地震波引起的井水位水温同震变化及其机理研究
|
摘要
地下水是地球系统的重要组成部分,地下水的运动变化与人类的生存环境息息相关。地震是地球活动的突发常发事件。地震与地下水动态的关系是近几十年来人们努力研究的内容,地震引起的井水位水温同震效应是当前大家非常关注的热点问题。本论文以这一命题为科学问题开展研究工作,试图揭示地下水位水温同震变化的时空分布规律、影响因素、相互关系和产生机理,这对于厘清地震与地下水动态的关系、研究地壳活动规律、减轻次生灾害、跟踪后续地震、追溯地震前兆等都具有重要的理论和实际意义。中国的地下水位水温观测台网,经过七、八十年代的起步建设,“九五”、“十五”期间的发展,观测井点数量分别达到了约420个和277个,尤其是数字化的观测仪器,采样率达到了分钟值,快速、连续和方便,为深入研究井水位水温同震效应奠定了坚实的数据基础。
本论文在对地下水位水温同震变化研究现状系统调研的基础上,发现了目前研究中存在的问题,主要表现在:多为定性分析,缺乏进一步的定量研究;水位水温同震变化相互关系未见有系统研究;对水温影响因素的研究较为缺乏;对水位水温同震变化机理的认识尚不深入和统一等。针对以上几个方面的问题,本论文首先选择代表性井点,对其多年连续观测数据进行细致分析,研究一井多震的水位水温同震变化规律;然后依照一定的原则对全国前兆台网数据库的观测井点进行筛选,系统收集被选井点水位水温对2008年汶川Ms 8.0级地震的响应,研究一震多井的水位水温同震变化特征和同井水位水温变化的组合类型;并挑选出以汶川震中为中心1000 km范围内的井点,收集其在2007年印尼苏门答腊Ms 8.5前后的观测资料,对比分析了汶川近震和苏门答腊远震引起的水位水温同震变化的异同及其影响因素;最后以北京塔院井为实验场地,对水位和不同深度的水温动态变化进行对比观测,探讨了水温动态的影响因素,并建立模型进行数值模拟,分析探讨水位水温同震变化的机理。
论文的主要工作可归纳为如下几个方面:
(1)水位同震变化的研究
水位同震变化可分为振荡和阶变(包括上升和下降),水位振荡是含水层对地震波的弹性响应,前人已有较多的研究,这里主要关注阶变型同震变化。本论文首先选择云南思茅大寨井和新疆乌鲁木齐新04井两口代表性井孔对水位同震阶变上升和下降变化特征进行研究;然后系统收集了井水位在2008年汶川Ms 8.0级和2007年印尼苏门答腊Ms 8.5前后的同震变化观测资料,分析对比了近震和远震引起的水位同震变化的异同。
对一井多震的的研究结果显示,同一口井的水位同震变化的方向不因地震的不同而改变,无论地震的方位、距离、大小和震源机制如何,上升的总是上升,下降的总是下降。水位同震变化幅度主要受震级和井震距控制,三者之间有良好的相关性。思茅井水位的同震变化总是上升,上升幅度随震级的增大而增大,随井震距的增大而减小,同时也受季节性降雨或者区域构造环境和局部应力状态变化的影响,地震波跨过红河断裂时会使井水位同震变化幅度偏离总体统计关系;分析认为渗透系数减小可能是水位同震上升的主要原因。
新04井水位的同震变化总是下降,下降幅度随震级的增大而增大,随井震距的增大而减小,同时也受本地应力状态的影响,在新04井周围发生的3个5级左右地震前后,井水位同震响应能力显著降低,可能是孔隙缩小、断层闭锁,水流受阻所致。新04井井孔穿过了断层破碎带,当受强震s波或面波能量的激发时,断层发生张合作用或蠕动,井水外泄可能是井水位同震下降变化的原因。
2008年汶川地震是地下水位水温观测网在“九五”、“十五”项目建设完成后国内发生的最为显著的地震,因此在一震多井分析时,选择以汶川地震为例。首先对前兆台网库中的井点进行了筛选。为兼顾下文对水温同震响应和同井水位水温同震变化组合类型的分析,选择观测井点时遵循以下原则:①井点要同时具有水位水温两个观测项,且在2008年5月12日地震前后仪器都正常工作;②汶川地震至少引起了水位水温中一个测项的同震变化;③水位水温具有相对稳定的动态背景。依照以上原则对前兆数据库有记录的井点进行了筛选,选出了96口井,并对其水位同震变化进行了分类统计。分类统计结果显示,2008年汶川Ms 8.0地震引起的水位同震变化以上升为主。
同时,在以上的96口井中,挑选出以汶川震中为中心1000 km范围内的32个井点,收集其对2007年印尼苏门答腊Ms 8.5级的响应,对比分析汶川近震和苏门答腊远震引起的水位水温同震变化的异同。对比结果显示,印尼远震引起水位同震上升或下降阶变的6口井,在汶川近震时其阶变方向仍然不变;在印尼远震引起水位振荡的18口井中,在汶川近震时除4口仍然振荡外,其余14口都产生了或上升或下降的同震阶变;在印尼远震无变化的8口井点中,在汶川地震时除1口无变化外,其余7口都产生了或上升或下降的同震阶变。
对比分析结果表明:无论对于远震还是近震来说,水位同震上升的比例都远大于同震下降变化的比例;相对于远震,近震引起的水位同震阶变井点数量大大增加,振荡和无变化井点数量减少;水位同震升降的方向不因地震的远近、大小、震源机制或地震方位的变化而改变,更多地受控于本地的地质构造环境和水文地质条件;地震波能量的变化不能改变水位同震变化的方向,当地震波能量足够大时,会使一些原来仅产生振荡或无同震响应的井孔的水位发生阶变。
(2)水温同震变化的研究
水温同震效应有上升和下降变化,但没有观测到振荡现象。水温同震变化与水位变化关系密切,对一井多震的研究结果表明,水位同震变化是水温同震变化的必要条件,有水温同震变化的震例都存在水位同震变化,而发生水位同震变化的震例则不一定有水温同震变化。研究结果显示,水位水温同震变化主要存在4种组合,它们分别是:水位上升水温上升型、水位下降水温上升型、水位下降水温下降型、水位振荡水温下降型。本论文首先选择云南思茅井、新疆新04井、北京左家庄井和塔院井作为上述4种组合的代表进行分析讨论;然后系统收集了从全国地下流体观测网中筛选出的96口井对2008年汶川Ms 8.0级地震的水温响应资料进行分析;最后在以汶川震中为中心1000 km范围内挑选出32个井点,收集其在2007年印尼苏门答腊Ms8.5前后的同震变化观测资料,分析对比了近震和远震引起的水温同震变化的异同及其影响因素。
水位上升水温上升型:以云南思茅井为例,水位、水温同震变化类型都为上升,变化起始时间上几乎同步,水位同震上升幅度为几十个cm的量级,最小12 cm,最大60.6 cm;水温同震上升幅度则为几十至上百10-4℃,两者之间的比例关系为几个℃/100m,量级与正常地温梯度3℃/100m基本相同。观测系统安装时的测量结果显示,水温探头附近为随深度增加的正梯度区,思茅井长时间水位水温动态对比也显示两者同向变化,结合水位水温变化幅度的比值与地温梯度量级相当的特点,分析认为水位同震上升是水温同震上升的直接原因。
水位下降水温上升型:以新疆新04井为例,水位同震变化为下降-恢复型,而水温同震变化主要表现为稳定-上升-恢复。地震引起的水位同震下降幅度最大58 mm,当水位同震阶变大于10 mm时,水温记录到同震变化,对应的水温上升幅度为几十个10-4℃,两者之间的比例关系可以达到～10℃/100m,远大于测量到的井孔地温梯度～3℃/100m。而且水温同震上升的起始时间不是与水位开始下降的时间同步,而是与水位下降后开始恢复的时间基本相当。
分析认为,新04井孔水温探头下方可能存在热的含水层,它与断层的共同作用,控制着井水位、水温的变化。当地震波激发井水沿断层外泄时,水位下降引起的水温降低与含水层补给热水产生的水温升高中和平衡,温度不发生明显变化(稳定);地震波过后井水沿断层外泄减少,而含水层中的热水继续流向井内,井水位上升,温度上升;井水位上升停止后,含水层中的热水也停止流向井内,受地温梯度的控制,井水温下降恢复到正常状态。
水位下降水温下降型:以左家庄井为例,温度探头位于井孔内有套管封闭的正梯度段内,水位水温同震变化为同步下降。记录到的水位同震变化幅度1-2 m,水温同震变化为～0.03—0.04℃,两者比例关系～2-3℃/100m。水温同震变化机理类似于思茅井水位水温同步变化,体现了正梯度状态下水位同震下降对水温同震变化的影响。
水位振荡水温下降型:以塔院井为例,井水位同震变化总是振荡,水温同震变化形态总是具有下降-上升-恢复的过程,不受地震方位和震源机制影响;水温同震变化总是发生在地震波到达和水震波开始之后。已记录到的最大水温同震变化达0.097℃。
分析认为,井孔中的水体受振荡激发而加速对流与掺混是导致塔院井水温同震下降的主要原因。水温探头一般放置在较深的部位,在正常情况下温度较高,当受到地震波的作用时,井孔中水体对流加速,深部热水体上涌,而浅部较泠水体下沉,水温探头将先观测到温度下降现象。随水震波的逐渐平息,探头附近井水温逐步恢复上升。
塔院井水温同震下降幅度随震级的增大而增大,随井震距的增大而减小。个别发生在井孔东侧的地震引起的温度同震变化小于理论计算值,可能是黄庄-高丽营断裂影响所致。
