en
×

分享给微信好友或者朋友圈

使用微信“扫一扫”功能。
作者简介:

钟建华,男,1957年生,博士,教授,主要从事沉积地质学和构造地质学的科研工作;中国矿业大学(北京)煤炭资源与安全开采国家重点实验室任兼职研究员;E-mail:957576033@qq.com。

参考文献
杜远生, Shi G, 龚一鸣, 许亚军. 2007. 东澳大利亚南悉尼盆地二叠系与地震沉积有关的软沉积物变形构造. 地质学报, 81(4): 511~518.
参考文献
杜远生. 2011. 中国地震事件沉积研究的若干问题探讨. 古地理学报, 13(6): 581~590.
参考文献
杜远生, 余文超. 2017. 地震和非地震引发的软沉积物变形构造. 古地理学报, 19(1): 65~72.
参考文献
乔秀夫, 宋天锐, 高林志, 彭阳, 李海兵, 高劢, 宋彪, 张巧大. 1994. 碳酸盐岩振动液化地震序列. 地质学报, 68(1): 16~36.
参考文献
乔秀夫, 高林志, 彭阳, 李海兵. 2001. 古郯庐带沧浪铺阶地震事件、层序及构造意义. 中国科学(D辑), 31(11): 911~918.
参考文献
苏德辰, 孙爱萍, 郭宪璞, 王思恩. 2013. 青藏高原东北缘祁连山地区古地震记录与区域构造意义. 岩石学报, 29(6): 2223~2232.
参考文献
苏德辰, 乔秀夫. 2018. 黏性沉积物中的古地震触变流动变形. 古地理学报, 20(4) : 609~622.
参考文献
钟建华, 曹孟春, 倪良田, 孙宁亮, 刘闯, 郝兵, 杨冠群, 宋冠先, 葛毓柱. 2018. 砂脉的研究现状与进展. 古地理学报, 20(1): 119~132.
参考文献
钟建华, 倪良田, 汪立东, 宋全友, 孙宁亮, 宋冠先薛纯琦, 郝兵, 邵珠福, 毛毳, 刘闯, 曹孟春. 2020a. 青岛灵山岛南背来石下白垩统大型砂脉的发现及其形成动力学研究. 地质论评, 66(6): 1457~1476.
参考文献
钟建华, 王永卓, 倪良田, 高春文, 周永柄, 毛毳, 樊晓东, 刘国志, 陈志鹏, 宋冠先, 王瑞, 黄亮, 刘强, 范玉娇, 任凯旋. 2020b. 砂脉(砂注入体)型储层与油藏——以松辽盆地长垣地区西部萨葡夹层为例. 地球科学与环境学报, 42(3): 285~296.
参考文献
钟建华, 孙宁亮, 郝兵, 倪良田, 宋冠先, 薛纯琦, 毛毳, 邵珠福, 葛毓柱, 陈彬, 刘圣鑫, 彭超锋, 谷东辉, 王永强, 李聪. 2020c. 山东灵山岛北背来石剖面下白垩统顺层砂脉的特征及其成因动力学分析. 古地理学报, 22(4): 635~646.
参考文献
Allen J R L. 1982. simple models for the shape and syMetry of tidal sand waves: statically stable equilibrium forms. Marine Geology, 48: 31~49.
参考文献
Allen J R L. 1994. Sedimentary Structure, their character and physical basis, Ⅱ. New York Elsevier Scientific Publishing Company Amsterdam, Oxford.
参考文献
Castilla R A, Audemard F A. 2007. Sand blows as a potential tool for magnitude estimation of pre-instrumental earthquakes. Journal of Seismology, 11(4): 473~487.
参考文献
Monnier D, Gay A, Imbert P, et al. 2015. Sand injectites network as a marker of the palaeo-stress field, the structural framework and the distance to the sand source: Example in the Vocontian Basin, SE France. Journal of Structural Geology, 79: 1~18.
参考文献
Denolle M A, Shearer P M. 2016. New perspectives on self-similarity for shallow thrust earthquakes. Journal of Geophysical Research, 121 (9): 6533~6565.
参考文献
Du Yongsheng, Shi G, Gong Yiming, Xu Yajun. 2007&. Permian soft-sediment deformation structures related to earhquake in the southern Sydney Basin, eastern Australia. Acta Geologica Sinica, 81(4): 511~518.
参考文献
Du Yongsheng. 2011&. Discussion about studies of earthquake event deposit in China. Journal of Palaeogeography, 13(6): 581~590.
参考文献
Du Yuansheng, Yu Wenchao. 2017&. Earthquakecaused and non-earthquake-caused soft-sediment deformations. Journal of Palaeogeography, 19(1) : 65~72.
参考文献
Fan W, Shearer P M. 2016. Local near instantaneously dynamically triggered aftershocks of large earthquakes. Science, 353(6304): 1133~1136.
参考文献
Galli P, Meloni F, Rossi A. 1999. Historical liquefaction in Italy: relationship between epicentral distance and seismic parameters// European Geophysical Society XXIIII General Assembly Natural Hazards NH3, The Hague, Netherlands, 19~23
参考文献
Galli P. 2000. New empirical relationships between magnitude and distance for liquefaction. Tectonophysics, 324(3): 169~187.
参考文献
Group E W. 2013. Liquefaction phenomena associated with the Emilia earthquake sequence of May–June 2012 (Northern Italy). Natural Hazards & Earth System Sciences, 13: 935~947.
参考文献
Hanks T C, Simsson D A. 1976. Geophysical assessment of peak accelerations. Bulletin of the Seismological Society of America, 66(3): 959~968.
参考文献
Heron D, Etienne J L. 2005. A complex subglacial clastic dyke swarm, Sólheimajkull, southern Iceland. Sedimentary Geology, 181(1/2): 25~37.
参考文献
Holzer T L, Hanks T C, Youd T L. 1989. Dynamics of Liquefaction During the 1987 Superstition Hills, California, Earthquake. Science, 244: 56~59.
参考文献
Ishihara K. 1985. in Proceedings of the Eleventh Interational Conference on Soil Mechanics and Foundation Engineering, San Francisco, 12 to 16 August 1985 (Balkema, Rotterdam, 1985), 321~376.
参考文献
Jobe Z R, Lowe D R, Morris W R. 2012. Climbing-ripple successions in turbidite systems: depositional environments, sedimentation rates and accumulation times. Sedimentology, 59(3): 867~898.
参考文献
Sims J D. 1973. Earthquake-Induced Structures in Sediments of Van Norman Lake, San Fernando, California. Science, 182, 161~163.
参考文献
Jolly R, Lonergan L. 2002. Mechanisms and controls on the formation of sand intrusions. Journal of the Geological Society, 159(5): 605~617.
参考文献
Kuribayashi, E. T. Iwasaki, F. Tatsuoka. 1975. Effects of stress—strain conditions on dynamic properties of sands. Soils & Foundations. 242: 105~114.
参考文献
Lowe D R. 1975. Water escape structures in coarse-grainedd sediments. Sedimentology, 22: 157~204.
参考文献
Moretti M, Alfaro P, Caselles O, et al. 1999. Modelling seismites with a digital shaking table. Tectonophysics, 304(4): 369~383.
参考文献
Nichols R J, Sparks R S J, Wilson C J N. 2010. Experimental studies of the fluidization of layered sediments and the formation of fluid escape structures. Sedimentology, 41(2): 233~253.
参考文献
Obermeier S F, Bleuer et al. 1991. Evidence of strong earthquake shaking in the lower Wabash Valley from prehistoric liquefaction. Science, 251: 1061.
参考文献
Obermeier S F. 1996. Use of liquefaction-induced features for paleoseismic analysis; an overview of how seismic liquefaction features can be distinguished from other features and how their regional distribution and properties of source sediment can be used to infer the locatio. Engineering Geology, 44(1): 1~76.
参考文献
Obermeier S F. 1998. Liquefaction evidence for strong earthquakes of Holocene and latest Pleistocene ages in the states of Indiana and Illinois, USA. Engineering Geology, 50(3~4): 227~254.
参考文献
Obermeier S F, Olson S M, Green R A. 2005. Field occurrences of liquefaction-induced features: a primer for engineering geologic analysis of paleoseismic shaking. Engineering Geology, 76(3~4): 209~234.
参考文献
Owen G, Moretti M, Alfaro P. 2011. Recognising triggers for soft-sediment deformation: Current understanding and future directions. Sedimentary Geology, 235(3~4): 133~140.
参考文献
Papathanassiou G, Pavlides S, Ganas A. 2006. The 2003 Lefkada earthquake: Field observations and preliminary microzonation map based on liquefaction potential index for the town of Lefkada. Engineering Geology, 82(1): 12~31
参考文献
Qiao Xiufu, Song Tianrui, Gao Linzhi, Peng Yang, Li Haibing, Gao Mai, Song Biao, Zhang Qiaoda. 1994&. Seismi sequence in carbonate rocks by vibrational liquefaction. Acta Geologica Sinica, 68(1) : 16~36.
参考文献
Qiao Xiufu, Gao Linzhi, Peng Yang, Li Haibing. 2001&. Seismic event, sequence and tectonic significance in Canglangpu Stage in Paleo-Tanlu Fault Zone. Science in China(Series D), 31(11): 911~918.
参考文献
Quigley M C S. Bastin, B A Bradley. 2013. Recurrent liquefaction in Christchurch, New Zealand, during the Canterbury earthquake sequence. Geology, 41: 419~422.
参考文献
Rodríguez-Pascua, M A Calvo J P, Vicente G D, et al. 2000. Seismites in lacustrine sediments of the Prebetic Zone, SE Spain, and their use as indicators of earthquake magnitudes during the Late Miocene. Sedimentary Geology, 135(1~4): 117~135.
参考文献
Seed H B, Tokimatsu L F, Harder L F, Chung R M J. 1985. Influence of spt procedures in soil liquefaction resistance evaluations. Geotech. Eng, 111: 1425.
参考文献
Shani-Kadmiel S M, Tsesarsky J N, Louie Z, Gvirtzman. 2012. Bulletin of the Seismological Society of America, 102(4): 1729~1739.
参考文献
Shani-Kadmiel S M, Tsesarsky J N, Louie Z, Gvirtzman. 2014. Bulletin of earthquake engineering. 12: 607~625.
参考文献
Simons M. S. E. Minson A. Sladen et al. 2011. The 2011 Magnitude 9. 0 Tohoku—Oki Earthquake: Mosaicking the Megathrust from Seconds to Centuries. Science, 332: 1421~1425.
参考文献
Su Dechen, Sun Aiping, Guo Xianpu, Wang Sien. 2013&. Seismites in the Qilianshan area of the northeastern margin of Qinghai—Tibetan Plateau and a pilot research on the regional tectonics. Acta Petrologica Sinica, 29(6): 2223~2232.
参考文献
Su Dechen, Qiao Xiufu. 2018&. Palaeoseismic thixotropic flow deformation in viscous sediments. Journal of Palaeogeography, 20(4) : 609~622.
参考文献
Sukhija B S, Rao M N, Reddy D V, Nagabhushanam P, Lakshmi BV, Gupta H K. 2000. Current Status of Paleoseismology——An overview. Journal of Geological Society of India, 55: 465~480.
参考文献
Tinsley J C, Youd T L, Perkins D M, Chen A T F. 1985. An Earth-Science Perspective. US Geological Survey Denver, CO US Geological Survey Professional Paper. 1360: 263~316.
参考文献
Warren J K. 2010. Evaporites through time: Tectonic, climatic and eustatic controls in marine and nonmarine deposits. Earth-Science Reviews, 98(3): 217~268.
参考文献
William L E. 2013. Injection-Induced Earthquakes. Science, 341(12): 1~7.
参考文献
Zhong Jianhua, Cao Mengchun, Ni liangtian, Sun Ningliang, Liu Chunag, Hao Bing, Yang Guangqun, Song Guangxian, Ge Yuzhu. 2018&. Situation of study and development tendency of sandy dykes. Journal of Palaeogeography, 20(1): 119~132.
参考文献
Zhong Jianhua, Ni Liangtian, Wang Lidong, Song Quanyou, Sun Ningliang, Song Guangxian, Xue Chunqi, Hao Bing, Shao Zhufu, Mao Cui, Liu Chuang, Cao Menchun. 2020a&. Discovery of large sand dikes in the Lower Cetaceous of southern Beilaishi, Lingshan Island, Qingdao, and their formation dynamics. Geological Review, 66(6): 1457~1476.
参考文献
Zhong Jianhua, Wang Yongzhuo, Ni Liangtian, Gao Chuanwen, Zhou Yongbin, Mao Cui, Fan Xiaodong, Liu Guozhi, Chen Zhipeng, Song Guangxian, Wang Rui, Huang Liang, Liu Quang, Fan Yujiao, Ren Kaixuan. 2020b&. Sand-dike (Sand injection) type of Reservoir and Oil Pool. Journal of Earth Science and Environment, 42(3): 285~296.
参考文献
Zhong Jianhua, Sun Ningliang, Hao Bing, Ni Liangtian, Song Guanxian, Xuan Chunqi, Mao Cui, Shao Zhufu, Ge Yuzhu, Chen Bin, Liu Shengxin, Peng Shaofeng, Fu Donghui, Wang Yongqiang, Li Cong. 2020c&. Characteristics and genetic dynamics analysis of Lower Cretaceous bedding sand veins in the Laishi Section of Lingshan Island, Shandong Province. Journal of Palaeogeography, 22(4): 635~646.
目录contents

