the effects of slope geometry on the dynamic response of a near-field mountain from the Wenchuan Earthquake

Fourier spectra and acceleration response spectra of near-field acceleration records of the 2008 Wenchuan Earthquake have been calculated. Relative fundamental frequencies (or predominant periods) were characterized. Then, the natural frequencies of a range of slopes with different geometric characteristics, such as height, slope ratio, and pattern, were analyzed. The seismic responses of the slopes were compared, and the variability of seismic response with the above geometric elements was found. Results show that if slope height increases, and provided that other conditions are unchanged, the natural frequency of the first mode of a double-surface slope will change as a power law. However, natural frequencies will diminish (based on a parabolic function) as the slope angle becomes large. Both the surface pattern and the number of surfaces on a slope can have a great impact on the seismic response of the slope. Moreover, within a certain range of slope heights or angles, either height or angle will also greatly influence the variability of the seismic response. The results of this research will be helpful to understanding seismic dynamic response features and explaining the ways that slope stability can be affected by earthquakes     

Seismic responses of the steel-strip reinforced soil retaining wall with full-height rigid facing from shaking table test

To investigate the seismic response of the steel-strip reinforced soil retaining wall with fullheight rigid facing in terms of the acceleration in the backfill, dynamic earth pressure in the backfill, the displacements on the facing and the dynamic reinforcement strain distribution under different peak acceleration, a large 1-g shaking table test was performed on a reduced-scale reinforced-earth retaining wall model. It was observed that the acceleration response in non-strip region is greater than that in potential fracture region which is similar with the stability region under small earthquake, while the acceleration response in potential fracture region is greater than that in stability region in middle-upper of the wall under moderately strong earthquakes. The potential failure model of the rigid wall is rotating around the wall toe. It also was discovered that the Fourier spectra produced by the inputting white noises after seismic wave presents double peaks, rather than original single peak, and the frequency of the second peak trends to increase with increasing the PGA (peak ground amplitude) of the excitation which is greater than 0.4 g. Additionally, the non-liner distribution of strip strain along the strips was observed, and the distribution trend was not constant in different row. Soil pressure peak value in stability region is larger than that in potential fracture region. The wall was effective under 0.1 g-0.3 g seismic wave according to the analyses of the facing displacement and relative density. Also, it was discovered that the potential failure surface is corresponds to that in design code, but the area is larger. The results from the study can provide guidance for a more rational design of reinforced earth retaining walls with full-height rigid facing in the earthquake zone.     

Dynamic response of a slope reinforced by double-row antisliding piles and pre-stressed anchor cables

Large-scale shaking table tests were conducted to study the dynamic response of a slope reinforced by double-row anti-sliding piles and prestressed anchor cables. The test results show that the reinforcement suppressed the acceleration amplification effectively. The axial force time histories are decomposed into a baseline part and a vibration part in this study. The baseline part of axial force well revealed the seismic slope stability, the peak vibration values of axial force of the anchor cables changed significantly in different area of the slope under seismic excitations. The peak lateral earth pressure acting on the back of the anti-sliding pile located at the slope toe was much larger than that acting on the back of the anti-sliding pile located at the slope waist. The test results indicate an obvious load sharing ratio difference between these two anti-slide piles, the load sharing ratio between the two anti-sliding piles located at the slope toe and the slope waist varied mainly in a range of 2-5. The anti-slide pile at the slope waist suppressed the horizontal displacement of the slope surface.