2008年汶川Ms 8.0地震引起的水温同震变化统计结果显示,在有水位或水温同震效应的96口观测井中,66口井记录到水温同震变化,上升和下降井孔各为31和35,以同震下降井孔所占比例略高;在66口井中,其中65口井都有相应的水位同震变化,只有1口井因特殊原因未观测到相应水位同震变化。同井水位水温同震变化关系的分类统计结果显示,在水位有同震升降阶变的77口井中,水位水温同方向变化(水位上升温度上升或者水位下降温度下降)的有31口井,反方向变化(水位上升温度下降或者水位下降温度上升)的有23口井,水温无变化的有23口井,以水位水温同方向变化占优势。对于水位振荡变化的18口井,水温无变化的井孔有7口,上升和下降的分别为2口和9口,水温以下降为主。
以汶川Ms 8.0地震为中心1000 km范围内的32个井点中,2007年印尼苏门答腊Ms 8.5级远震与汶川Ms 8.0近震引起的水温同震变化的比较显示,相对于远震,近震引起的水温同震阶变井点数量增加,无变化井点数量减少。当地震波能量足够大时,会使一些原来无同震响应的井孔发生水温阶变,但大多数井水温同震升降的性质都不因地震的远近、大小、震源机制或方位的变化而改变。两个发生水温同震升降性质变化的井点是由于水的自流状态改变或水位同震变化由振荡转为阶变而引起。
对一震多井和远近震例对比研究结果表明,水温同震变化是以水位的同震变化为前提条件,水温同震升降的性质不因地震的不同而改变,同震变化的幅度与震级、井震距、季节、地温梯度、探头放置位置等因素有关。
(3)不同深度井水温对比观测及水位水温动态关系的数值模拟研究
为探讨井水温动态变化的影响因素,以北京塔院井为实验场地,使用新增的一台高精度温度仪,进行详细的水温梯度测量和不同深度水温动态连续测量,把新探头获取的不同深度的水温动态数据与原有的水位水温仪(2001年7月安装)观测数据进行对比研究。同时参考唐山矿井和西昌太和井的观测资料,对比分析了井孔水温动态的影响因素。
详细的水温梯度测量显示,塔院井在距井口约105～180 m处存在一负梯度段,分析认为负梯度带的出现可能是局部段落有冷流体的存在和渗入所致。不同深度的实验测试结果表明,在井孔浅部,水温梯度变化大,其潮汐效应不明显;在深部水温正梯度段,水温水位同向变化;在负梯度段,水温水位反向变化;水温梯度越大,潮汐效应幅度越大。水温探头放置处的水温梯度的差异及其与含水层的相对位置是水温动态复杂性的重要原因。
引入了流体运动的控制方程和热传导方程。COMSOL Multiphysics是专为描述和模拟各种物理现象而开发的一个专业有限元数值分析软件,可模拟所有可用偏微分方程(PEDS)描述的物理过程。本论文选用最新版本COMSOL 4.1对塔院井水位水温同震变化进行数值模拟研究。分析中,对于井水面的升降或波动变化使用COMSOL软件中的移动网格法(ALE)来实现。
为了解释水位震荡水温下降的现象,建立了一个三维模型进行模拟分析。几何模型取井半径r=0.25 m,井深H=100 m,上下温差3℃。由于井孔及其所包含的水体构成了一个轴对称结构,研究中将三维模型简化为二维轴对称模型。在井孔下部侧壁施加周期为5 s和20 s正弦波组成的压力波动,模拟地震波对井水的作用。结果表明,在井水上表面出现了明显的水位振荡变化;在井孔中部施加压力波的上方出现水温振荡下降现象,与许多井水位振荡水温同震下降的实际观测结果一致。在实际观测中,可能由于水温探头的延迟效应和分钟值的采样率,难于观测到水温的振荡效应。
通过对井水位水温同震变化的系统实验观测和分析研究,本研究取得的主要新认识如下:
①同一口井水位同震升降的性质主要受井孔局部水文地质条件控制,不因地震方位、大小、距离和震源机制的不同而改变,上升的井总是上升,下降的井总是下降。
②单井水位升降幅度随震级的增大而增大,随井震距的增大而减小,三者之间有良好的相关性;地震波跨越断层或者说来自于某个方向的地震波、季节性的降水会使同震变化的幅度偏离总体统计关系;井孔附近孕震可能使同震响应能力发生改变。
③同一口井的水位同震变化是水温同震变化的必要条件,水温同震响应总是出现在地震波到达和水位同震变化开始之后;水温同震变化的幅度受到震级、井震距、季节、地温梯度、探头放置位置等因素的影响。
④无论对于远震还是近震来说,同一地震引起的水位同震变化以上升为主,可能是渗透率的降低所致;同震水位下降的井多穿过了断层破碎带,地震波激发使断层发生张合作用或蠕动,井水外泄可能是井水位同震下降的原因。
⑤相对于远震,近震引起的水位水温同震阶变井点数量大大增加,振荡和无变化井点数量减少;当地震波能量足够大时,会使一些原来仅产生振荡或无同震响应的井孔的水位水温发生阶变,但不能使水位水温阶变的性质发生变化。个别发生水温同震升降性质变化的井点是由于水的自流状态改变或水位同震变化由振荡转为阶变而引起。
⑥水温探头放置处的水温梯度差异及其与含水层的相对位置是水温动态复杂性的重要原因。在深部正梯度段,水温水位同向变化;在负梯度段,水温水位反向变化。水温梯度越大,潮汐效应和同震变化幅度越大。塔院井负梯度带的出现可能是局部段落有冷流体的存在和渗入所致。
⑦同井中水位水温同震阶变的关系以两者同方向变化(水位上升温度上升或者水位下降温度下降)比反向变化(水位上升温度下降或者水位下降温度上升)占优势;同震水位振荡型井孔,其深部水温同震变化以下降-恢复为主,其原因是井孔中的水体受振荡激发而加速对流与掺混所致。
由于时间紧、工作量大、数值模拟软件晚到、实验井(塔院井)改造等原因,论文仍然存在许多问题,例如每个井孔有自己的独特的特征,要想对一些观测现象深入分析研究,需要有井孔的水文地质、井孔结构、温度梯度等有详细的资料,而很多井孔的资料都很不完整;一井多震和一震多井的研究实例有限;在数值模拟分析中,流体运动的复杂性,有些现象很难模拟等问题,都有待于将来进一步深入研究。
Groundwater is an important component of the earth system, of which the movement and change are closely associated with human living environments. Earthquakes are sudden and frequent events of on the earth. The relationship between earthquakes and groundwater change has been the subject that scientists make great efforts to study for tens of years. Coseismic effects of well-water level and temperature caused by earthquakes is currently the outstanding topic receiving much attention. In this thesis, focused on this problem that, an attempt is made to reveal the distribution rule in space and time, influence factors, mutual relations and formation mechanisms of coseismic variations of groundwater level and temperature. It is of great significance in both theory and practice to clarifying relationship between earthquakes and groundwater variation, studies of crustal movement, reduction of secondary disasters, tracking following earthquakes, and earthquake precursor retrospection. The observation network of groundwater level and temperature in China has experienced prelimilary constructions in 70’s and 80’s, developments in ninth 5-year plan and tenth 5-year plan, and the numbers of observation wells have reached 420 and 277 respectively. Especially, the sample rate of digital observation apparatus, has reached one value per minute, is rapid, continuous, and convenient, which lays solid data foundation for a detailed study on the coseismic effects of well-water level and temperature.