    摘要

    地震液化阈值是一个非常重要的科学问题,一般认为M5地震不会形成大面积的液化。2009年12月14日,中国新疆哈密市发生了M5.1中等地震,震源深度仅4 km。地震砂脉网格在平面从几十厘米到2 m以上;砂脉纵断面呈楔形、倾斜(平均75.10°),分选较好。通常情况下,它们通过液化和流化分异作用发生在细粒丰富的盐渍砂砾层(SSGL,Salinization Sand—Gravel Layer)和盐粒砂砾层(SGSGL,Salt-Grain—Sand—Gravel Layer)中,尽管没有细粒盖层和源砂,但这些盐渍的砂砾层极易在盐溶后发生活化,颗粒之间的摩擦力骤降,液化上涌而形成砂脉。液化边界距震中可达80 km,甚至可能达到120 km,相当于M7.0~8.0级地震的液化最远距离。哈密地区之所以能在M5.1地震作用下形成远程砂脉,主要由于以下5个优势:① 砂泥的盐溶液中细粒组分容易发生液化流化。浓盐水能够降低颗粒的剪切能力,平均降低25%~75%左右,使地震液化阈值降低到 0.15~0.05 g(以0.2 g为一般阈值)。与此同时,浓盐水由于密度大,盐水可使淡水最小流化速度(Umf)降低12.51%~21.58%,有利于流化。 ② 广泛分布的盐渍砂砾层和盐粒砂砾层。 ③ 震源极浅(深度仅4 km)。 ④ 基底极浅(深度0~3 m)。 ⑤ 表层盐屑混合盖+盐渍砂砾层+盐粒砂砾层+极浅基岩基底组成了特殊的三明治结构。通过对液化流化的形成机理研究表明,砂主要来自砂脉底部的砂砾层的流化分选,流化分选会在砂脉底部的砂砾层中形成一个分选晕。

    Abstract

    Objectives: The earthquake liquefaction threshold is a very important scientific problem. It is generally believed that the earthquake of magnitude 5 will not result in large-scale liquefaction. On December 14, 2009, a moderate 5.1 magnitude earthquake struck Hami City in western China’s Xinjiang Province, with a focal depth of only 4 kilometers (Figure 1). This paper is to discuss the mechanism of extensive liquefaction caused by Hami 5.1 earthquake.

    Methods:The authors made a detailed field investigation in Hami, Xinjiang, and obtained a lot of valuable geological information, especially sand dikes and obtained a lot of valuable salt solution activation data through the laboratory fluidization experiment and salt solution activation experiment, which provided valuable data for further study of liquefaction activation of saline solution.

    Results: The liquefaction boundary can be 80 kilometers away from the epicenter, and may even reach 120 km, which is equivalent to the epicenter distance of a 7.0~8.0 magnitude earthquake. It is undeniable that the formation of Hami sand dikes, especially the unusually significant liquefaction and fluidization. Seismic sand dikes grid plane from tens of centimeters to more than 2 meters; The longitudinal section is wedge-shaped and oblique (average 75.10°), and the sorting is good. The dip Angle of sand dikes ranged from 52° to 90°, with an average of 75.4° (250 sand dikes were taken as examples), and more than 96% of sand dikes had steep dip Angle (inclination >60°). The Hami sand dikes are convex in both plane and section, but are flat and even grooved in some places due to flood erosion. The protrusion height of hami sand dikes is generally 2~3 cm ~ 5~6 cm, and the maximum is 10 cm. The experimental results show that under the condition of salt water concentration of 23% and 9%~17%, the shear stress inside sand dikes can be greatly reduced by salt solution, which is 52.63%~85.20% and 12.51%~21.58%, respectively, which is conducive to the formation of sand dikes by seismic liquefaction and fluidization. An unusual assemblage of sedimentary facies and basement (sandwich-like assemblage) is thought to facilitate large-scale liquefaction. In addition, solid salt dissolves to form a thinner fluid layer that can form an overpressured fluid on its own without seismic liquefaction.

    Conclutions: ① A large number of sand dikes are developed in the study area, which are continuously distributed, medium—small scale, polygon on the plane and wedge on the section; most of it leans. It is believed to have been a moderate earthquake of M5.1 on December 14, 2009. ② Due to the shallow basement, thin cap layer, and intense salinization, Hami, under the very limited shock impact of such a moderate intensity earthquake just entering the liquefaction threshold (M5.1), formed a wide distribution of seismic liquefaction sands extending at least 80 km outward from the epicenter, and possibly up to 120 km. ③ The good characteristics of dike sorting are definitely due to the liquefaction and fluidization sorting of SSGL and SGSGL, rather than a pure parent sand unit. Through fluidization experiment and calculation, fluidization distribution and fluidization boundary diagram of fluidization sand with low particle size are obtained. ④ Experimental studies on the minimum fluidization velocity (Umf) of particles <0.125~40 M are reported, and two empirical formulas are proposed: Umf=6.612×D0.6277 and Umf=7.7443×D0.6293. The maximum fluid excess pore pressure and seismic shock pressure are calculated to be about 513.448~637.29 kg/m2. They are quite different from the classical formulas. The maximum fluid overpressure and seismic impact pressure (about 513.448~637.29 kg/m2) were calculated from the overpressure values of 40 M gravel and the elevation of sand dikes. ⑤ The study shows that the sand forming sand dikes mainly comes from the fluidization sorting of the sand and gravel layer below the bottom of the dike. ⑥ The reason why Hami can form long distance sand veins under M5.1 earthquake is mainly due to the following six advantages: ① The salt solution of sand mud sand is easy to liquefy and fluidize. Concentrated brine and semi-dissolved salt can reduce the shear capacity of particles by 25%~75% on average, and the seismic liquefaction threshold is reduced to 0.15~0.05 g (with 0.2 g as the general threshold). At the same time, with the increase of density, the minimum fluidization velocity (Umf) of fresh water can be reduced by 12.51%~21.58%, which is conducive to fluidization. ② Widely distributed SSGL and SGSGL; ③ The source is very shallow (only 4km deep); ④ very shallow basement (depth of about 0~3 m); ⑤ The special sandwich structure of the salt—debris-mixed-cover +SSGL+SGSGL+ extremely shallow bedrock basement. Based on the excess pore pressure value of 40M gravel and the height of sand veins above the ground, the maximum fluid excess pore pressure and seismic shock pressure are calculated to be about 513.448~637.290 kg/m2.

    关键词

    砂脉液化盐溶活化流化地震哈密

  • 地震及其液化一直是地球科学研究的热门话题(Sims et al.,1973; Holzer et al.,1989; Obermeier et al.,1991; 2016; 杜远生等,200720112017乔秀夫等,19942001William,2013; 苏德辰等,20132018钟建华等,20182020a,b,c)。Sims(1973)发现地震可以形成软沉积物变形构造(SSDS,soft-sediment deformation structure),揭开了液化与地震之间关联性。近年来,人们对由已知地震产生的液化或流化(Owen et al.,2011; Simons et al.,2011; Quigley et al.,2013;Group,2013)引起的SSDS的兴趣激增。这些现象已被广泛应用于识别古地震和沉积学、构造、构造物理以及岩土工程。作为识别地震和非地震变形的标准,最常见的SSDS是砂岩岩脉。然而,砂脉的形成机制仍有许多不确定性,也很容易由其他因素引起(Heron,2005Sukhija et al.,2000; Damien et al.,2015;杜远生等,2017钟建华等,20182020a,b,c)。到目前为止,还没有人研究过在M5.1地震形成的砂砾层的液化及砂脉成因机制的问题,为此,本研究有3个目的:首先,详细记录和描述盐渍砂砾层(SSGL, Salinization Sand—Gravel Layer)和盐粒砂砾层(SGSGL, Sand—Gravel-Bearing Salt Crust)中以砂脉为代表的广泛液化(约12000 km2);其次,以2009年12月14日哈密M5.1地震为例,研究中震级地震形成砂脉的异常动力学过程和形成机制;第三,笔者等初步阐明了液化和流化过程中盐溶解活化效应的机理。

  • 图1 新疆哈密2019年地震位置及砂脉出露点: (a)哈密及其外围的1737年~2014年地震分布图;(b)1737年~2014年新疆哈密地区地震群及2009年及其他时期砂脉的空间分布

  • Fig.1 Earthquake location and exposed location of sand dikes in Hami, Xinjiang: (a) seismic distribution map of Hami and its periphery from 1737 to 2014; (b) the spatial distribution of sand-dike exposure in the Hami area of Xinjiang during 1737~2014 and in 2009 and other periods

  • 1 地质背景

  • 笔者等本文的重点研究区域在哈密市雅满苏镇(图1),研究区域长约120 km,宽约120 km,近NW—SE(图1b)。哈密地区发育了干旱气候条件下的辫状河沉积。沉积了一套分选差的砂砾、化学沉积岩以及风成砂等,这套沉积是砂脉发育的主要物质基础。这套辫状河沉积位于干旱地区,一年中大部分时间都是干旱少雨。由于具有广阔的补给区,在雨季时,会遭到严重的洪水淹没,大量的碎屑等风化物质被冲入河道和砂滩,导致石炭系黑色变质岩、红褐色泥质砂岩和新近系泥质砂砾岩等以不整合的方式快速沉积。河道主要被0~3 m厚(多为1~2 m)的冲积层所充填,冲积层主要为砂砾层,有少量的砂层和泥层。发育砂脉的砂砾层主要有以下3种类型:第一种是SSGL,含盐较少,盐呈细晶状,在蒸发表面成薄膜状,松散易碎(图2a;图3a—d,f,h,j),广泛地发育在戈壁上;第二种是盐粒砂砾层(SGSGL)(图2b—e),含盐中等,明显可见盐颗粒,盐粒直径从不足1 cm到2 cm左右,呈颗粒状产于砂砾之间(图2b、c中的红色箭头所指);但经常呈盐盖状发育,较坚硬,外形呈屋檐状的垂悬(图2d,e),可形成超压圈闭;第三种是含砂砾盐壳(SGBSC,Sand—Gravel-Bearing Salt Crust),或含砂砾盐盖(SGBSC,Sand—Gravel-Bearing Salt Cover),或含砂砾盐层(SGBSC,Sand—Gravel-Bearing Salt Layer),含盐很高,外观上可以见到非常明显的盐块;砂砾含量相对较少,在20%~50%之间(图2f),多呈悬浮状发育在盐块中,整体坚硬如石。其中砂脉少发育,一旦发育规模就很大。当砂砾含量很少时,含砂砾岩壳又可以成为盐壳,此时砂脉分布非常广泛,几乎遍及全区,无论是在地势较低的河床上,还是在地势较高的戈壁上均可以见到,但砂脉主要发育在前2种砂砾层中。以下简要介绍其特点。

  • 2 砂脉的几何学及其结构构造特点

  • 研究区发育了大量裸露的连续性很好的砂脉(图3;图4),并且可以大范围直接进行三维观察,为详细研究砂脉提供了良好的条件。按产出环境可分为两大类,一类是产于戈壁上的砂脉(图3);另一类是产在水洼中的砂脉(图4),其特点主要是:

  • (1)砂脉主要发育在SSGL和SGSGL中,底部无单一的砂源,液化层上缺乏细粒盖层(图3;图4),缺少发育砂脉的基本条件:即底部有可供液化的砂源和砂源顶部有可形成高压流体的圈闭层。这是哈密砂脉成因的最典型特殊之处,颠覆了地震液化砂脉形成的基本理论。

  • (2)砂脉形态在平面上呈多角形或多边形(图3a、b、d;图4a、b、d),在剖面上,它们多半呈楔形(图3f、g、h);图4c、f),大致对称(图4b、c)或明显不对称的形态(图3f、h)。砂脉高度、砂脉宽度与砂脉网格直径之间存在明显的正相性,在研究区的任何位置测得的砂脉的倾角分布范围大。笔者等对比了砂脉参数与距震中距离的关系,总体来看,从震中到80 km左右砂脉高度和倾角逐渐增大,到80 km以外砂脉高度和倾角逐渐减小。

  • (3)与砂脉壁相邻的砂砾层见陡倾或接近直立层,砂砾叠瓦状结构(图3h、i),砾石有时与砂脉外缘平行,形成薄的含砾边缘层;靠顶部有时倒转(图3i)中的红色椭圆),表明了砂脉形成过程中,液化与流化砂和水向上流动形成了持续向上牵引,使两侧砂砾层紧靠砂脉的砾石在牵引力作用下,发生移动变位,呈倾斜状;也揭示了砂脉两侧的砂砾层在砂脉形成过程中呈软化甚至液化状态。

  • (4)大多数砂脉内部发育了结构或组构,无论在平面和垂向视图中,砂脉内部具有条带构造:中部为灰色,外部为褐色。在挖开的平面和垂直视图中,砂脉有2~3个灰色和黄褐色色带(图3e)。除条带构造外,多数砂脉内部具有片状构造(图3h、i)。此外,还有一个重要的特征,砂脉中间带的粒度较大,边缘带的粒度较小,这说明砂脉的初始能量大,中带的流化速度(Umf)较大,边缘带的流化速度(Umf)较小。

  • (5)绝大部分砂脉呈倾斜状,砂脉的倾角范围为52°~90°,平均为75.10°(250个砂脉统计),96%以上砂脉倾角陡(倾角>60°)(图3f)。

  • (6)哈密砂脉在平面和剖面上均呈上凸形(图3a—c,j;图4a—d),砂脉突出高度一般为数厘米,最大可达13 cm(图3j),由于洪水的侵蚀,有些地方表面平整甚至有沟槽。

  • (7)砂脉组分以细—粗砂为主(图3e—k;图4b—d,f),与周缘的砂砾层相比,分选明显要好。这些细粗砂主要由石英、长石和方解石及岩屑组成,它们来自于流化分异或流化分选(后文详述)。

  • 图2 发育砂脉的3种含盐量不同的砂砾层

  • Fig.2 Three different sand and gravel layers with different salt content developed sand dikes

  • (a)SSGL含盐较少,盐粒呈微晶状,较松散易碎,其中发育了大型楔状砂脉,高近1 m,最大宽度达35 cm。雅满苏风力发电厂附近。(b)盐粒砂砾层(SGSGL),明显可见盐颗粒,盐呈颗粒状产于砂砾之间(红色箭头)。较松散易碎。其中发育了中型楔状砂脉,高约0.3 m,最大宽度达13 cm,大山口火车站附近。(c)盐粒砂砾层(SGSGL),明显可见盐颗粒(红色箭头),盐呈颗粒状产于砂砾之间(红色箭头),直径最大可达 1 cm,其中还发育了一些小孔(蓝色箭头),较坚硬,大山口火车站东15 km处。(d)盐粒砂砾层(SGSGL),明显可见盐颗粒,较硬,其中发育了大型楔状砂脉,高1m左右,最大宽度达25 cm,哈密南湖发电厂附近。(e)盐渍砂砾层(SSGL),表面见盐膜,较松散易碎,其中发育了大型板片状砂脉,高0.62 m,最大宽度达20 cm,底部有一团砂根(蓝色箭头,地质锤所指)。哈密南湖发电厂附近。(f)含砂砾盐壳(Sand—Gravel-Bearing Salt Crust),含20%~30%的砂砾,厚20~40 cm。砂砾少时也可以归属于盐壳(Salt Crust),坚硬如石。靠哈密南湖发电厂附近

  • (a) SSGL (Salinization Sand—Gravel Layer) contains less salt and its salt grains are microcrystalline. More loose and brittle. Large wedge-shaped sand veins with a height of nearly 1 m and a maximum width of 35 cm were developed. Near Yamansu Wind Power Plant. (b) The salt-grained sand and gravel layer (SGSGL) , where salt particles are evident and occur in a granular form between the sand and gravel (red arrow) . More loose and brittle. The medium wedge-shaped sand veins, about 0.3 m high and 13 cm wide, were developed. Near Dashankou Railway Station. (c) The SGSGL, where salt particles (red arrow) are visible, and the salt is in a granular form between the sand and gravel (red arrow) , up to 1 cm in diameter. There are also some small holes developed in it (blue arrows) , which are harder.15 km east of Dashankou Railway Station. (d) SGSGL, where salt particles are evident, harder. Large wedge-shaped sand veins with a height of about 1 m and a maximum width of 25 cm were developed. Near the Hami South Lake Power Plant. (e) SSGL with salt film on the surface. More loose and brittle. Large lamella sand veins, 0.62 m high and up to 20 cm wide, are developed, with a mass of sand roots at the bottom (blue arrow, geological hammer) . Near the Hami South Lake Power Plant. (f) Sand—gravel bearing salt crust, containing 20% to 30% gravel. It is 20~40 cm thick. Less gravel can also be classified as salt crust; as hard as stone. Near the Hami South Lake Power Plant

  • 以上简要介绍了哈密2009年M5.1地震形成的砂脉的主要特点,实际上哈密地区还有一些埋藏的老砂脉,甚至在花岗岩和变质岩中都有砂脉,所以砂脉的特点和形成机制是非常复杂的,有待今后报道。

  • 3 讨论

  • 3.1 SSGL和SGSGL液化和流化的盐溶效应

  • 砂脉必须有液化和流化共同作用才能形成,但液化和流化具有差异性(Lowe,1975)。长期以来,人们认识到的一个问题:砂源、细粒盖层、饱和度、超孔隙压力和一定震级的地震波动和晃动对砂脉的形成至关重要。经典理论强调了砂脉形成中液化最容易发生在粗粉砂到具有细粒盖层的细砂中(Moretti et al.,1999;Crespellani et al.,2001; Owen et al.,2011)。但与具有细粒盖层的粗粒砂脉相比,没有细粒盖层的粗粒砂脉的研究较少。笔者等通过研究无细粒盖层和无纯砂源的特殊条件下的SSGL和SGSGL液化,表明其原因与盐溶活化密切相关。

  • 图3 新疆哈密戈壁上的SSGL和SGSGL中的砂脉

  • Fig.3 Sand dikes in SSGL and SGSGL in gobi of Hami area, Xinjiang

  • (a)在重盐渍化砂砾层中形成的多边形砂脉的野外照片。它们是在2009年的M5.1地震中形成的,而直线砂砾脊是在2005年~2006年的道路施工过程中由一台前装载机产生的。所以也揭示了砂脉是2005年~2006年以后形成的。(b)平原上许多细微的、随机方向的砂脉,形成多边形特征(直径50~100 cm),有低、宽的脊(5~6 cm高,约10~2 cm宽)。表面覆盖2~3 mm厚的盐膜。(c)盐渍化的砂砾层的剖面上可以见到大量砂脉;间距平均1 m左右,多微倾斜。(d)多边形砂脉网开挖平面图(深度约5~10 cm)。(e)一个三联砂脉的顶部平面视图。它可以分为3个区域。(f)垂直沟槽图,显示砂质砾石主体中陡峭倾斜的砂脉,与砂质砂源连通性好,表明砂脉呈楔形倾斜。(g)一个垂直的壕沟视图显示了一个玻璃状的砂脉在一个泥岩宿主。砂脉中上段相邻的夹层细粒砾石和细粒砂层的上翻表明注入砂强烈的向上挤压作用。(h)砂脉两侧的砾石被牵引倾斜(黄色椭圆),砂脉中发育了层状或片状组构(蓝色线)。(i)砂脉两侧的砾石被牵引倾斜(黄色椭圆和蓝色箭头所指)、甚至倒转(红色椭圆),砂脉中发育了层状或片状组构(蓝色线)。(j)竖直面上可见砂脉凸出成脊状,比两侧分别高13 cm和8.5 cm。(k)发育在弱SGSGL中的砂脉的三维图,可见呈楔形(红色箭头),由分选很好的中粗砂组成。在其右侧还有一条发育不完善的山脉(蓝色箭头)

  • (a) Field photograph of polygonal sand dikes formed in heavily salinized sand and gravel beds. They were formed during a 5.1 magnitude earthquake in 2009, while the straight grit ridges were created by a front loader during road construction in 2005~2006. So it also reveals that the sand dikes formed after 2005~2006. (b) Many fine, randomly oriented sand dikes in the plain, forming polygonal features (50~100 cm in diameter) with low, broad ridges (5~6 cm high and about 10~20 cm wide) . The surface is covered with 2~3 mm thick salt film. (c) A large number of sand dikes can be seen on the profile of salinized sand and gravel beds. The average distance is about 1 m, more than slightly inclined. (d) Excavation plan of polygonal sand dike network (depth about 5~10 cm) . (e) Top plane view of a perfect triple sand dike, it can be divided into three areas. (f) Vertical grooves showing steeply sloping dikes in the sandy gravel body with good connectivity to the sandy sand source, indicating a wedge dip in the dikes. (g) A vertical trench view shows a glassy dike of sand in a mudstone host, the upturning of adjacent interbedded fine-grained gravel and fine-grained sand layers in the middle and upper sections of the dike indicates strong upward compression of the injected sand. (h) The gravel on both sides of the dikes is drawn and tilted (yellow ovals) , and layered or lamellar fabric develops in the dikes (blue lines) . (i) The gravel on either side of the dikes is drawn to tilt (yellow ovals and blue arrows) or even reverse (red ovals) , and layered or lamellar fabric develops in the dikes (blue lines) . (j) Ridge-like sand dikes can be seen on the vertical plane, 13 cm and 8.5 cm higher than on both sides, respectively. (k) Three-dimensional view of sand dikes developed in weak SGSGL, visible as wedge-shaped (red arrow) , consisting of well-sorted medium—coarse sand, there is also a poorly developed sand dike to the right (blue arrow)

  • 3.1.1 盐溶的液化活化效应及其机理。

  • (1)盐溶的液化活化效应。盐溶作用最显著的特征是盐溶的活化作用。未含盐的砂砾层是惰性的,颗粒与颗粒之间是直接紧密接触的(图5a),要发生液化的难度很大。如果砂砾层被盐水饱和,情况就会发生巨大变化,最大的变化是会发生盐溶活化,可分为两种过程或两种机制:一种是盐结晶胶结;另一种是盐溶解。盐结晶胶结是通过结晶形成薄膜或盐颗粒来扩大细颗粒之间的距离,使颗粒断开时形成大量孔隙(图2c;图5b),砂砾的体积加大,图5中的b比a的厚度增加了H1。在野外可以观察到被盐渍化的砂砾层明显隆起数厘米,当在其上行走时会踩出深数厘米的坑穴,而没有被盐渍的砂砾层就不会有这种现象。

  • (2)模拟实验。为获得盐溶对液化的影响程度和产生机理,笔者等设计了盐溶模拟实验。设置的盐溶模拟实验时充分考虑了哈密SSGL和SGSGL实际情况(图2),用超细盐粉与各种粒度的砂和细砾混合成固—固混合物,设定了盐含量9%、17%和23% 3种比例(图6),充分搅拌后加饱和水浸泡。获得了3种浓度的盐水(可能有部分未溶解的盐粒)与砂砾的混合物;9%的盐砂混合物的盐水浓度相当于盐渍砂砾层的盐水浓度(图2a;图3a—c),17%的盐砂混合物的盐水浓度相当于小盐粒砂砾层的盐水浓度(图2b);23%的盐砂混合物的盐水浓度接近于饱和盐水浓度,但可能低于大盐粒砂砾层的盐水浓度(图2c)。观察表明,一旦遇水盐颗粒会迅速溶解到水中形成高浓度或饱和盐水层,哈密含盐很高的SGSGL可能还有部分盐颗粒因为盐水饱和而不被溶解,但盐颗粒边缘被半溶解软化(图5c),导致大多数砂粒彼此不接触,一些细颗粒失去了部分支撑(即剪应力)呈悬浮状,有效应力大幅降低,孔隙流体发生超孔隙压力,出现原位静止液化(图5c、d)。这就是盐溶活化机理。

  • 图4 哈密大山口水洼中的砂脉及其内部结构图

  • Fig.4 Sand dikes in Dashankou, Hami and its internal structure

  • (a)哈密大山口水洼中的砂脉部分全景图;宽3~20 cm;高数厘米;中心或边缘多有一条宽约1 cm的裂缝;季节性洪水使盐渍层溶解,晒干后会返盐。(b)图(a)中红框的放大图,可见楔形砂脉,宽13 cm,高29 cm。主要由中砂组成,脉中心为粗砂,并含少量细砾,呈直立形态。(c)图(b)的红框放大图,可见呈楔形,砂脉中心有一垂向裂缝。底部为细砾层,发育了一个流化分选晕(蓝色虚线)。(d)砂脉的平面与剖面,由于流水侵蚀作用,砂脉在平面上不大明显(细红色箭头);剖面上砂脉明显(粗红色箭头),中心多有一条裂缝,呈微倾斜状。(e)考察队正在挖探槽,水洼中央的砂脉被泥层覆盖,但其下砂脉发育的很好。远处可见盐渍(红色箭头)。(f)图是(e)中挖出的探槽中见尖楔形的砂脉(红色箭头)和两次洪水沉积的泥层覆盖(Ⅰ、Ⅱ)

  • (a) Panoramic view of sand dikes in pools at Dashankou, Hami; 3~5 cm to nearly 20 cm wide; a few centimeters high; an extra slit about 1 cm wide in the center or edge; seasonal flooding dissolves the salt layer, which dries and returns to salt. (b) An enlarged view of the red box in the Figure (a) shows a wedge of sand dikes, 13 cm wide and 29 cm high. It is mainly composed of medium sand, and the center of dikes is mainly coarse sand with a few fine gravel. Stand upright. (c) Enlarged view of the red box in the figure (b) shows a wedge shape with a crack in the center. At the bottom is a fine gravel layer with a fluidized sorting halo developed (blue dashed line) . (d) The plane and section of the sand dikes, which are less visible on the plane due to erosion by running water and sedimentation (fine red arrows) . In the section, the sand dikes are obvious (red arrow) , and there is a crack in the center, slightly tilted. (e) The team is digging a trough. The central dike of the bubble has been covered with mud, but the dike below is well developed. Visible in the distance saline (red arrow) . (f) Sharp-wedged sand dikes (red arrows) in the trench dug in Figure (e) and mud overlays deposited by two floods (Ⅰ, Ⅱ)