Fist, this thesis makes a systemic investigation is to the current research status on coseismic variations of groundwater level and temperature. It finds that there are many existing problems. For example, most work is of qualitative analysis, lacking further quantitative analysis. There is no report on systemic research on the relationship between groundwater level and temperature in coseismic variations. Researches on the influence factors on water temperature are rare. The understanding on the mechanism of coseismic variations of groundwater level and temperature is still controversial.
Targeting the problems above, some representative wells are selected for a detail analysis of the continuous observation data of several years, and for the study of the coseismic variation rules of groundwater level and temperature under one well-multi earthquakes conditions. The observation wells in the database of national precursor network are selected according to specific principles. The observation data of groundwater level and temperature of the selected wells are then colleclected systemically, which are around the time when the Wenchuan Ms 8.0 (2008) event occurred, for studying the coseismic variation of groundwater level and temperature in the case of one earthquake-multi wells, and the combination type of groundwater level and temperature variations in the same well. Those wells within the range of 1000 km with the center of Wenchuan epicenter are selected to collect their observation data around Sumatra Ms8.5, Indonesia (2007) event. Next this thesis analyzes and compares the differences and similarities in coseismic variations of groundwater level and temperature caused by nearby and remote earthquakes and their influence factors. Finally taking the Tayuan well in Beijing as the test site, apply a comparison observation on groundwater level, dynamic variations of groundwater temperature at different depths, discuss the influence factors for variations of groundwater temperature. And models are constructed to perform numerical modeling and explore the mechanism of coseismic variations of groundwater level and temperature.
The main work in the thesis can be concluded as following.
1 Coseismic variations of groundwater level
Coseismic variations of groundwater level can be divided to oscillation type and step type (including ascent and descent). The groundwater oscillation is the elastic response of aquifer to earthquake waves, which has been well studied. So this work concerns the coseismic variations of the step type. Two representative wells, the Dazhai well in Simao, Yunan Province and Xin 04 well in Urumchi, Xinjiang are selected firstly for the study of ascending and descending characteristics in coseismic variations of groundwater level. Then, the coseismic variation data of groundwater levels before and after the Wenchuan Ms 8.0 (2008) and Sumatra Ms 8.5, Indonesia (2007) events are collected for analysis and comparison of the differences and similarities in coseismic variations of groundwater level and temperature caused by nearby and remote earthquakes and their influence factors.
The research results of the case one well-multi earthquakes indicate that the coseismic variation directions of groundwater level at a same well do not change with different earthquakes. It keeps ascent or descent for any orientations, distances, magnitudes and mechanisms of the earthquakes. The amplitudes of coseismic variations of groundwater level depend upon the seismic magnitudes and well-earthquake distances, of which the three are well correlated.
The coseismic variation of groundwater level in the Simao well is always ascending with growing amplitude with increasing magnitudes of earthquakes and declining amplitude with increasing well-earthquake distances. At the same time, it is also affected by seasonal rainfalls, regional tectonic settings and change of local stress state. The magnitude of coseismic variations of groundwater level will deviate from the general statistic relation when earthquake waves cross the Red River fault or a shock takes place near the well. Analysis suggests that the decrease of permeability may be the primary reason for coseismic ascending of groundwater level.
The coseismic variation of groundwater level in the Xin 04 well is always descending, with growing amplitude with increasing magnitudes of earthquakes and decreasing amplitude with increasing well-earthquake distances. Meanwhile it is also affected by local stress status. When three earthquakes with magnitudes around Ms. 5.0 occurred near the Xin 04 well, the coseismic response ability of groundwater level decreased notably, which may be due to pore shrinking, fault locking, and stream blocking. The Xin 04 well penetrates a fault fractured zone on which will open-close action or creep would occur when it is triggered by S waves or surface waves of strong shocks, where the outflow of well water may be the reason for coseismic descending of groundwater level.
Wenchuan Ms 8.0 earthquake (2008) was the most remarkable one that occurred in China when the projects of groundwater level and temperature in ninth 5-year plan and tenth 5-year plan finished. So Wenchuan earthquake is selected to analyze the effect of one earthquake-multi wells. The wells recorded in the precursor networks are selected firstly. With giving attention to the combination analysis of coseismic effects of groundwater temperature and coseismic variations of groundwater level and temperature in same well, the observation wells are selected following such principles:①the well station must have the two observation items as groundwater level and temperature, and work normally before, during and after the earthquake on May 12, 2008.②the Wenchuan earthquake has caused coseismic variations of at least one item of groundwater level and temperature.③the dynamic background of groundwater level and temperature is relatively stable. According to these principles, 96 wells are selected from the recorded wells in the database of precursor networks, and the coseismic variations of groundwater level in these wells are classified, the statistics shows that the coseismic variations of groundwater level caused by the Wenchuan Ms 8.0 earthquake (2008) are dominated by water ascent.
Among the above 96 wells, 32 wells in the area of 1000 km centered by the epicenter of the Wenchuan event are selected to collect their responses to Sumatra Ms 8.5, Indonesia (2007), in order to have further comparison of the differences and similarities in coseismic variations of groundwater level and temperature caused by nearby and remote earthquakes. The result shows that 6 wells which exhibited ascent or descent related with the Sumatra Ms 8.5, Indonesia, 2007 kept same change when the Wenchuan event occurred. And among the 18 wells with oscillation variations of groundwater level associated with the Sumatra event of 2007, except 4 wells had oscillations related with the Wenchauan shock, the other 14 wells showed ascent or descent changes when the Wenchuan event took place. In the rest 8 wells which looked like quiet re during the remote Indonesia remote earthquake, only one well kept invariant while the other 7 wells showed coseismic ascent or descent.
The comparative analysis suggests that the proportion of coseismic ascent of groundwater level is far greater than that of coseismic descent for either remote or nearby earthquakes. The number of wells with coseismic stepwise changes of underground water caused by nearby earthquakes is larger, while that of wells with oscillation or without change is small with respect to that by remote events. The directions of coseismic ascent or descent of groundwater level do not depend on distances, magnitudes, mechanisms or azimuths of earthquakes, instead are controlled by local geological settings and hydrologic conditions. If the energy of seismic waves is strong enough, some wells that only have oscillations or no coseismic responses originally will show stepwise variations of groundwater level.
2 Coseismic variations of groundwater temperature
There is ascent or descent in coseismic responses of groundwater temperature, but no oscillation is observed. Such changes are closely related with coseismic variations of groundwater level. The case study of one well-multi earthquakes suggests that coseismic variation of groundwater level seems to be the necessary condition for coseismic variation of groundwater temperature. It means that the earthquakes that have coseismic variations of groundwater temperature also show coseismic variations of groundwater level, while the events that have coseismic variations of groundwater level are not always accompanied by coseismic variations of groundwater temperature. In the research, the coseismic variations of groundwater level and temperature can be divided into four main types as ascending groundwater level and ascending groundwater temperature, descending groundwater level and ascending groundwater temperature, descending groundwater level and descending groundwater temperature, and oscillating groundwater level and descending groundwater temperature. In the thesis, the Simao well inYunnan, Xin 04 well in Xinjiang, Zuojiazhuang well in Beijing, and Tayuan well in Beijing are selected as the representing four cases for further analysis and discussion. Then the response data of groundwater temperatures before, during and after the Wenchuan Ms 8.0 (2008) are collected systemically from the 96 wells which are selected from the national underground fluid observation networks. 32 wells in the area of 1000 km centered by the epicenter of the Wenchuan event are selected to collect their responses to Sumatra Ms 8.5, Indonesia (2007). The differences and similarities together with their influence factors in coseismic variations of groundwater temperature caused by nearby and remote earthquakes are analyzed and compared.