  • 笔者等用应力计测定了盐溶砂砾层颗粒之间的内聚力或剪切力。盐溶(包括半溶解盐颗粒)可以大大降低砂层和砾石层内部的内聚力或剪切力,在20%~80%(图6a—e)。笔者把这种盐溶使砂层和砾石层颗粒的内聚力或剪切力减小效果称为“盐溶效应”。具体如下:①盐溶效应与盐水浓度有关,一个简单的规律是盐水浓度越高盐溶效应越大,如果把淡水的盐溶效应定为0的话,那么盐水浓度9%时,细砾和细中粗砂的盐溶效应为25%~75%(图6a—e),但粉砂+黏土除外(图6f),可能与细颗粒,尤其是黏土之间的吸附有关;盐水浓度17%和23%时,细砾和中粗砂的盐溶效应略大于25%~75%(图6a—e),但粉砂+黏土除外(图6f);②盐溶效应与颗粒粒度密切相关,细砾的盐溶效应最好,约75%(图6a);粗砂的盐溶效应在65%左右(图6b、c);粗细砂的盐溶效应在60%左右;但更令人惊讶的是,细粉砂级颗粒不是盐溶活化液化的敏感组分(图6f),可能与细粉砂和黏土表面的凝聚力较大有关,但Seed(1985)等用液化粉砂的穿透实验来确定(美国)震中的位置,表明了粉砂对地震液化的响应性很好。当地震液化阈值在0.20 g时(Obermeier,1996;1998),盐溶会显著降低地震液化阈值,使地震液化阈值在0.15~0.05 g(另文讨论),盐渍化地区极易发生地震液化,这就是盐溶液化活化的机理。笔者把这个过程称为“盐溶—液化—活化效应”,简称“SDLME”(Salt-Dissoved—Liquefaction—Mobilization—Effection),其机理主要与高浓度盐水悬浮降低了颗粒的有效应力而增加了流体压力有关。

  • 图5 盐粒半液化的形成示意图

  • Fig.5 Sketch for the salt semi-liquefaction

  • (a)不含盐的干砂砾层。黄色圆形球代表砂粒,颗粒间剪切力较大,内聚力较强;可以承受很大的有效应力。(b)含盐的砂砾层。白色的点是盐颗粒。盐结晶使砂粒断开和“膨化”现象,颗粒之间的距离明显增大,高度增加了H1。(c)盐颗粒部分溶解的状态。大多数颗粒已被半溶解盐颗粒断开,如在大部分可液化的沉积物中,约65%~78%。盐粒外围的红色圆圈表示溶解的超咸水和半溶解的盐膜,中心的白色点是未溶解的盐粒。(d)盐颗粒大部分溶解,仅在较大的盐粒中心有少量残余盐粒。由于重力作用砂粒发生沉降,间距减小,颗粒的厚度减薄H2。一些颗粒已断开,如在部分可液化的沉积物中,约70%~85%。颗粒之间的剪切力大幅降低,有效应力失去60%~80%。孔隙流体密度和压力增大。注:高度降低了H2H2H1

  • (a)Plain dry grit without salt. The yellow round ball represents the sand grain, the shear force between grains is larger, the cohesion is stronger; it can withstand a lot of effective stress. (b) Salt-bearing sand and gravel. The white dots are salt particles. Salt crystallization causes the disconnection and "puffing" of sand grains, the distance between grains increases significantly, and the height increases H1. (c) A state of partial dissolution of salt particles. Most particles have been broken off by semi-dissolved salt particles, as in most liquifiable sediments, about 65% to 78%. The red circles around the salt grains represent dissolved supersalt water and semi-dissolved salt film; the white dot in the center is an undissolved grain of salt. (d) most of the salt grains are dissolved, with only a few residual salt grains in the centre of the larger grain. due to the action of gravity, sand particles settle, the spacing decreases, and the thickness of particles decreases H2. some particles have broken off, as in partially liquifiable sediments, about 70% to 85%. the shear force between particles is greatly reduced, and the effective stress is lost about 60%~80%(fig.6). Pore fluid density and pressure increase. Note: the height is reduced by H2, H2H1

  • 3.1.2 盐溶对流化有显著的活化

  • (1)盐溶对流化的活化及其机理。盐溶活化效应的第2个方面是对流化的积极影响。盐溶液化效应与盐水的密度、黏度和表面张力有关,而盐水的流化效应与盐水的密度、黏度有关。实验表明,低浓度到高浓度(9%~23%)盐水(含未溶解的盐粒)具有明显的流化增强效应。由于其黏度和密度较大,可以使最小流化速度(Umf)明显降低(图7a)。

  • 通过计算公式(Nichols et al.,2010)可知,弹性模量与黏度系数的两次幂成反比。在20℃时淡水的黏度系数约为1.00 cp,而20%浓度的盐水的黏度系数约为2.00 cp。根据这两个公式,笔者等得到了不同的计算结果。实验还证实了9%~23%盐水浓度(包括未溶解和半溶解的盐粒)的饱和盐水可使流化速度(Umf)值平均降低12.51%~21.58%(图7a)。因此,在中、高浓度盐水条件下,可以形成比淡水更小的Umf,液化和流化更容易发生,也是高浓度盐水流化活化的重要机制,是由M5.1地震的震荡使哈密SSGL和SGSGL广泛发育砂脉的重要机制之一。同时笔者等通过实验得到两个公式,可以用来计算淡水中密度约为2.5 g/cm3的颗粒在20℃时的Umf:

  • Umf=6.612×D0.6277

  • D为颗粒半径,在6 cm的管中,<0.125~10 mm的颗粒适当;

  • Umf=7.7443×D0.6293

  • D为颗粒半径,12 cm的管中,10~40 mm的颗粒适当(图7b)。笔者等还得到了不同盐水浓度的Umf公式(图7a)。

  • (2)盐溶的流化分异分选。盐溶的流化可以形成分异,不同粒度的颗粒的Umf不同(图7a),与河流的分选相似,河流是水平方向的分选,而流化是垂向上的分选。从哈密野外实际情况看流化分异分选可以分为层分异分选和点分异分选。 ① 层分异分选:在哈密野外常看到在盐屑混合盖中砂质含量较多(图8a),有的甚至发育了分选较好的砂层(图8b)。液化流化在重力作用下发生重力分异,砂级颗粒向上运移,砾石级颗粒向下运移,形成正粒度性,使砂砾石层的分选大大提高(图8),同时又使盐屑混合盖顶部细粒颗粒增加,使盐屑混合盖的密封性大大提高,为下部超压的形成创造了基础; 砂脉发源在下部的砾石层中,在砂层中得到扩展。超高压流体使液化的砾石层中的砾石向上运移,形成罕见的砾脉(图8b)。当SSGL和SGSGL发生完全液化并形成黏性流体时,大部分颗粒很容易通过间隙液体相互分离并漂浮。当液化层(LSGL)受到地震波的冲击以及受盐屑混合盖及基底的冲击和拉分时,颗粒和孔隙液体也随之向下和向上移动,形成湍流流场,不同粒径的颗粒有不同的沉降速度。两种不同粒径球体的质量比为QD1/QD2=(D1/D2)3,两个直径不同的球体的沉降速度比为VD1D2=(D1/D2)3。因此颗粒越大沉降速度V越大,且是颗粒直径比值三次方。因此,图9中的红色大颗粒其沉降速度将远快于图9中的绿球小颗粒,所以很容易形成液化分异或分选。因此,粗粒的砾石聚集在液化层底部,细粒的砂聚集在液化层顶部(图9c、d),为在分选很差的SSGL和SGSGL获得砂源和形成砂脉奠定了物质基础。这就是哈密液化层液化分异的重要机理,也是SSGL和SGSGL形成分选很好的砂脉提供了良好的前提条件。②点分异分选:在层分异分选的基础上在砂脉根部会形成点分异分选(图2e;图10a、b)。砂脉分选良好的特征揭示了在SSGL和SGSGL发生了非常好的流化分异或流化分选(图8d、7e;图8)。在某些砂脉的底部有时会集中一团分选较好的砂,周缘都是砂砾(图2e;图10a、b),表明在砂脉底部有一个较小的分选很好的区域,与面状分异分选相比是一个“点”,所以将其称为“点分异分选”。流化分异使细粒物质向砂脉根部汇聚(图10a、b),与此同时又使砂脉底部提供细粒物质的砂砾层中的细粒物质被搬运走,使砂脉底部的砂砾层形成了一个分选晕(图10c),靠近砂脉底部的流体速度较快,较大的颗粒可以被搬运到砂脉中,向外由于球形扩大而使流体的流速变低,流体的搬运能力下降,被搬运的粒度逐渐变小,由粗砂变为中砂,再由中砂变为细砂,最后变为泥(图10c)。这一现象在野外露头常常可以见到。也正是这种流化分选和流化晕的形成使砂砾层中发育了分选很好的砂脉。在砂脉内也会发生流化分异,有明显的粒度分带,Obermeier(1998)也发现在砂脉内较粗的颗粒粒度有向上变细的现象,可能在一定程度上与液化和流化分异有关。对于粗颗粒,在所有浓度下的受阻沉降率要高得多(Jobe et al.,2012)。

  • 图6 0~10000 min不同粒度的砂层的应力随盐水浓度和粒径的变化

  • Fig.6 The cohesiveness of sand and gravel layers with salinity and particle size varies from 0 to 10000 min

  • (a)2~4 mm砾石在淡水和盐水浓度9%、17%和23%条件下的内聚力。可见颗粒的内聚力随盐水浓度的增加明显减小,从淡水的20 N到盐水浓度23%的2 N,减少了10倍。1~10000 min淡水和盐水的内聚力基本没有变化。 (b)1~2 mm(粗砂)颗粒在淡水和盐水浓度9%、17%和23%条件下的内聚力。可见颗粒的内聚力随盐水浓度的增加明显减小,从淡水的8 N到盐水浓度23%的2 N,减少4倍。1~10000 min淡水和盐水的内聚力基本没有变化。(c)0.5~1 mm(中砂)颗粒在淡水和盐水浓度9%、17%和23%条件下的内聚力。可见颗粒的内聚力随盐水浓度的增加明显减小,从淡水的6 N到盐水浓度23%的2 N,减少了约3倍。 1~10000 min淡水和盐水的内聚力基本没有变化。(d)0.25~0.5 mm(细砂)颗粒在淡水和盐水浓度9%、17%和23%条件下的内聚力。可见颗粒的内聚力随盐水浓度的增加明显减小,从淡水的7 N到盐水浓度23%的2 N,减少了3倍多一点。(e)0.125~0.25 mm(粉砂)颗粒在淡水和盐水浓度9%、17%和23%条件下的内聚力。可见颗粒的内聚力随盐水浓度的增加明显减小,从淡水的7 N到盐水浓度23%的2.5 N,减少了近3倍。(f)≤0.125 mm的颗粒在淡水和盐水浓度9%、17%和23%条件下的内聚力。可见颗粒的内聚力随盐水浓度的增加明显减小,无论淡水还是盐水浓度23%的颗粒在开始时内聚力近于0,但是随着时间的增加内聚力迅速增加,从开始到约70 min后增加到20余牛顿

  • (a)The cohesion of 2~4 mm gravel in fresh water and salinity of 9%, 17% and 23%. The cohesion of the particles decreases with the increase of salinity, from 20 N in fresh water to 2 N in 23% salinity, which decreases by 10 times. From the beginning to 10000 minutes, the cohesion of fresh water and salt water is basically unchanged. (b) Cohesion of 1~2 mm (coarse sand) particles in fresh water and salinity of 9%, 17% and 23%. The cohesiveness of the particles decreases with the increase of salinity, from 8 N in fresh water to 2 N in 23% salinity, which decreases by 4 times. The cohesiveness of fresh and saltwater remained basically unchanged from the beginning to 10000 minutes. (c) Cohesion of 1~2 mm particles in fresh water and salinity of 9%, 17% and 23%. The cohesiveness of the particles decreases with the increase of salinity, from 8 N in fresh water to 2 N in 23% salinity, which decreases by 4 times. The cohesiveness of fresh and saltwater remained basically unchanged from the beginning to 10000 minutes. (d) Cohesion of 0.5~1.0 mm (medium sand) particles in fresh water and salinity of 9%, 17% and 23%. The cohesiveness of the particles decreases obviously with the increase of salinity, from 6 N in fresh water to 2 N in 23% salinity, which decreases by about 3 times. The cohesiveness of fresh and saltwater remained basically unchanged from the beginning to 10000 minutes. (d) Cohesion of 0.25~0.5 mm (fine sand) particles in fresh water and salinity of 9%, 17% and 23%. The cohesiveness of the particles decreases significantly with the increase of salinity, from 7 N in fresh water to 2 N in 23% salinity, which decreases by a little more than three times. (e) Cohesion of 0.125~0.25 mm (silty sand) particles in fresh water and salinity of 9%, 17% and 23%. The cohesiveness of the particles decreases significantly with the increase of salinity, from 7 N in fresh water to 2.5 N in 23% salinity, which decreases nearly 3 times. (f) Cohesion ≤0.125 mm (clay) particles under fresh water and salinity conditions of 9%, 17% and 23%. It can be seen that the cohesion of particles decreases significantly with the increase of salinity, at the beginning, the cohesion of both fresh water and particles with salinity of 23% is close to 0, but with the increase of time, the cohesion increases rapidly, from the beginning to about 70 minutes later to more than 20 N

  • 图7 淡水及不同盐水浓度盐水中变粒径颗粒流化速度 (a) 结果来自直径6 cm的有机玻璃管;(b) 结果来自12 cm的有机玻璃管

  • Fig.7 Fluidization velocity of particles with variable size in fresh water and salt water with different salinities(a) The results come from a 6 cm diameter plexiglass tube; (b) the results come from a 12 cm diameter plexiglass tube

  • 图8 液化—流化分异形成的正粒序性

  • Fig.8 Normal grain grading resulting from liquefaction—fluidization differentiation

  • (a)液化—流化在重力作用下发生重力分异,砂级颗粒向上运移,砾石级颗粒向下运移。(b)液化—流化在重力作用下发生重力分异,砂级颗粒向上运移,砾石级颗粒向下运移,形成正粒度性

  • (a) Gravity differentiation occurs in liquefaction—fluidization under the action of gravity, and sand-grade particles migrate upward and gravel-grade particles migrate downward. (b) Gravity differentiation occurs in liquefaction—fluidization under the action of gravity. Sand grade particles migrate upward and gravel grade particles migrate downward, resulting in normal grain grading

  • 流化分异在岩屑混合盖的底部形成细粒富集带,有利于封闭圈闭流体和超压,所以在没有地震震荡的情况下也能形成超孔隙压力,甚至有可能喷流形成砂脉和砂火山,也是SSGL和SGSGL比正常SGL(砂砾层)更容易液化形成砂脉的根本原因。

  • 3.2 哈密砂脉的的形成模式

  • 众所周知,形成液化砂脉必须要满足3个条件:一是细粒盖层,用来圈闭超压;二是供形成砂脉的砂源;三是有超压,用来把液化层中砂向上输运。但哈密砂脉的成因有自己的特点,非一般砂脉所有。以下简述:

  • (1)隐性的“细粒盖层”—盐屑混合盖。一般认为,如果没有细小的颗粒盖层(泥页岩等)圈闭超孔隙压力流体,就不可能形成超孔隙压力而发育砂脉。野外观测表明,研究区域内不发育这样的细粒盖层(图3a—d,f,h),图4是砂脉形成好以后洪水形成的泥层,不是砂脉发育之前形成的。笔者发现了2种不同寻常的“帽子”,第一种是盐渍砂砾层(SSGL)(图3)和盐粒砂砾层(SGSGL)(图2a—c);第二种是盐壳盖层(SCC,Salt Crust Cover)(图2))。干燥后盐壳盖层非常致密坚硬,与盐岩几无差异,其中有大量明显可见的盐。这种盐壳可以认为是细粒盖层。而第一种SSGL和SGSGL相对较软(有的可以很硬),没有明显可见的大盐块。盐渍砂砾层(SSGL)(图3)和盐粒砂砾层(SGSGL)的暴露表面一般有一层薄薄的盐膜(图3a—d),但当被淡水润湿或饱和时,盐会迅速溶解到盐水中,形成一种超浓的盐水,可对超孔隙压力流体起到良好的密封作用。由于两个含SSGL和SGSGL都属于含盐碎屑层,再加上有时会混有部分盐壳盖层,为简便起见,笔者将其称为盐屑混合盖(SGSGL)。由于强蒸发作用,地表盐渍化比地下更强烈。地表盐水浓度始终高于地下。SGSGL内的盐水浓度始终高于其下方的液化砂砾层(LSSGL,Liquefied Salt Sand—Gravel Layer)。一个非常重要的事实是,固体盐一旦溶解就会变成液体,从而增加了可液化SSGL和SGSGL中的流体。这可能形成一个单独的饱和盐水层(图5b,c)。这现象已被一个简单的溶解实验证明。这不仅导致了SGSGL和LSSGL之间的密度变化,也导致了密度密封的形成。这是SGSGL作为细粒盖层的机理。因此,从这一点来看,研究区基本不缺乏细粒盖层。需要指出的是,SGSGL有时也会成为液化层,为砂脉的形成提供了一些砂源。

  • 图9 不同粒度的颗粒在液化中的沉降分异或分选

  • Fig.9 Settlement differentiation or sorting of particles of different size in liquefaction

  • (a)液化刚开始的状态;颗粒之间紊流发育。(b)在重力作用下,大颗粒开始下沉,发生液化分异。(c)在重力作用下,液化分异形成正粒序性。(d)3种不同粒度的颗粒的流速矢量图,从大到小速度依次减小。(e)重力作用下液化层中3种粒度颗粒的分异位置。(f)重力作用下液化层完成了分异分选,形成了很好的正粒序性,砂级颗粒富集在液化层顶部,砾石富集在液化层下部

  • (a) The state of liquefaction in the beginning; turbulent flow development between particles. (b) Under the action of gravity, large particles begin to sink and liquefaction differentiation occurs. (c) Under the action of gravity, liquefaction differentiation formed normal grain grading. (d) Velocity vector graphs of three particles with different sizes, decreasing from large to small. (e) The different locations of three particle sizes in the liquefaction layer under gravity. (f) Under the action of gravity, the separation and sorting of the liquefied layer are completed, and a good normal grain grading is formed; sand grade particles are enriched in the top of the liquefied layer, and gravel is enriched in the lower part of the liquefied layer

  • (2)SSGL和SGSGL液化和流化分异形成的砂源。有大量的野外证据表明形成哈密砂脉的源砂不是直接来自砂滩和线状砂滩上的某个砂层。既然没有砂层作为母砂源,那么砂脉中的砂从何而来?前文的分析充分表明通过SSGL和SGSGL液化和流化分异可以获得分选较好的砂。SSGL和SGSGL的液化和流化分化可能在砂脉形成和发展的不同时期活跃,液化在先,流化在后,但相辅相成,互相成就。哈密地区的砂脉来自于分选较差的SSGL和SGSGL内,并通过液化和流化分化(或流化分化)在SSGL和SGSGL内先微距运移到裂缝底部(图2f;图10a,b)和岩屑混合盖底部(图10a,b)。这种模式在此定义为砂颗粒可以从分选很差的SSGL和SGSGL中形成等粒度积累(图9;图10d)(Sims et al.,1973),由于在液化过程颗粒密度、颗粒大小以及因为上升流体的流动速度与碎屑混合物(Holzer et al.,1989)(主要是在流化过程中)的差异会出现分选差异,形成分选差异颗粒密度、颗粒大小是主要的,上升流体的流动速度与碎屑混合物是次要的。此外,现场粒度数据表明,这些分选良好的砂质砂脉很可能来自于砂脉下方更深、分选较差的SSGL和SGSGL,不是一个砂源单元。模拟试验证明了这一点:粗砂、中砂和细砂的液化敏感性较好(图6b—e);相反,粉砂的液化敏感性较差(图6f)。这一试验结果可以解释为什么哈密砂脉主要由细至粗砂组成(图3c—i;图4b—d),而泥和粉砂很少。液化和流化分化的过程和机理将在后面详细介绍。

  • (3)超浅的基岩基底。哈密发育砂脉的地区基岩基底普遍很浅,多在0~3 m,主要为晚古的变质岩(图11a)。除了变质岩基底外,另一种基底是风化壳,以砖红色古土壤为主(图11b),还有一些火山岩。刚性的基底可以很好地传播地震波,并且吸收很弱,加上埋藏又浅,为形成低级别大范围地震液化砂脉也创造了条件。

  • 图10 液化SSGL和SGSGL的流化分化形成的分选晕示意图

  • Fig.10 Schematic diagram of fluidization differentiation range of liquefied SSGL and SGSGL

  • (a)大型砂脉及其SSGL和SGSGL;砂脉高近1 m,顶部最宽可达30 cm,底部尖窄,呈缝隙状;裂缝底部有一个透镜状的砂根。(b)图(a)中红框的放大。可以见到裂缝和砂根。砂根宽30 cm左右,厚10 cm左右;其顶部已经挤入到裂缝中。(c)液化SSGL和SGSGL的流化分化形成的分选晕示意图,可以分为四个圈层,不同的圈层碎屑粒度不同,靠砂根处粒度最粗(白色圆点),向外粒度逐渐变细(红色圆点)

  • (a) Large sand dikes with their SSGL and SGSGL; the sand dikes are nearly 1 m high, the widest at the top can reach 30 cm, and the bottom tip is narrow and crevice like; there is a lenticular sand root at the bottom of the crack. (b) Enlargement of the red box in Fig. (a) cracks and sand roots can be seen. The sand root is about 30 cm wide and 10 cm thick; its top has been squeezed into a crack. (c) Schematic diagram of separation halos formed by fluidization differentiation of liquefied SSGL and SGSGL, which can be divided into four circles, with different circles having different particle sizes. The grain size is the coarsest near the sand root (white dots) , and gradually tapering outward (red dots)

  • (4)哈密砂脉的的形成模式。因此,根据现场观测和室内实验,笔者等初步提出哈密液化模型如图12所示。

  • 通过研究笔者等提出了哈密砂脉的形成模式(图12)。这个模式的三构素是:上部的岩屑混合盖,中部的SSGL和SGSGL液化层,底部的基岩基底。在顶部的SSGL和SGSGL(有时可能有盐壳)统称为“盐屑混合盖”(图10a,b;图12a),SSGL和SGSGL顶部由于蒸发泵流作用而含盐更高,由于盐水浓度高而不容易被完全溶解,所以相对下部的SSGL和SGSGL则表现出刚性。发生盐溶后砂砾颗粒会因失去支撑而下沉,在盐屑混合盖下方会形成一个较薄的盐水层,从而形成一个良好的盐水封闭层(图12b—d)。 Warren(2010)也发现密集的残余海底盐水常年淤积,因此保存了一个封闭的地层,为形成高压流体创造了条件。而盐屑混合盖由于盐饱和溶解相对较弱而相对较硬,形成一个相对刚性的盖子。盐溶会使砂砾层的内聚力大大减弱,盐溶实验为解释盐溶液化活化效应提供了支持数据(图6)。从图6可以看出,淡水中颗粒的内聚力是咸水中颗粒的内聚力的3~4倍(图6),并且随着时间的延长衰减较小,从开始到10000 min后咸水中颗粒的内聚力基本变化不大(图6),揭示了一旦盐溶发生后在很长的时间里都具有盐溶活性。在岩屑混合盖或含砂砾岩壳下方的盐水界面处,地下水将变成过饱和盐水,而SGSGL将停止溶解,并形成一个更紧密的密封层。因此,岩屑混合盖或含砂砾岩壳将漂浮在过饱和盐水上,并产生超孔隙压力(图12b—d)。

  • 图11 砂砾岩盖层之下的变质岩基岩

  • Fig.11 Metamorphic bedrock beneath a sandy conglomerate cap

  • (a)变质岩基底与极薄的沉积砂砾盖层,其中发育了中型砂脉。(b)古近系基底与极薄的沉积砂砾盖层,其中发育了大量的砂脉

  • (a) Metamorphic rock basement with a very thin sedimentary sand and gravel cap, in which a medium sand vein is developed. (b) A large number of sand veins developed in the Paleogene basement and the very thin sedimentary sand and gravel cover

  • 当受到地震波的冲击时,SSGL和SGSGL与地下水反向移动,形成一个独立的空间(图12c中的分离层)。此分离层欠压,对顶部岩屑混合盖或含砂砾岩壳有种吸引作用,使其快速向液化的SSGL和SGSGL运动,形成撞击压力,使孔隙流体压力骤增(图12c)。盐溶液不仅降低了液化层内部的内聚力,而且还形成了盐颗粒半溶解的“软化”液化。同时也使液化砂砾层(LSGL)处于超孔隙压力状态,因此它基本上同时处于静态液化状态(图12b)。Allen(1982)也发现静态液化源于无黏性颗粒的静止质量中孔隙流体压力的充分增加。地震波先沿基底到达,使盖层与基底之间出现相位差,在基底与盖层之间出现拉分空隙(图12c)和负压带,使岩屑混合盖猛地撞击SSGL和SGSGL,加大了其液化和孔隙流体压力,使LSSGL中的孔隙流体压力超压;液化和流化使颗粒发生重力分异,较小的颗粒向上运移,反之相反,形成正粒序性(图9c、f;图1d)),较小的颗粒富集在液化流化层的上部,形成一个封闭性相对较好的隔层(图2f、e)在半周期后基底与盖层之间又相对运动,在液化层中形成相对运动而产生高压带(图12d)。地震使基底、液化SSGL和SGSGL和盐屑混合盖(“三明治”结构)像手风琴一样产生背离和相向运动,基底和盐屑混合盖冲击和拉分使SSGL和SGSGL出现频繁的高压和负压(图12c、d),加速了SSGL和SGSGL的液化,使液化流体和上部的液化分异和分选形成的细粒物质喷出地表(图10e),形成砂脉和砂火山。

  • 综上所述,盐溶活化作用在砂脉形成中的重要性有3个方面或3种机理:一是盐溶,尤其是半溶的盐粒使SSGL和SGSGL的颗粒之间的内应力或剪切力大大降低(图5c、d),使其在未受地震振荡时提前进入液化状态,可大大降低地震振荡的液化能量或液化阈值;其次,盐溶导致流体的密度加大,使SSGL和SGSGL中的大多数颗粒被盐溶液悬浮,降低甚至失去有效应力而形成超孔隙压力,促进液化;第三,由于砂砾层在地震之前就发生了液化,可把砂(砾石)缩短进入流化层的形成时间,等于大大延长了地震振荡的时间,从而延长砂脉的形成时间。需要强调的是在这种不利条件下,如果没有盐溶的液化活化是绝对不可能形成砂脉的。

  • 图12 盐溶活化与地震液化过程的“三明治”概念模型与砂脉的形成模式

  • Fig.12 Sandwich conceptual model of salt dissolution and seismic liquefaction process and sand dike formation