Type of ascending groundwater level and ascending groundwater temperature
Taking the Simao well in Yunnan as an example, the coseismic variations of both groundwater level and groundwater temperature are of ascent, which start nearly at the same time. The amplitude of coseismic ascent of groundwater level has a magnitude of tens of centimeters with minimum value as 12 cm and maximum value as 60.6 cm. The amplitude of coseismic ascent of groundwater temperature ranges from tens to hundred 10-4℃. The ratio between them is several℃/100m which is roughly the same as the normal geothermal gradient of 2-3℃/ 100m. The measurements when the observation system was installed show a positive gradient area where water temperature increases with depth near the groundwater temperature sensor. The long-term dynamic comparison of groundwater level and temperature in the Simao well also shows changes in a same direction. Considering that the ratio between the scopes of groundwater level and temperature is equivalent to the order magnitude of groundwater temperature gradient, it is indicated that coseismic ascent of groundwater level is the direct reason for coseismic ascent of groundwater temperature.
Type of descending groundwater level and ascending groundwater temperature
At the Xin 04 well in Xinjiang, the coseismic variations of groundwater level are of descending-resuming type, while that of groundwater temperature mainly displays as stability-ascent-recover. The start time of coseismic ascent of groundwater temperature is comparable to the time when groundwater level changes from descending to resuming. The maximum amplitude of coseismic descending of groundwater level is 58 mm. When the coseismic step of groundwater level is greater than 10 mm, the coseismic variations of groundwater temperature also take place with corresponding amplitude of groundwater temperature ascending as tens of 10-4℃. The ratio between them can reach 10℃/100m which is greater than normal geothermal gradient of 3℃/100m. Moreover, the starting time of coseismic ascending of groundwater temperature is not synchronous with the time when groundwater level begins descending, but basically equivalent to the time when groundwater level begins recovering after descending.
Further analysis indicates hot aquifer may exist below the groundwater temperature sensor in the Xin 04 well. This hot aquifer will act together with the fault and control the variations of groundwater level and temperature in the well. When the well water is triggered by seismic waves and leaks out, the temperature ascent caused by groundwater level descent will counteract the groundwater temperature ascending caused by hot water supplied by aquifers and come to a balance, so the temperature will not change evidently. After the seismic waves are over, the leakage of well water decreases, while the hot water in aquifers will flow into the well continuously, which results in ascent of well water level and temperature. When the well water level stops ascending, the hot water in aquifers will also cease to flow into the well, and the well water temperature resumes to normal status under the control of the geothermal gradient.
Type of descending groundwater level and descending groundwater temperature
In the Zuojiazhuang well in Beijing, the temperature sensor is located at the positive gradient section sealed by well casing. The coseismic variations of groundwater level and temperature are of synchronous descending. The recorded amplitude of coseismic variation of groundwater level is in the range of 1-2 m; the recorded coseismic variation of groundwater temperature is in the range of 0.03-0.04℃; the ratio between them is in the range of 2-3℃/100 m. The mechanism of coseismic variation of groundwater temperature is similar to that observed at the Simao well in Yunnan Province, which suggests the influence of the coseismic descending of groundwater level on that of groundwater temperature in a positive gradient status.
Type of oscillating groundwater level and descending groundwater temperature
In the case of the Tayuan well in Beijing, the coseismic change is always oscillating, while that of groundwater temperature is usually of a descending-ascending-resuming process and independent of azimuths and mechanisms of earthquakes. The coseismic variations of groundwater temperature often occur at the time when seismic waves arrive and after the water level seismic wave starts. The recorded maximum coseismic variation of groundwater temperature can reach 0.097℃.
Further analysis suggests that the water body in the well will be excited by oscillation, resulting in accelerated convection and mixture, which is the main cause for coseismic descending of groundwater temperature in the Tayuan well. The water temperature sensor is commonly placed at a relatively deep section of the well with relatively high temperature in normal conditions. When the seismic waves come, the water body in the well will accelerate convection and deep hot water ascends, while the shallow cold water falls down, so the water temperature sensor will observe temperature descending firstly. With the water level seismic waves calming down, the well water temperature around the sensor will resume ascending gradually.
The amplitude of coseismic descent of groundwater temperature increases with the growing magnitudes of earthquakes and decreases with the increasing well-earthquake distance. The coseismic variations of groundwater temperature caused by individual earthquakes east of the well are smaller than theoretic values, which may be due to the influence of the Huangzhuang-Gaoliyin fault there.
Classification statistics of coseismic variations of groundwater temperature caused by Wenchuan Ms 8.0 earthquake (2008) is made. Among 96 observation wells that have coseismic effects of groundwater level and temperature, there are 66 wells that exhibit coseismic variations of groundwater temperature, in the 66 wells, 65 wells show corresponding coseismic variations of groundwater level, the well that no corresponding coseismic variations of groundwater level were observed may be caused by some special reasons. In the classification statistics of coseismic variations of groundwater level and temperature in same well, there are 77 wells that show coseismic ascending or descending steps of groundwater level, of which 31 wells show changes of groundwater level and temperature in same directions; 23 wells with changes in opposite directions; and 23 wells without variation of groundwater temperature. The wells with changes of groundwater level and temperature in same directions are predominant. Among the 18 wells with oscillation of groundwater level, 7 wells show no changes in groundwater temperature; 2 wells show temperature ascending; 9 wells show temperature descending; and descent dominates the groundwater temperature.
Among the 32 wells with a range of 1000 km centered by the epicenter of the Wenchuan Ms 8.0 event in 2008, the coseismic variations of groundwater temperature by this event and the Sumatra Ms 8.5 of 2007 are compared. It indicates that there are more fewer wells without such changes, with respect to the case of remote earthquakes. If the energy of seismic wave is strong enough, some wells without coseismic responses originally will show water temperature steps. Of course, the properties of most coseismic ascending or descending of groundwater temperature will not change with the variations in distances, magnitudes, mechanisms or azimuths of the earthquakes. The two wells that occurred character change in coseismic ascending or descending of groundwater temperature were caused by artesian flowing change or the coseismic variation of groundwater level changing from oscillating to step.
Case comparison of one earthquake-multi wells and nearby and remote earthquakes also indicate that coseismic variations of groundwater temperature occur on the premise of coseismic variations of groundwater level. The properties of coseismic ascending and descending of groundwater temperature will not change with different earthquakes. The amplitude of coseismic variation is related with such factors as magnitudes, well-earthquake distance, season, geothermal gradient, and location of temperature sensor.
3 Test of well water temperature sections and numerical modeling of dynamic relation between groundwater level and temperature
The Tayuan well in Beijing is selected as the test site for discussing influence factors of well-water temperature variation. With a new high-precision temperature meter, detail measurement of water temperature gradients and dynamic continuous measurements of groundwater temperature at different depths have been carried out. The dynamic data of groundwater temperature at different depths in the Tayuan well are obtained with new sensor that allow a comparative study with data obtained by intrinsic groundwater level and temperature meter (which was installed in July, 2001). At the same time, the observed data from the Tangshan mine well and Taihe well in Xichang are also referenced to analyze the influences of groundwater gradient on tide and postseismic effects.