  • (a)[岩屑混合盖或含砂砾岩壳]—[SGSGL—SSGL]—[基底]的“三明治”结构。SGSGL—SSGL中的砾石具有叠瓦结构。当SSGL和SGSGL被干燥时,碎屑颗粒被盐粒胶结和隔离(图2b、c)。(b)当SSGL和SGSGL被淡水饱和时,盐膜或盐颗粒逐渐溶解,颗粒断开,盐溶解产生盐溶液,在SSGL和SGSGL下方形成无沉淀盐水层,颗粒之间没有接触,就像地震下完全可液化的砂子在摇摆晃动。最后,岩屑混合盖或含砂砾岩壳将漂浮在含盐的地下水上。注意:颗粒不能相互接触,如完全液化的砂子。(c)SSGL和SGSGL或基底向相反方向移动,形成高压。有利于砂的再活化,导致液化和液化分化;岩屑混合盖和基底的相反运动使LSSGL中的孔隙流体压力欠压。(d)在受到强烈震荡的SSGL和SGSGL中发生重力沉降分异,较大的砾石颗粒向下坠落,并发生紧密堆积,并形成更紧密堆积的粗粒(注意: 卵石的长轴重新排列与水平面的平行)。较小的砂级颗粒和黏土向上浮起,形成一个正粒序性层,在LSSGL和SGSGL顶部形成一个更紧密的膜,并伴随着饱和盐水(“自孕”密封层。图7d为红色薄层)。岩屑混合盖和基底的相向运动使LSSGL中的孔隙流体压力超压。(e)LSSGL超孔隙压力流体与上部砂喷发,导致砂脉的形成,与此同时泄压。LSSGL出现紧密排列密度增高

  • (a) “Sandwich” structure of [debris mixed cap or sandy conglomerate crust]—[SGSGL—SSGL]—[basement]. The gravel in SGSGL—SSGL has imbricate structure. When SSGL and SGSGL are dried, the clastic particles are cemented and isolated by salt grains(Fig.2b, c). (b) When SSGLand SGSGL is saturated with fresh water, the salt film or salt particles gradually dissolve, the particles break off, the salt solution is dissolved, resulting in the formation of a precipitation-free brine layer below the SGSGL, with no contact between the particles, just like a fully liquefied sand swaying under an earthquake. Finally, the SGSGL will float on saline groundwater. Below the SGSGL, at the SGSGL—water interface, the groundwater will turn into suoversaturated brine, and the SGSGL will stop dissolving and form a tighter seal. As a result, the SGSGL will float on susaturated salt water and create pressure. When hit by seismic waves, the SGSGL and groundwater move in reverse, forming a separate chamber. Note: particles should not be in contact with each other, such as fully liquefied sand. (c) The SGSGL and/or basement move in opposite directions, resulting in high pressure. This facilitates sand reactivation, leading to liquefaction and liquefaction differentiation, and the formation of more closely packed coarse-grained grains (note: the long axial load of pebbles is newly arranged parallel to the horizontal plane) . The fine particles will be squeezed into the fine pores of the SGSGL to form a more compact film, accompanied by a saturated brine ( “self-fertilizing ” seal; Fig.5d shows the brown red thin layer) . Note: particles should not be in contact with each other, such as fully liquefied sand. (d) Overpressured fluid and sand ejection form dikes. In strongly disturbed SGSGLS, expansion fractures, fluid and sand dikes, and fluid differentiation occur. Today, the extruded salt-rich mortar remains on the surface in the form of a mixture of salt-cemented sand and silt. The increase of pore fluid pressure leads to the corresponding reduction of SSGL and SGSGL strength, and the pore fluid pressure is equal to the overlying rock pressure and seismic fluctuation and shaking, forming an overpressure system. The larger particles in the LSSGL are densely packed. (e) Eruption of LSSGL overpressurized pore water, resulting in the formation of sand dikes. The density of LSSGL is very high

  • (5)砂脉形成的异常动力学过程。砂脉系统是非常复杂的动态系统,它真实地展示了固体和流体相互作用的物理原理。Lowe(1975)认为液化过程中存在5种应力。阐明哈密砂脉形成的动力学过程不仅有助于更好的理解哈密砂脉的形成,而且有助于理解古代类似砂脉的形成过程(Holzer et al.,1989)。如图1研究区为一个只有4 km深的非常浅的M5.1地震。哈密地区地震液化主要包括饱和、相对松散、粗粒沉积层的静抗剪强度的损失,主要是由于盐溶作用的显著活化作用使组成颗粒发生分解堆积。这不仅是地震横波(Holzer et al.,1989)的动态驱动,还包括所有其他地震地面震动,以及SGSGL和分离层的联合作用(图10)。完全由地震引起的液化属于循环液化。这种循环液化是由一个反向荷载的重复作用造成的,就像在地震期间循环液化在很大程度上与经典理论不相容(Allen,1982)。地波特别是瑞利波具有典型的低速、低频、高振幅特征,容易引起地面的隆起和翻滚。通常,地震地波占地震波动和晃动总能量的67%,它们的衰减也非常缓慢。瑞利波是一种沿自由地波传播的波,SGSGL与空中界面可视为自由界面。此外,SSGL与SGSGL底部和基底顶部的两个接触界面也是自由表面(图10),强振幅有利于盐屑混合盖和基底对SSLG和SGSGL的影响(图10c,d)。

  • 图10c和图10d是盐屑混合盖与基底相位相差180°的状态(地震期间的部分状态),这两个过程更有利于SSGL和SGSGL的液化。由于它们有不同的密度、深度和厚度,它们将有不同的频率和初始相位,即使在相同的地震波下。实例观察表明,上覆岩层的作用力大小对液化的影响很大。在黄河三角洲边滩的实际观察发现,当推土机通过一个松散的水饱和细粒层时,立即液化,并有许多小型淤泥火山迅速发生。此外,经验表明在松散的、被水饱和细粒层践踏时会迅速发生液化,形成液化并发育小型砂泥火山。

  • 这些现象表明,盐溶活化对SSGL和SGSGL液化的影响非常重要。较小的颗粒(如黏土和砂子)和水聚集在不可液化的SGSGL和下层SSGL之间的界面上。较粗的颗粒(如细卵石和粗—中砂)出现在基底与下垫面SSGL的界面上,这是地震液化过程中及之后颗粒再沉积或重新排列的结果。这导致了高孔隙压力和重新活化砂,形成液化分异(图10e)。盐溶液化的SSGL与SGSGL对盐屑混合盖和刚性基底的震动冲击非常敏感。因此,笔者将这两种复合动力学(盐溶液化和地震液化)称为“异常动力学”。

  • 此外,从理论上讲,主震过程中的S波和瑞利波破坏SGSGL或SSGL,导致震中周围形成一组圆形的冲击诱发裂缝(p波也对这组裂缝有贡献),主震过程中的Lev波使SGSGL在震中周围以一组径向激振裂隙的形式破裂。因此,Rayleigh波和Lev波的组合冲击使SGSGL产生方向随机的网络形式的裂缝。由于各种地质因素,这些多边形网络不再是简单的四边形,而是复杂的多边形形状。Rayleigh波、Lev波和横波冲击形成的合力使SSGL和SGSGL发生裂纹。这个力有多大?这不仅是一个有趣的问题,而且是一个重要的问题。通过对从裂缝中输送到地面的最大卵石(平均直径为4 cm)的弹性模量(Umf)计算,得出地震冲击压力仅为513.448~637.29 kg/m2左右(另文专述)。这说明裂隙的形成主要依赖于横波的弯曲效应。此外,地震震动似乎使叠瓦构造向流动方向重新排列或增加了叠瓦构造的倾角。

  • 此外,哈密地区的液化和砂脉的形成也得益于“盆地效应”(Shani-Kadmiel et al.,2012)和“聚焦效应”(Shani-Kadmiel et al.,2014)的放大效应,这两种效应都发生在非常接近地表凹陷的地方。此外,Denolle(2016)指出,地壳结构较浅的沉积盆地可以通过捕获和放大地震波,将地震放大幅度提高到原来的3倍。哈密地区不同寻常的砂脉是由多种不同寻常的过程形成的,其中有些过程是相互关联(结晶与盐溶解)。正是由于上述不同寻常的因素,才产生了不同寻常的结果:远程的液化效应和广泛分布的砂脉。

  • 3.3 远程液化和最大震中距的确定

  • 哈密砂脉一个主要的关注点是远程液化。液化地点与最大震中距离与力矩震级(震级限制法)之间的关系被频繁地使用,以至于人们对其有效的液化阈值M5似乎没有什么怀疑(Sims et al.,1973Kuribayashi et al.,1975Tinsley et al.,1985Obermeier,1996Galli,2000Jolly et al.,2002Obermeier et al.,2005;Papathanassiou et al.,2005;Fan et al.,2016)。地震可以使孔隙流体产生超孔隙压力而且松软的沉积物发生液化(Ishihara et al.1985),但是达不到阈值(≤M5,0.2 g)也是枉然的。这意味着,即使可液化地点距震中距离很小,小于5.0级的地震也不足以引起液化。然而,Galli(2000)指出,意大利历史液化数据库也解释了大量产生液化的中等地震(Ms≥4.2)(Hanks et al.,1976;Papathanassiou et al.,2005;Galli et al.,19992000Castilla et al.,2007)。此外,Rodríguez-Pascua(2000)认为地震液化阈值在M4.5左右。所以,从一些学者的观点看,哈密M5.1地震形成液化也不是没有可能的,但形成如此大范围(图1b)的液化确难以置信。

  • 从长期来讲一个地区的地震是不可避免的,对于构造活跃地区经常在同一个空间反复发生地震也是常见的。因此,在一个地震频发的地区很难准确地确定砂脉是由哪一次地震形成的。笔者收集了研究区及邻区近280年的地震资料,并在图1中标记了震级与震中距离的分布,其量级为M5.0~M7.5(图1a,b)。这是根据Obermeier图解法(1998)的震级限定方法,结合图1a和图1b中哈密境内及邻区砂脉分布格局及其影响范围得出的。图1a也绘制了哈密>M5.75地震各区域连续液化范围。从图1a和图1b中可以看出,从哈密到雅满苏镇,M6.0级以上的地震(以及M5.75)都发生在研究区域的左下角。本研究区有可能受到6级以上地震的影响,本地区的液化肯定是由6级以下地震形成的。在主研究区附近,在过去280年里至少发生了4次M5地震。笔者还绘制了4次略大于M5地震的液化变形包络线(脉岸)范围(图1b)。笔者认为这极不可能由另一个更低级别或远程触发地震产生的影响,所以笔者可以很容易地得出这样的结论:这些砂脉区域内从哈密到雅满苏是由2009年12月14日M5.1雅满苏地震产生(图1b黄色标记)。其他证据还有:

  • 首先,在2009年的震中附近,直径约80 km的范围内,砂脉不太可能受到公元1737年到公元2009年的任何其他地震的影响(图1b)。在南湖发电厂附近发现的砂脉是在2005~2006年公路建成后形成的,此后只有2009年12月14日M5.1地震级别最高,形成液化砂脉的可能性最大。

  • 其次,从砂脉的几何形态、保存新鲜度(主要是未被盐渍、掩埋或大量截断)、发生特征和分布连续性(图1),结合历史震中的区域分布格局(图1),都表明这些砂脉不可能是由其他地震形成的,包括2014年的地震。有充足的理由认为,2009年M5.1地震液化至少可达距震中80 km以远的哈密南湖电厂。综上所述,超浅层地震、超浅层基底以及分布广泛的SGSGL和SSGL是2009年M5.1地震远程液化的有利因素。相反,虽然离震中很近,但由于在震中附近现代沉积发育不良,砂脉发育不佳。

  • 再者,哈密最西端的80~120 km的砂脉在2009 年M5.1和公元1914年之间的过渡M7.5 震中地区,最新的砂脉不含盐且特别松散,可能是由2009年M5.1地震形成的。而那些高盐砂脉可能是由其他历史上的地震形成的,比如1914年的M7.5大地震。最西面的砂脉从黄色实线80 km到120 km黄色虚线(图1b),也陷入1914年M7.5地震的有效范围,是最强的“county-recorded”地震事件,刚刚超过100年哈密的研究区域。然而,在距离震中120 km以西的地方,没有发现砂脉(图1c)。据此推测,这些80~120 km范围内的砂脉也可能是2009年M5.1地震形成的。

  • 4 结论

  • (1)研究区发育了大量的砂脉,连续分布、中小型规模、平面上多边形,剖面上楔形;大部分倾斜,被认为是形成于2009年12月14日M5.1的中等规模的地震。

  • (2)由于有基底浅、盖层薄和强烈盐化,使哈密在这样一个中等强度的、刚进入液化阈值的地震(M5.1)的非常有限的震荡冲击下,形成了从震中向外延伸80~120 km的广泛分布的地震液化砂脉。

  • (3)实验表明,在浓度为9%、17%和23%的盐水(含大量半溶盐粒)条件下,盐溶作用可大幅度降低颗粒之间的剪应力,使有效应力大幅降低,随盐水浓度的增加颗粒之间的剪应力分别下降25%~75%左右,有利于地震液化和流化形成砂脉。但细颗粒、尤其是黏土基本没有盐溶效应,可能与黏土的吸附有关。

  • (4)砂脉分选良好的特征肯定是由于盐渍砂砾层(SSGL,Salinization Sand—Gravel Layer)和盐粒砂砾层(SGSGL,Salt-Grain—Sand—Gravel Layer)的液化和流化分选,而不是在砂脉之下有一个砂源。通过流化实验和计算,得到了液化砂砾层的流化速度分布和流化边界图。

  • (5)笔者等通过对<0.125~40 mm颗粒的最小流化速度Umf进行了精细的实验研究,提出了两个经验公式:Umf=6.612×D0.6277Umf=7.7443×D0.6293

  • (6)在M5.1地震的作用下形成广泛分布的哈密砂脉主要有5个特征:① 砂砾层中的盐溶解后会促使液化流化。平均降低58.3%~70.0%,使地震液化阈值降低到0.15~0.05 g(以0.2 g为一般阈值)。与此同时,浓盐水由于密度加大,还可使淡水最小流化速度(Umf)降低12.51%~21.58%,而有利于流化;② 广泛分布的SSGL和SGSGL;③ 震源极浅(4 km深);④ 基底极浅(深度约0~3 m);⑤ 盐屑混合盖+SSGL和SGSGL+极浅基岩基底的特殊的“三明治”结构。

  • 致谢:本文在实地考察、室内实验中受益于许多合作者和研究生,包括刘选、刘圣鑫、张丹峰、邵珠福、王韶洁、刘晓光、白静、陈红、王桂林、王永强、冯国强、薛艳秋、谷东辉、彭超峰、李丰田、王超宁、张良等等,同时,感谢哈密地震数据中心提供的所有地震目录。

  • 参考文献

    • 杜远生, Shi G, 龚一鸣, 许亚军. 2007. 东澳大利亚南悉尼盆地二叠系与地震沉积有关的软沉积物变形构造. 地质学报, 81(4): 511~518.

    • 杜远生. 2011. 中国地震事件沉积研究的若干问题探讨. 古地理学报, 13(6): 581~590.

    • 杜远生, 余文超. 2017. 地震和非地震引发的软沉积物变形构造. 古地理学报, 19(1): 65~72.

    • 乔秀夫, 宋天锐, 高林志, 彭阳, 李海兵, 高劢, 宋彪, 张巧大. 1994. 碳酸盐岩振动液化地震序列. 地质学报, 68(1): 16~36.

    • 乔秀夫, 高林志, 彭阳, 李海兵. 2001. 古郯庐带沧浪铺阶地震事件、层序及构造意义. 中国科学(D辑), 31(11): 911~918.

    • 苏德辰, 孙爱萍, 郭宪璞, 王思恩. 2013. 青藏高原东北缘祁连山地区古地震记录与区域构造意义. 岩石学报, 29(6): 2223~2232.

    • 苏德辰, 乔秀夫. 2018. 黏性沉积物中的古地震触变流动变形. 古地理学报, 20(4) : 609~622.

    • 钟建华, 曹孟春, 倪良田, 孙宁亮, 刘闯, 郝兵, 杨冠群, 宋冠先, 葛毓柱. 2018. 砂脉的研究现状与进展. 古地理学报, 20(1): 119~132.

    • 钟建华, 倪良田, 汪立东, 宋全友, 孙宁亮, 宋冠先薛纯琦, 郝兵, 邵珠福, 毛毳, 刘闯, 曹孟春. 2020a. 青岛灵山岛南背来石下白垩统大型砂脉的发现及其形成动力学研究. 地质论评, 66(6): 1457~1476.

    • 钟建华, 王永卓, 倪良田, 高春文, 周永柄, 毛毳, 樊晓东, 刘国志, 陈志鹏, 宋冠先, 王瑞, 黄亮, 刘强, 范玉娇, 任凯旋. 2020b. 砂脉(砂注入体)型储层与油藏——以松辽盆地长垣地区西部萨葡夹层为例. 地球科学与环境学报, 42(3): 285~296.

    • 钟建华, 孙宁亮, 郝兵, 倪良田, 宋冠先, 薛纯琦, 毛毳, 邵珠福, 葛毓柱, 陈彬, 刘圣鑫, 彭超锋, 谷东辉, 王永强, 李聪. 2020c. 山东灵山岛北背来石剖面下白垩统顺层砂脉的特征及其成因动力学分析. 古地理学报, 22(4): 635~646.

    • Allen J R L. 1982. simple models for the shape and syMetry of tidal sand waves: statically stable equilibrium forms. Marine Geology, 48: 31~49.

    • Allen J R L. 1994. Sedimentary Structure, their character and physical basis, Ⅱ. New York Elsevier Scientific Publishing Company Amsterdam, Oxford.

    • Castilla R A, Audemard F A. 2007. Sand blows as a potential tool for magnitude estimation of pre-instrumental earthquakes. Journal of Seismology, 11(4): 473~487.

    • Monnier D, Gay A, Imbert P, et al. 2015. Sand injectites network as a marker of the palaeo-stress field, the structural framework and the distance to the sand source: Example in the Vocontian Basin, SE France. Journal of Structural Geology, 79: 1~18.

    • Denolle M A, Shearer P M. 2016. New perspectives on self-similarity for shallow thrust earthquakes. Journal of Geophysical Research, 121 (9): 6533~6565.

    • Du Yongsheng, Shi G, Gong Yiming, Xu Yajun. 2007&. Permian soft-sediment deformation structures related to earhquake in the southern Sydney Basin, eastern Australia. Acta Geologica Sinica, 81(4): 511~518.

    • Du Yongsheng. 2011&. Discussion about studies of earthquake event deposit in China. Journal of Palaeogeography, 13(6): 581~590.

    • Du Yuansheng, Yu Wenchao. 2017&. Earthquakecaused and non-earthquake-caused soft-sediment deformations. Journal of Palaeogeography, 19(1) : 65~72.

    • Fan W, Shearer P M. 2016. Local near instantaneously dynamically triggered aftershocks of large earthquakes. Science, 353(6304): 1133~1136.

    • Galli P, Meloni F, Rossi A. 1999. Historical liquefaction in Italy: relationship between epicentral distance and seismic parameters// European Geophysical Society XXIIII General Assembly Natural Hazards NH3, The Hague, Netherlands, 19~23

    • Galli P. 2000. New empirical relationships between magnitude and distance for liquefaction. Tectonophysics, 324(3): 169~187.

    • Group E W. 2013. Liquefaction phenomena associated with the Emilia earthquake sequence of May–June 2012 (Northern Italy). Natural Hazards & Earth System Sciences, 13: 935~947.

    • Hanks T C, Simsson D A. 1976. Geophysical assessment of peak accelerations. Bulletin of the Seismological Society of America, 66(3): 959~968.

    • Heron D, Etienne J L. 2005. A complex subglacial clastic dyke swarm, Sólheimajkull, southern Iceland. Sedimentary Geology, 181(1/2): 25~37.

    • Holzer T L, Hanks T C, Youd T L. 1989. Dynamics of Liquefaction During the 1987 Superstition Hills, California, Earthquake. Science, 244: 56~59.

    • Ishihara K. 1985. in Proceedings of the Eleventh Interational Conference on Soil Mechanics and Foundation Engineering, San Francisco, 12 to 16 August 1985 (Balkema, Rotterdam, 1985), 321~376.

    • Jobe Z R, Lowe D R, Morris W R. 2012. Climbing-ripple successions in turbidite systems: depositional environments, sedimentation rates and accumulation times. Sedimentology, 59(3): 867~898.

    • Sims J D. 1973. Earthquake-Induced Structures in Sediments of Van Norman Lake, San Fernando, California. Science, 182, 161~163.

    • Jolly R, Lonergan L. 2002. Mechanisms and controls on the formation of sand intrusions. Journal of the Geological Society, 159(5): 605~617.

    • Kuribayashi, E. T. Iwasaki, F. Tatsuoka. 1975. Effects of stress—strain conditions on dynamic properties of sands. Soils & Foundations. 242: 105~114.

    • Lowe D R. 1975. Water escape structures in coarse-grainedd sediments. Sedimentology, 22: 157~204.

    • Moretti M, Alfaro P, Caselles O, et al. 1999. Modelling seismites with a digital shaking table. Tectonophysics, 304(4): 369~383.

    • Nichols R J, Sparks R S J, Wilson C J N. 2010. Experimental studies of the fluidization of layered sediments and the formation of fluid escape structures. Sedimentology, 41(2): 233~253.

    • Obermeier S F, Bleuer et al. 1991. Evidence of strong earthquake shaking in the lower Wabash Valley from prehistoric liquefaction. Science, 251: 1061.

    • Obermeier S F. 1996. Use of liquefaction-induced features for paleoseismic analysis; an overview of how seismic liquefaction features can be distinguished from other features and how their regional distribution and properties of source sediment can be used to infer the locatio. Engineering Geology, 44(1): 1~76.

    • Obermeier S F. 1998. Liquefaction evidence for strong earthquakes of Holocene and latest Pleistocene ages in the states of Indiana and Illinois, USA. Engineering Geology, 50(3~4): 227~254.

    • Obermeier S F, Olson S M, Green R A. 2005. Field occurrences of liquefaction-induced features: a primer for engineering geologic analysis of paleoseismic shaking. Engineering Geology, 76(3~4): 209~234.

    • Owen G, Moretti M, Alfaro P. 2011. Recognising triggers for soft-sediment deformation: Current understanding and future directions. Sedimentary Geology, 235(3~4): 133~140.

    • Papathanassiou G, Pavlides S, Ganas A. 2006. The 2003 Lefkada earthquake: Field observations and preliminary microzonation map based on liquefaction potential index for the town of Lefkada. Engineering Geology, 82(1): 12~31

    • Qiao Xiufu, Song Tianrui, Gao Linzhi, Peng Yang, Li Haibing, Gao Mai, Song Biao, Zhang Qiaoda. 1994&. Seismi sequence in carbonate rocks by vibrational liquefaction. Acta Geologica Sinica, 68(1) : 16~36.

    • Qiao Xiufu, Gao Linzhi, Peng Yang, Li Haibing. 2001&. Seismic event, sequence and tectonic significance in Canglangpu Stage in Paleo-Tanlu Fault Zone. Science in China(Series D), 31(11): 911~918.

    • Quigley M C S. Bastin, B A Bradley. 2013. Recurrent liquefaction in Christchurch, New Zealand, during the Canterbury earthquake sequence. Geology, 41: 419~422.

    • Rodríguez-Pascua, M A Calvo J P, Vicente G D, et al. 2000. Seismites in lacustrine sediments of the Prebetic Zone, SE Spain, and their use as indicators of earthquake magnitudes during the Late Miocene. Sedimentary Geology, 135(1~4): 117~135.

    • Seed H B, Tokimatsu L F, Harder L F, Chung R M J. 1985. Influence of spt procedures in soil liquefaction resistance evaluations. Geotech. Eng, 111: 1425.

    • Shani-Kadmiel S M, Tsesarsky J N, Louie Z, Gvirtzman. 2012. Bulletin of the Seismological Society of America, 102(4): 1729~1739.

    • Shani-Kadmiel S M, Tsesarsky J N, Louie Z, Gvirtzman. 2014. Bulletin of earthquake engineering. 12: 607~625.

    • Simons M. S. E. Minson A. Sladen et al. 2011. The 2011 Magnitude 9. 0 Tohoku—Oki Earthquake: Mosaicking the Megathrust from Seconds to Centuries. Science, 332: 1421~1425.

    • Su Dechen, Sun Aiping, Guo Xianpu, Wang Sien. 2013&. Seismites in the Qilianshan area of the northeastern margin of Qinghai—Tibetan Plateau and a pilot research on the regional tectonics. Acta Petrologica Sinica, 29(6): 2223~2232.

    • Su Dechen, Qiao Xiufu. 2018&. Palaeoseismic thixotropic flow deformation in viscous sediments. Journal of Palaeogeography, 20(4) : 609~622.

    • Sukhija B S, Rao M N, Reddy D V, Nagabhushanam P, Lakshmi BV, Gupta H K. 2000. Current Status of Paleoseismology——An overview. Journal of Geological Society of India, 55: 465~480.

    • Tinsley J C, Youd T L, Perkins D M, Chen A T F. 1985. An Earth-Science Perspective. US Geological Survey Denver, CO US Geological Survey Professional Paper. 1360: 263~316.

    • Warren J K. 2010. Evaporites through time: Tectonic, climatic and eustatic controls in marine and nonmarine deposits. Earth-Science Reviews, 98(3): 217~268.

    • William L E. 2013. Injection-Induced Earthquakes. Science, 341(12): 1~7.

    • Zhong Jianhua, Cao Mengchun, Ni liangtian, Sun Ningliang, Liu Chunag, Hao Bing, Yang Guangqun, Song Guangxian, Ge Yuzhu. 2018&. Situation of study and development tendency of sandy dykes. Journal of Palaeogeography, 20(1): 119~132.

    • Zhong Jianhua, Ni Liangtian, Wang Lidong, Song Quanyou, Sun Ningliang, Song Guangxian, Xue Chunqi, Hao Bing, Shao Zhufu, Mao Cui, Liu Chuang, Cao Menchun. 2020a&. Discovery of large sand dikes in the Lower Cetaceous of southern Beilaishi, Lingshan Island, Qingdao, and their formation dynamics. Geological Review, 66(6): 1457~1476.

    • Zhong Jianhua, Wang Yongzhuo, Ni Liangtian, Gao Chuanwen, Zhou Yongbin, Mao Cui, Fan Xiaodong, Liu Guozhi, Chen Zhipeng, Song Guangxian, Wang Rui, Huang Liang, Liu Quang, Fan Yujiao, Ren Kaixuan. 2020b&. Sand-dike (Sand injection) type of Reservoir and Oil Pool. Journal of Earth Science and Environment, 42(3): 285~296.