From the detailed measurements of water temperature gradient, it exists a negative gradient section at 105-180 m away from the mouth of the well. the occurrence of negative gradient section may be caused by cold fluid existing at or flowing into local sections The test results at different depths indicate that water temperature gradients are large and the tidal effects are not evident at the shallow part of the wells. The groundwater level and temperature show change in same directions in the positive gradient section of deep groundwater temperature; and in opposite directions in the negative gradient section. The larger the groundwater gradient is, the bigger the amplitudes of tidal effects become. Differences of water temperature gradients around the sensor and its relative position to aquifer are the important factors that cause complex dynamic variations of water temperature.
Control equations of fluid motion and heat exchange equations are introduced in this study. COMSOL Multiphysics is a professional FE numerical analysis software that is developed for describing and modeling varies of physical phenomena; it can model those physical procedures expressed with partial difference equations. Numerical modeling of coseismic variations of groundwater level and temperature is carried out with COMSOL 4.1 software. The ascending and descending or oscillating of groundwater level are realized with Arbitrary Lagrangian-Eulerian (ALE) method in COMSOL software.
A three-dimensional model is constructed to model and explain the phenomenon of water level oscillating and water temperature descending. The geometry of the model includes 0.25 m of well radius (r) and 100 m of well depth (H). The temperature difference between the top and bottom of the well is 3℃. Because the borehole and the water comprised in it construct an axial symmetry structure, the three-dimensional model is simplified to a two-dimensional axial symmetry model. Compressive waves composed of sine waves of 5 s and 20 s are exerted at the side wall of the bottom section of the well to simulate the effect of seismic waves on well groundwater. The results show that obvious groundwater level oscillation appears on the upper surface of well water; the groundwater temperature oscillates downward in the middle part of the well where compressive waves are exerted, which is coincided with many observations of groundwater level oscillations and coseismic descending of groundwater temperature. Because of the delayed effect of water temperature sensor and sample rate of one value per minute, it is difficult to observe the oscillation effects of groundwater temperature.
By systemic test observations and analysis of coseismic variations of groundwater level and temperature, this thesis has obtained some insights presented below.
(1) The coseismic ascending or descending of groundwater level and temperature at the same well is mainly controlled by local geohydrologic conditions, and will not change with the orientations, magnitudes, distances, and mechanisms of earthquakes. Those wells with groundwater ascending are always in an ascending state, while those with groundwater descending wells are always in a descending regime.
(2) The ascending or descending amplitude of groundwater level in a single well increases with growing magnitudes of earthquakes and decreases with increasing well-earthquake distances. There is a good correlation among them. Moreover, such factors as seismic waves crossing faults or coming from a special direction, and seasonal rainfall will cause the amplitude of coseismic variation deviate from general statistics relation. Earthquakes preparation near the well may probably reduce coseismic response capability.
(3) Coseismic variation of groundwater level is the necessary condition for coseismic variation of groundwater temperature at the same well. The coseismic response of groundwater temperature usually appears at the time when seismic waves arrive and coseismic variations of groundwater level begin. The amplitude of coseismic variation of groundwater temperature is affected by such factors as magnitude, well-earthquake distance, seasonal rainfall, geothermal gradient, locations of water temperature sensor.
(4) The coseismic variations of groundwater level caused by a same earthquake are mainly ascending, which is true for either nearby or remote earthquakes and may be caused by permeability decrease. Most wells with coseismic descending of groundwater level may penetrate fault fracture zones. The seismic waves will cause the faults to open and close or creep. The leakage of well water may cause coseismic descending of groundwater level.
(5) Compared with remote earthquakes, the number of wells with coseismic steps of groundwater level and temperature caused by nearby earthquakes is much larger, while the number of wells with oscillations or without changes is less. If the energy of seismic waves is strong enough, the groundwater level and temperature of some wells that only oscillate or show no coseismic responses originally will show steps, but the seismic waves cannot change the step character of groundwater level and temperature. The character change of coseismic ascending or descending of groundwater temperature in individual well was caused by variation in water artesian flowing or coseismic variation of groundwater level changing from oscillation to step.
(6) Differences in groundwater temperature gradients around the water temperature sensor and its relative position to aquifer are important factors causing complexity in dynamic variation of groundwater temperature. The groundwater level and temperature show syn-directiion change at deep sections of positive gradients but reverse- direction change at sections of negative gradients. The larger the groundwater gradient is, the bigger the amplitudes of tidal effects and coseismic variations become. The occurrence of negative gradient section in Tayuan well may be caused by cold fluid flowing into local sections of the well.
(7) In the same well, the relation of coseismic stop of groundwater level and temperature shows that changing in same direction is dominant to changeing in opposite directions. For wells with coseismic oscillations of groundwater level, the deep groundwater temperature variations are mainly of coseismic descending-resuming. The reason is that the water body in the well is triggered by oscillations so that water convection and mixture are accelerated
Because some objective factors such as limited time available, large amount of work, numerical modeling software that can be used only lately, and rebuilding of the test well (Tayuan well) , there are many problems that exist in the article or remain unsolved. For example, each well has its special characters; the detailed data of geohydrologic condition, structure of borehole, geothermal grident, etc. are needed to have thorough analysis of observations, but these data of many wells are incomplete. There are limited examples for the case studies of one well-multi earthquakes and one earthquake-multi wells. The movement of fluid is complex; some phenomena are hard to be modeled. These problems will be studied further in the future.
本论文在对地下水位水温同震变化研究现状系统调研的基础上,发现了目前研究中存在的问题,主要表现在:多为定性分析,缺乏进一步的定量研究;水位水温同震变化相互关系未见有系统研究;对水温影响因素的研究较为缺乏;对水位水温同震变化机理的认识尚不深入和统一等。针对以上几个方面的问题,本论文首先选择代表性井点,对其多年连续观测数据进行细致分析,研究一井多震的水位水温同震变化规律;然后依照一定的原则对全国前兆台网数据库的观测井点进行筛选,系统收集被选井点水位水温对2008年汶川Ms 8.0级地震的响应,研究一震多井的水位水温同震变化特征和同井水位水温变化的组合类型;并挑选出以汶川震中为中心1000 km范围内的井点,收集其在2007年印尼苏门答腊Ms 8.5前后的观测资料,对比分析了汶川近震和苏门答腊远震引起的水位水温同震变化的异同及其影响因素;最后以北京塔院井为实验场地,对水位和不同深度的水温动态变化进行对比观测,探讨了水温动态的影响因素,并建立模型进行数值模拟,分析探讨水位水温同震变化的机理。
论文的主要工作可归纳为如下几个方面:
(1)水位同震变化的研究
水位同震变化可分为振荡和阶变(包括上升和下降),水位振荡是含水层对地震波的弹性响应,前人已有较多的研究,这里主要关注阶变型同震变化。本论文首先选择云南思茅大寨井和新疆乌鲁木齐新04井两口代表性井孔对水位同震阶变上升和下降变化特征进行研究;然后系统收集了井水位在2008年汶川Ms 8.0级和2007年印尼苏门答腊Ms 8.5前后的同震变化观测资料,分析对比了近震和远震引起的水位同震变化的异同。
对一井多震的的研究结果显示,同一口井的水位同震变化的方向不因地震的不同而改变,无论地震的方位、距离、大小和震源机制如何,上升的总是上升,下降的总是下降。水位同震变化幅度主要受震级和井震距控制,三者之间有良好的相关性。思茅井水位的同震变化总是上升,上升幅度随震级的增大而增大,随井震距的增大而减小,同时也受季节性降雨或者区域构造环境和局部应力状态变化的影响,地震波跨过红河断裂时会使井水位同震变化幅度偏离总体统计关系;分析认为渗透系数减小可能是水位同震上升的主要原因。
新04井水位的同震变化总是下降,下降幅度随震级的增大而增大,随井震距的增大而减小,同时也受本地应力状态的影响,在新04井周围发生的3个5级左右地震前后,井水位同震响应能力显著降低,可能是孔隙缩小、断层闭锁,水流受阻所致。新04井井孔穿过了断层破碎带,当受强震s波或面波能量的激发时,断层发生张合作用或蠕动,井水外泄可能是井水位同震下降变化的原因。
2008年汶川地震是地下水位水温观测网在“九五”、“十五”项目建设完成后国内发生的最为显著的地震,因此在一震多井分析时,选择以汶川地震为例。首先对前兆台网库中的井点进行了筛选。为兼顾下文对水温同震响应和同井水位水温同震变化组合类型的分析,选择观测井点时遵循以下原则:①井点要同时具有水位水温两个观测项,且在2008年5月12日地震前后仪器都正常工作;②汶川地震至少引起了水位水温中一个测项的同震变化;③水位水温具有相对稳定的动态背景。依照以上原则对前兆数据库有记录的井点进行了筛选,选出了96口井,并对其水位同震变化进行了分类统计。分类统计结果显示,2008年汶川Ms 8.0地震引起的水位同震变化以上升为主。
同时,在以上的96口井中,挑选出以汶川震中为中心1000 km范围内的32个井点,收集其对2007年印尼苏门答腊Ms 8.5级的响应,对比分析汶川近震和苏门答腊远震引起的水位水温同震变化的异同。对比结果显示,印尼远震引起水位同震上升或下降阶变的6口井,在汶川近震时其阶变方向仍然不变;在印尼远震引起水位振荡的18口井中,在汶川近震时除4口仍然振荡外,其余14口都产生了或上升或下降的同震阶变;在印尼远震无变化的8口井点中,在汶川地震时除1口无变化外,其余7口都产生了或上升或下降的同震阶变。
对比分析结果表明:无论对于远震还是近震来说,水位同震上升的比例都远大于同震下降变化的比例;相对于远震,近震引起的水位同震阶变井点数量大大增加,振荡和无变化井点数量减少;水位同震升降的方向不因地震的远近、大小、震源机制或地震方位的变化而改变,更多地受控于本地的地质构造环境和水文地质条件;地震波能量的变化不能改变水位同震变化的方向,当地震波能量足够大时,会使一些原来仅产生振荡或无同震响应的井孔的水位发生阶变。
(2)水温同震变化的研究
水温同震效应有上升和下降变化,但没有观测到振荡现象。水温同震变化与水位变化关系密切,对一井多震的研究结果表明,水位同震变化是水温同震变化的必要条件,有水温同震变化的震例都存在水位同震变化,而发生水位同震变化的震例则不一定有水温同震变化。研究结果显示,水位水温同震变化主要存在4种组合,它们分别是:水位上升水温上升型、水位下降水温上升型、水位下降水温下降型、水位振荡水温下降型。本论文首先选择云南思茅井、新疆新04井、北京左家庄井和塔院井作为上述4种组合的代表进行分析讨论;然后系统收集了从全国地下流体观测网中筛选出的96口井对2008年汶川Ms 8.0级地震的水温响应资料进行分析;最后在以汶川震中为中心1000 km范围内挑选出32个井点,收集其在2007年印尼苏门答腊Ms8.5前后的同震变化观测资料,分析对比了近震和远震引起的水温同震变化的异同及其影响因素。
水位上升水温上升型:以云南思茅井为例,水位、水温同震变化类型都为上升,变化起始时间上几乎同步,水位同震上升幅度为几十个cm的量级,最小12 cm,最大60.6 cm;水温同震上升幅度则为几十至上百10-4℃,两者之间的比例关系为几个℃/100m,量级与正常地温梯度3℃/100m基本相同。观测系统安装时的测量结果显示,水温探头附近为随深度增加的正梯度区,思茅井长时间水位水温动态对比也显示两者同向变化,结合水位水温变化幅度的比值与地温梯度量级相当的特点,分析认为水位同震上升是水温同震上升的直接原因。
水位下降水温上升型:以新疆新04井为例,水位同震变化为下降-恢复型,而水温同震变化主要表现为稳定-上升-恢复。地震引起的水位同震下降幅度最大58 mm,当水位同震阶变大于10 mm时,水温记录到同震变化,对应的水温上升幅度为几十个10-4℃,两者之间的比例关系可以达到～10℃/100m,远大于测量到的井孔地温梯度～3℃/100m。而且水温同震上升的起始时间不是与水位开始下降的时间同步,而是与水位下降后开始恢复的时间基本相当。
分析认为,新04井孔水温探头下方可能存在热的含水层,它与断层的共同作用,控制着井水位、水温的变化。当地震波激发井水沿断层外泄时,水位下降引起的水温降低与含水层补给热水产生的水温升高中和平衡,温度不发生明显变化(稳定);地震波过后井水沿断层外泄减少,而含水层中的热水继续流向井内,井水位上升,温度上升;井水位上升停止后,含水层中的热水也停止流向井内,受地温梯度的控制,井水温下降恢复到正常状态。
水位下降水温下降型:以左家庄井为例,温度探头位于井孔内有套管封闭的正梯度段内,水位水温同震变化为同步下降。记录到的水位同震变化幅度1-2 m,水温同震变化为～0.03—0.04℃,两者比例关系～2-3℃/100m。水温同震变化机理类似于思茅井水位水温同步变化,体现了正梯度状态下水位同震下降对水温同震变化的影响。
水位振荡水温下降型:以塔院井为例,井水位同震变化总是振荡,水温同震变化形态总是具有下降-上升-恢复的过程,不受地震方位和震源机制影响;水温同震变化总是发生在地震波到达和水震波开始之后。已记录到的最大水温同震变化达0.097℃。
分析认为,井孔中的水体受振荡激发而加速对流与掺混是导致塔院井水温同震下降的主要原因。水温探头一般放置在较深的部位,在正常情况下温度较高,当受到地震波的作用时,井孔中水体对流加速,深部热水体上涌,而浅部较泠水体下沉,水温探头将先观测到温度下降现象。随水震波的逐渐平息,探头附近井水温逐步恢复上升。
塔院井水温同震下降幅度随震级的增大而增大,随井震距的增大而减小。个别发生在井孔东侧的地震引起的温度同震变化小于理论计算值,可能是黄庄-高丽营断裂影响所致。
2008年汶川Ms 8.0地震引起的水温同震变化统计结果显示,在有水位或水温同震效应的96口观测井中,66口井记录到水温同震变化,上升和下降井孔各为31和35,以同震下降井孔所占比例略高;在66口井中,其中65口井都有相应的水位同震变化,只有1口井因特殊原因未观测到相应水位同震变化。同井水位水温同震变化关系的分类统计结果显示,在水位有同震升降阶变的77口井中,水位水温同方向变化(水位上升温度上升或者水位下降温度下降)的有31口井,反方向变化(水位上升温度下降或者水位下降温度上升)的有23口井,水温无变化的有23口井,以水位水温同方向变化占优势。对于水位振荡变化的18口井,水温无变化的井孔有7口,上升和下降的分别为2口和9口,水温以下降为主。
以汶川Ms 8.0地震为中心1000 km范围内的32个井点中,2007年印尼苏门答腊Ms 8.5级远震与汶川Ms 8.0近震引起的水温同震变化的比较显示,相对于远震,近震引起的水温同震阶变井点数量增加,无变化井点数量减少。当地震波能量足够大时,会使一些原来无同震响应的井孔发生水温阶变,但大多数井水温同震升降的性质都不因地震的远近、大小、震源机制或方位的变化而改变。两个发生水温同震升降性质变化的井点是由于水的自流状态改变或水位同震变化由振荡转为阶变而引起。
对一震多井和远近震例对比研究结果表明,水温同震变化是以水位的同震变化为前提条件,水温同震升降的性质不因地震的不同而改变,同震变化的幅度与震级、井震距、季节、地温梯度、探头放置位置等因素有关。
(3)不同深度井水温对比观测及水位水温动态关系的数值模拟研究
为探讨井水温动态变化的影响因素,以北京塔院井为实验场地,使用新增的一台高精度温度仪,进行详细的水温梯度测量和不同深度水温动态连续测量,把新探头获取的不同深度的水温动态数据与原有的水位水温仪(2001年7月安装)观测数据进行对比研究。同时参考唐山矿井和西昌太和井的观测资料,对比分析了井孔水温动态的影响因素。
详细的水温梯度测量显示,塔院井在距井口约105～180 m处存在一负梯度段,分析认为负梯度带的出现可能是局部段落有冷流体的存在和渗入所致。不同深度的实验测试结果表明,在井孔浅部,水温梯度变化大,其潮汐效应不明显;在深部水温正梯度段,水温水位同向变化;在负梯度段,水温水位反向变化;水温梯度越大,潮汐效应幅度越大。水温探头放置处的水温梯度的差异及其与含水层的相对位置是水温动态复杂性的重要原因。
引入了流体运动的控制方程和热传导方程。COMSOL Multiphysics是专为描述和模拟各种物理现象而开发的一个专业有限元数值分析软件,可模拟所有可用偏微分方程(PEDS)描述的物理过程。本论文选用最新版本COMSOL 4.1对塔院井水位水温同震变化进行数值模拟研究。分析中,对于井水面的升降或波动变化使用COMSOL软件中的移动网格法(ALE)来实现。
为了解释水位震荡水温下降的现象,建立了一个三维模型进行模拟分析。几何模型取井半径r=0.25 m,井深H=100 m,上下温差3℃。由于井孔及其所包含的水体构成了一个轴对称结构,研究中将三维模型简化为二维轴对称模型。在井孔下部侧壁施加周期为5 s和20 s正弦波组成的压力波动,模拟地震波对井水的作用。结果表明,在井水上表面出现了明显的水位振荡变化;在井孔中部施加压力波的上方出现水温振荡下降现象,与许多井水位振荡水温同震下降的实际观测结果一致。在实际观测中,可能由于水温探头的延迟效应和分钟值的采样率,难于观测到水温的振荡效应。
通过对井水位水温同震变化的系统实验观测和分析研究,本研究取得的主要新认识如下:
①同一口井水位同震升降的性质主要受井孔局部水文地质条件控制,不因地震方位、大小、距离和震源机制的不同而改变,上升的井总是上升,下降的井总是下降。
②单井水位升降幅度随震级的增大而增大,随井震距的增大而减小,三者之间有良好的相关性;地震波跨越断层或者说来自于某个方向的地震波、季节性的降水会使同震变化的幅度偏离总体统计关系;井孔附近孕震可能使同震响应能力发生改变。
③同一口井的水位同震变化是水温同震变化的必要条件,水温同震响应总是出现在地震波到达和水位同震变化开始之后;水温同震变化的幅度受到震级、井震距、季节、地温梯度、探头放置位置等因素的影响。
④无论对于远震还是近震来说,同一地震引起的水位同震变化以上升为主,可能是渗透率的降低所致;同震水位下降的井多穿过了断层破碎带,地震波激发使断层发生张合作用或蠕动,井水外泄可能是井水位同震下降的原因。
⑤相对于远震,近震引起的水位水温同震阶变井点数量大大增加,振荡和无变化井点数量减少;当地震波能量足够大时,会使一些原来仅产生振荡或无同震响应的井孔的水位水温发生阶变,但不能使水位水温阶变的性质发生变化。个别发生水温同震升降性质变化的井点是由于水的自流状态改变或水位同震变化由振荡转为阶变而引起。
⑥水温探头放置处的水温梯度差异及其与含水层的相对位置是水温动态复杂性的重要原因。在深部正梯度段,水温水位同向变化;在负梯度段,水温水位反向变化。水温梯度越大,潮汐效应和同震变化幅度越大。塔院井负梯度带的出现可能是局部段落有冷流体的存在和渗入所致。
⑦同井中水位水温同震阶变的关系以两者同方向变化(水位上升温度上升或者水位下降温度下降)比反向变化(水位上升温度下降或者水位下降温度上升)占优势;同震水位振荡型井孔,其深部水温同震变化以下降-恢复为主,其原因是井孔中的水体受振荡激发而加速对流与掺混所致。
由于时间紧、工作量大、数值模拟软件晚到、实验井(塔院井)改造等原因,论文仍然存在许多问题,例如每个井孔有自己的独特的特征,要想对一些观测现象深入分析研究,需要有井孔的水文地质、井孔结构、温度梯度等有详细的资料,而很多井孔的资料都很不完整;一井多震和一震多井的研究实例有限;在数值模拟分析中,流体运动的复杂性,有些现象很难模拟等问题,都有待于将来进一步深入研究。
Groundwater is an important component of the earth system, of which the movement and change are closely associated with human living environments. Earthquakes are sudden and frequent events of on the earth. The relationship between earthquakes and groundwater change has been the subject that scientists make great efforts to study for tens of years. Coseismic effects of well-water level and temperature caused by earthquakes is currently the outstanding topic receiving much attention. In this thesis, focused on this problem that, an attempt is made to reveal the distribution rule in space and time, influence factors, mutual relations and formation mechanisms of coseismic variations of groundwater level and temperature. It is of great significance in both theory and practice to clarifying relationship between earthquakes and groundwater variation, studies of crustal movement, reduction of secondary disasters, tracking following earthquakes, and earthquake precursor retrospection. The observation network of groundwater level and temperature in China has experienced prelimilary constructions in 70’s and 80’s, developments in ninth 5-year plan and tenth 5-year plan, and the numbers of observation wells have reached 420 and 277 respectively. Especially, the sample rate of digital observation apparatus, has reached one value per minute, is rapid, continuous, and convenient, which lays solid data foundation for a detailed study on the coseismic effects of well-water level and temperature.
Fist, this thesis makes a systemic investigation is to the current research status on coseismic variations of groundwater level and temperature. It finds that there are many existing problems. For example, most work is of qualitative analysis, lacking further quantitative analysis. There is no report on systemic research on the relationship between groundwater level and temperature in coseismic variations. Researches on the influence factors on water temperature are rare. The understanding on the mechanism of coseismic variations of groundwater level and temperature is still controversial.
Targeting the problems above, some representative wells are selected for a detail analysis of the continuous observation data of several years, and for the study of the coseismic variation rules of groundwater level and temperature under one well-multi earthquakes conditions. The observation wells in the database of national precursor network are selected according to specific principles. The observation data of groundwater level and temperature of the selected wells are then colleclected systemically, which are around the time when the Wenchuan Ms 8.0 (2008) event occurred, for studying the coseismic variation of groundwater level and temperature in the case of one earthquake-multi wells, and the combination type of groundwater level and temperature variations in the same well. Those wells within the range of 1000 km with the center of Wenchuan epicenter are selected to collect their observation data around Sumatra Ms8.5, Indonesia (2007) event. Next this thesis analyzes and compares the differences and similarities in coseismic variations of groundwater level and temperature caused by nearby and remote earthquakes and their influence factors. Finally taking the Tayuan well in Beijing as the test site, apply a comparison observation on groundwater level, dynamic variations of groundwater temperature at different depths, discuss the influence factors for variations of groundwater temperature. And models are constructed to perform numerical modeling and explore the mechanism of coseismic variations of groundwater level and temperature.
The main work in the thesis can be concluded as following.
1 Coseismic variations of groundwater level
Coseismic variations of groundwater level can be divided to oscillation type and step type (including ascent and descent). The groundwater oscillation is the elastic response of aquifer to earthquake waves, which has been well studied. So this work concerns the coseismic variations of the step type. Two representative wells, the Dazhai well in Simao, Yunan Province and Xin 04 well in Urumchi, Xinjiang are selected firstly for the study of ascending and descending characteristics in coseismic variations of groundwater level. Then, the coseismic variation data of groundwater levels before and after the Wenchuan Ms 8.0 (2008) and Sumatra Ms 8.5, Indonesia (2007) events are collected for analysis and comparison of the differences and similarities in coseismic variations of groundwater level and temperature caused by nearby and remote earthquakes and their influence factors.
The research results of the case one well-multi earthquakes indicate that the coseismic variation directions of groundwater level at a same well do not change with different earthquakes. It keeps ascent or descent for any orientations, distances, magnitudes and mechanisms of the earthquakes. The amplitudes of coseismic variations of groundwater level depend upon the seismic magnitudes and well-earthquake distances, of which the three are well correlated.