    • Zhong Jianhua, Sun Ningliang, Hao Bing, Ni Liangtian, Song Guanxian, Xuan Chunqi, Mao Cui, Shao Zhufu, Ge Yuzhu, Chen Bin, Liu Shengxin, Peng Shaofeng, Fu Donghui, Wang Yongqiang, Li Cong. 2020c&. Characteristics and genetic dynamics analysis of Lower Cretaceous bedding sand veins in the Laishi Section of Lingshan Island, Shandong Province. Journal of Palaeogeography, 22(4): 635~646.

  • 参考文献

    • 杜远生, Shi G, 龚一鸣, 许亚军. 2007. 东澳大利亚南悉尼盆地二叠系与地震沉积有关的软沉积物变形构造. 地质学报, 81(4): 511~518.

    • 杜远生. 2011. 中国地震事件沉积研究的若干问题探讨. 古地理学报, 13(6): 581~590.

    • 杜远生, 余文超. 2017. 地震和非地震引发的软沉积物变形构造. 古地理学报, 19(1): 65~72.

    • 乔秀夫, 宋天锐, 高林志, 彭阳, 李海兵, 高劢, 宋彪, 张巧大. 1994. 碳酸盐岩振动液化地震序列. 地质学报, 68(1): 16~36.

    • 乔秀夫, 高林志, 彭阳, 李海兵. 2001. 古郯庐带沧浪铺阶地震事件、层序及构造意义. 中国科学(D辑), 31(11): 911~918.

    • 苏德辰, 孙爱萍, 郭宪璞, 王思恩. 2013. 青藏高原东北缘祁连山地区古地震记录与区域构造意义. 岩石学报, 29(6): 2223~2232.

    • 苏德辰, 乔秀夫. 2018. 黏性沉积物中的古地震触变流动变形. 古地理学报, 20(4) : 609~622.

    • 钟建华, 曹孟春, 倪良田, 孙宁亮, 刘闯, 郝兵, 杨冠群, 宋冠先, 葛毓柱. 2018. 砂脉的研究现状与进展. 古地理学报, 20(1): 119~132.

    • 钟建华, 倪良田, 汪立东, 宋全友, 孙宁亮, 宋冠先薛纯琦, 郝兵, 邵珠福, 毛毳, 刘闯, 曹孟春. 2020a. 青岛灵山岛南背来石下白垩统大型砂脉的发现及其形成动力学研究. 地质论评, 66(6): 1457~1476.

    • 钟建华, 王永卓, 倪良田, 高春文, 周永柄, 毛毳, 樊晓东, 刘国志, 陈志鹏, 宋冠先, 王瑞, 黄亮, 刘强, 范玉娇, 任凯旋. 2020b. 砂脉(砂注入体)型储层与油藏——以松辽盆地长垣地区西部萨葡夹层为例. 地球科学与环境学报, 42(3): 285~296.

    • 钟建华, 孙宁亮, 郝兵, 倪良田, 宋冠先, 薛纯琦, 毛毳, 邵珠福, 葛毓柱, 陈彬, 刘圣鑫, 彭超锋, 谷东辉, 王永强, 李聪. 2020c. 山东灵山岛北背来石剖面下白垩统顺层砂脉的特征及其成因动力学分析. 古地理学报, 22(4): 635~646.

    • Allen J R L. 1982. simple models for the shape and syMetry of tidal sand waves: statically stable equilibrium forms. Marine Geology, 48: 31~49.

    • Allen J R L. 1994. Sedimentary Structure, their character and physical basis, Ⅱ. New York Elsevier Scientific Publishing Company Amsterdam, Oxford.

    • Castilla R A, Audemard F A. 2007. Sand blows as a potential tool for magnitude estimation of pre-instrumental earthquakes. Journal of Seismology, 11(4): 473~487.

    • Monnier D, Gay A, Imbert P, et al. 2015. Sand injectites network as a marker of the palaeo-stress field, the structural framework and the distance to the sand source: Example in the Vocontian Basin, SE France. Journal of Structural Geology, 79: 1~18.

    • Denolle M A, Shearer P M. 2016. New perspectives on self-similarity for shallow thrust earthquakes. Journal of Geophysical Research, 121 (9): 6533~6565.

    • Du Yongsheng, Shi G, Gong Yiming, Xu Yajun. 2007&. Permian soft-sediment deformation structures related to earhquake in the southern Sydney Basin, eastern Australia. Acta Geologica Sinica, 81(4): 511~518.

    • Du Yongsheng. 2011&. Discussion about studies of earthquake event deposit in China. Journal of Palaeogeography, 13(6): 581~590.

    • Du Yuansheng, Yu Wenchao. 2017&. Earthquakecaused and non-earthquake-caused soft-sediment deformations. Journal of Palaeogeography, 19(1) : 65~72.

    • Fan W, Shearer P M. 2016. Local near instantaneously dynamically triggered aftershocks of large earthquakes. Science, 353(6304): 1133~1136.

    • Galli P, Meloni F, Rossi A. 1999. Historical liquefaction in Italy: relationship between epicentral distance and seismic parameters// European Geophysical Society XXIIII General Assembly Natural Hazards NH3, The Hague, Netherlands, 19~23

    • Galli P. 2000. New empirical relationships between magnitude and distance for liquefaction. Tectonophysics, 324(3): 169~187.

    • Group E W. 2013. Liquefaction phenomena associated with the Emilia earthquake sequence of May–June 2012 (Northern Italy). Natural Hazards & Earth System Sciences, 13: 935~947.

    • Hanks T C, Simsson D A. 1976. Geophysical assessment of peak accelerations. Bulletin of the Seismological Society of America, 66(3): 959~968.

    • Heron D, Etienne J L. 2005. A complex subglacial clastic dyke swarm, Sólheimajkull, southern Iceland. Sedimentary Geology, 181(1/2): 25~37.

    • Holzer T L, Hanks T C, Youd T L. 1989. Dynamics of Liquefaction During the 1987 Superstition Hills, California, Earthquake. Science, 244: 56~59.

    • Ishihara K. 1985. in Proceedings of the Eleventh Interational Conference on Soil Mechanics and Foundation Engineering, San Francisco, 12 to 16 August 1985 (Balkema, Rotterdam, 1985), 321~376.

    • Jobe Z R, Lowe D R, Morris W R. 2012. Climbing-ripple successions in turbidite systems: depositional environments, sedimentation rates and accumulation times. Sedimentology, 59(3): 867~898.

    • Sims J D. 1973. Earthquake-Induced Structures in Sediments of Van Norman Lake, San Fernando, California. Science, 182, 161~163.

    • Jolly R, Lonergan L. 2002. Mechanisms and controls on the formation of sand intrusions. Journal of the Geological Society, 159(5): 605~617.

    • Kuribayashi, E. T. Iwasaki, F. Tatsuoka. 1975. Effects of stress—strain conditions on dynamic properties of sands. Soils & Foundations. 242: 105~114.

    • Lowe D R. 1975. Water escape structures in coarse-grainedd sediments. Sedimentology, 22: 157~204.

    • Moretti M, Alfaro P, Caselles O, et al. 1999. Modelling seismites with a digital shaking table. Tectonophysics, 304(4): 369~383.

    • Nichols R J, Sparks R S J, Wilson C J N. 2010. Experimental studies of the fluidization of layered sediments and the formation of fluid escape structures. Sedimentology, 41(2): 233~253.

    • Obermeier S F, Bleuer et al. 1991. Evidence of strong earthquake shaking in the lower Wabash Valley from prehistoric liquefaction. Science, 251: 1061.

    • Obermeier S F. 1996. Use of liquefaction-induced features for paleoseismic analysis; an overview of how seismic liquefaction features can be distinguished from other features and how their regional distribution and properties of source sediment can be used to infer the locatio. Engineering Geology, 44(1): 1~76.

    • Obermeier S F. 1998. Liquefaction evidence for strong earthquakes of Holocene and latest Pleistocene ages in the states of Indiana and Illinois, USA. Engineering Geology, 50(3~4): 227~254.

    • Obermeier S F, Olson S M, Green R A. 2005. Field occurrences of liquefaction-induced features: a primer for engineering geologic analysis of paleoseismic shaking. Engineering Geology, 76(3~4): 209~234.

    • Owen G, Moretti M, Alfaro P. 2011. Recognising triggers for soft-sediment deformation: Current understanding and future directions. Sedimentary Geology, 235(3~4): 133~140.

    • Papathanassiou G, Pavlides S, Ganas A. 2006. The 2003 Lefkada earthquake: Field observations and preliminary microzonation map based on liquefaction potential index for the town of Lefkada. Engineering Geology, 82(1): 12~31

    • Qiao Xiufu, Song Tianrui, Gao Linzhi, Peng Yang, Li Haibing, Gao Mai, Song Biao, Zhang Qiaoda. 1994&. Seismi sequence in carbonate rocks by vibrational liquefaction. Acta Geologica Sinica, 68(1) : 16~36.

    • Qiao Xiufu, Gao Linzhi, Peng Yang, Li Haibing. 2001&. Seismic event, sequence and tectonic significance in Canglangpu Stage in Paleo-Tanlu Fault Zone. Science in China(Series D), 31(11): 911~918.

    • Quigley M C S. Bastin, B A Bradley. 2013. Recurrent liquefaction in Christchurch, New Zealand, during the Canterbury earthquake sequence. Geology, 41: 419~422.

    • Rodríguez-Pascua, M A Calvo J P, Vicente G D, et al. 2000. Seismites in lacustrine sediments of the Prebetic Zone, SE Spain, and their use as indicators of earthquake magnitudes during the Late Miocene. Sedimentary Geology, 135(1~4): 117~135.

    • Seed H B, Tokimatsu L F, Harder L F, Chung R M J. 1985. Influence of spt procedures in soil liquefaction resistance evaluations. Geotech. Eng, 111: 1425.

    • Shani-Kadmiel S M, Tsesarsky J N, Louie Z, Gvirtzman. 2012. Bulletin of the Seismological Society of America, 102(4): 1729~1739.

    • Shani-Kadmiel S M, Tsesarsky J N, Louie Z, Gvirtzman. 2014. Bulletin of earthquake engineering. 12: 607~625.

    • Simons M. S. E. Minson A. Sladen et al. 2011. The 2011 Magnitude 9. 0 Tohoku—Oki Earthquake: Mosaicking the Megathrust from Seconds to Centuries. Science, 332: 1421~1425.

    • Su Dechen, Sun Aiping, Guo Xianpu, Wang Sien. 2013&. Seismites in the Qilianshan area of the northeastern margin of Qinghai—Tibetan Plateau and a pilot research on the regional tectonics. Acta Petrologica Sinica, 29(6): 2223~2232.

    • Su Dechen, Qiao Xiufu. 2018&. Palaeoseismic thixotropic flow deformation in viscous sediments. Journal of Palaeogeography, 20(4) : 609~622.

    • Sukhija B S, Rao M N, Reddy D V, Nagabhushanam P, Lakshmi BV, Gupta H K. 2000. Current Status of Paleoseismology——An overview. Journal of Geological Society of India, 55: 465~480.

    • Tinsley J C, Youd T L, Perkins D M, Chen A T F. 1985. An Earth-Science Perspective. US Geological Survey Denver, CO US Geological Survey Professional Paper. 1360: 263~316.

    • Warren J K. 2010. Evaporites through time: Tectonic, climatic and eustatic controls in marine and nonmarine deposits. Earth-Science Reviews, 98(3): 217~268.

    • William L E. 2013. Injection-Induced Earthquakes. Science, 341(12): 1~7.

    • Zhong Jianhua, Cao Mengchun, Ni liangtian, Sun Ningliang, Liu Chunag, Hao Bing, Yang Guangqun, Song Guangxian, Ge Yuzhu. 2018&. Situation of study and development tendency of sandy dykes. Journal of Palaeogeography, 20(1): 119~132.

    • Zhong Jianhua, Ni Liangtian, Wang Lidong, Song Quanyou, Sun Ningliang, Song Guangxian, Xue Chunqi, Hao Bing, Shao Zhufu, Mao Cui, Liu Chuang, Cao Menchun. 2020a&. Discovery of large sand dikes in the Lower Cetaceous of southern Beilaishi, Lingshan Island, Qingdao, and their formation dynamics. Geological Review, 66(6): 1457~1476.

    • Zhong Jianhua, Wang Yongzhuo, Ni Liangtian, Gao Chuanwen, Zhou Yongbin, Mao Cui, Fan Xiaodong, Liu Guozhi, Chen Zhipeng, Song Guangxian, Wang Rui, Huang Liang, Liu Quang, Fan Yujiao, Ren Kaixuan. 2020b&. Sand-dike (Sand injection) type of Reservoir and Oil Pool. Journal of Earth Science and Environment, 42(3): 285~296.

    • Zhong Jianhua, Sun Ningliang, Hao Bing, Ni Liangtian, Song Guanxian, Xuan Chunqi, Mao Cui, Shao Zhufu, Ge Yuzhu, Chen Bin, Liu Shengxin, Peng Shaofeng, Fu Donghui, Wang Yongqiang, Li Cong. 2020c&. Characteristics and genetic dynamics analysis of Lower Cretaceous bedding sand veins in the Laishi Section of Lingshan Island, Shandong Province. Journal of Palaeogeography, 22(4): 635~646.