The coseismic variation of groundwater level in the Simao well is always ascending with growing amplitude with increasing magnitudes of earthquakes and declining amplitude with increasing well-earthquake distances. At the same time, it is also affected by seasonal rainfalls, regional tectonic settings and change of local stress state. The magnitude of coseismic variations of groundwater level will deviate from the general statistic relation when earthquake waves cross the Red River fault or a shock takes place near the well. Analysis suggests that the decrease of permeability may be the primary reason for coseismic ascending of groundwater level.
The coseismic variation of groundwater level in the Xin 04 well is always descending, with growing amplitude with increasing magnitudes of earthquakes and decreasing amplitude with increasing well-earthquake distances. Meanwhile it is also affected by local stress status. When three earthquakes with magnitudes around Ms. 5.0 occurred near the Xin 04 well, the coseismic response ability of groundwater level decreased notably, which may be due to pore shrinking, fault locking, and stream blocking. The Xin 04 well penetrates a fault fractured zone on which will open-close action or creep would occur when it is triggered by S waves or surface waves of strong shocks, where the outflow of well water may be the reason for coseismic descending of groundwater level.
Wenchuan Ms 8.0 earthquake (2008) was the most remarkable one that occurred in China when the projects of groundwater level and temperature in ninth 5-year plan and tenth 5-year plan finished. So Wenchuan earthquake is selected to analyze the effect of one earthquake-multi wells. The wells recorded in the precursor networks are selected firstly. With giving attention to the combination analysis of coseismic effects of groundwater temperature and coseismic variations of groundwater level and temperature in same well, the observation wells are selected following such principles:①the well station must have the two observation items as groundwater level and temperature, and work normally before, during and after the earthquake on May 12, 2008.②the Wenchuan earthquake has caused coseismic variations of at least one item of groundwater level and temperature.③the dynamic background of groundwater level and temperature is relatively stable. According to these principles, 96 wells are selected from the recorded wells in the database of precursor networks, and the coseismic variations of groundwater level in these wells are classified, the statistics shows that the coseismic variations of groundwater level caused by the Wenchuan Ms 8.0 earthquake (2008) are dominated by water ascent.
Among the above 96 wells, 32 wells in the area of 1000 km centered by the epicenter of the Wenchuan event are selected to collect their responses to Sumatra Ms 8.5, Indonesia (2007), in order to have further comparison of the differences and similarities in coseismic variations of groundwater level and temperature caused by nearby and remote earthquakes. The result shows that 6 wells which exhibited ascent or descent related with the Sumatra Ms 8.5, Indonesia, 2007 kept same change when the Wenchuan event occurred. And among the 18 wells with oscillation variations of groundwater level associated with the Sumatra event of 2007, except 4 wells had oscillations related with the Wenchauan shock, the other 14 wells showed ascent or descent changes when the Wenchuan event took place. In the rest 8 wells which looked like quiet re during the remote Indonesia remote earthquake, only one well kept invariant while the other 7 wells showed coseismic ascent or descent.
The comparative analysis suggests that the proportion of coseismic ascent of groundwater level is far greater than that of coseismic descent for either remote or nearby earthquakes. The number of wells with coseismic stepwise changes of underground water caused by nearby earthquakes is larger, while that of wells with oscillation or without change is small with respect to that by remote events. The directions of coseismic ascent or descent of groundwater level do not depend on distances, magnitudes, mechanisms or azimuths of earthquakes, instead are controlled by local geological settings and hydrologic conditions. If the energy of seismic waves is strong enough, some wells that only have oscillations or no coseismic responses originally will show stepwise variations of groundwater level.
2 Coseismic variations of groundwater temperature
There is ascent or descent in coseismic responses of groundwater temperature, but no oscillation is observed. Such changes are closely related with coseismic variations of groundwater level. The case study of one well-multi earthquakes suggests that coseismic variation of groundwater level seems to be the necessary condition for coseismic variation of groundwater temperature. It means that the earthquakes that have coseismic variations of groundwater temperature also show coseismic variations of groundwater level, while the events that have coseismic variations of groundwater level are not always accompanied by coseismic variations of groundwater temperature. In the research, the coseismic variations of groundwater level and temperature can be divided into four main types as ascending groundwater level and ascending groundwater temperature, descending groundwater level and ascending groundwater temperature, descending groundwater level and descending groundwater temperature, and oscillating groundwater level and descending groundwater temperature. In the thesis, the Simao well inYunnan, Xin 04 well in Xinjiang, Zuojiazhuang well in Beijing, and Tayuan well in Beijing are selected as the representing four cases for further analysis and discussion. Then the response data of groundwater temperatures before, during and after the Wenchuan Ms 8.0 (2008) are collected systemically from the 96 wells which are selected from the national underground fluid observation networks. 32 wells in the area of 1000 km centered by the epicenter of the Wenchuan event are selected to collect their responses to Sumatra Ms 8.5, Indonesia (2007). The differences and similarities together with their influence factors in coseismic variations of groundwater temperature caused by nearby and remote earthquakes are analyzed and compared.
Type of ascending groundwater level and ascending groundwater temperature
Taking the Simao well in Yunnan as an example, the coseismic variations of both groundwater level and groundwater temperature are of ascent, which start nearly at the same time. The amplitude of coseismic ascent of groundwater level has a magnitude of tens of centimeters with minimum value as 12 cm and maximum value as 60.6 cm. The amplitude of coseismic ascent of groundwater temperature ranges from tens to hundred 10-4℃. The ratio between them is several℃/100m which is roughly the same as the normal geothermal gradient of 2-3℃/ 100m. The measurements when the observation system was installed show a positive gradient area where water temperature increases with depth near the groundwater temperature sensor. The long-term dynamic comparison of groundwater level and temperature in the Simao well also shows changes in a same direction. Considering that the ratio between the scopes of groundwater level and temperature is equivalent to the order magnitude of groundwater temperature gradient, it is indicated that coseismic ascent of groundwater level is the direct reason for coseismic ascent of groundwater temperature.
Type of descending groundwater level and ascending groundwater temperature
At the Xin 04 well in Xinjiang, the coseismic variations of groundwater level are of descending-resuming type, while that of groundwater temperature mainly displays as stability-ascent-recover. The start time of coseismic ascent of groundwater temperature is comparable to the time when groundwater level changes from descending to resuming. The maximum amplitude of coseismic descending of groundwater level is 58 mm. When the coseismic step of groundwater level is greater than 10 mm, the coseismic variations of groundwater temperature also take place with corresponding amplitude of groundwater temperature ascending as tens of 10-4℃. The ratio between them can reach 10℃/100m which is greater than normal geothermal gradient of 3℃/100m. Moreover, the starting time of coseismic ascending of groundwater temperature is not synchronous with the time when groundwater level begins descending, but basically equivalent to the time when groundwater level begins recovering after descending.
Further analysis indicates hot aquifer may exist below the groundwater temperature sensor in the Xin 04 well. This hot aquifer will act together with the fault and control the variations of groundwater level and temperature in the well. When the well water is triggered by seismic waves and leaks out, the temperature ascent caused by groundwater level descent will counteract the groundwater temperature ascending caused by hot water supplied by aquifers and come to a balance, so the temperature will not change evidently. After the seismic waves are over, the leakage of well water decreases, while the hot water in aquifers will flow into the well continuously, which results in ascent of well water level and temperature. When the well water level stops ascending, the hot water in aquifers will also cease to flow into the well, and the well water temperature resumes to normal status under the control of the geothermal gradient.
Type of descending groundwater level and descending groundwater temperature
In the Zuojiazhuang well in Beijing, the temperature sensor is located at the positive gradient section sealed by well casing. The coseismic variations of groundwater level and temperature are of synchronous descending. The recorded amplitude of coseismic variation of groundwater level is in the range of 1-2 m; the recorded coseismic variation of groundwater temperature is in the range of 0.03-0.04℃; the ratio between them is in the range of 2-3℃/100 m. The mechanism of coseismic variation of groundwater temperature is similar to that observed at the Simao well in Yunnan Province, which suggests the influence of the coseismic descending of groundwater level on that of groundwater temperature in a positive gradient status.
Type of oscillating groundwater level and descending groundwater temperature
In the case of the Tayuan well in Beijing, the coseismic change is always oscillating, while that of groundwater temperature is usually of a descending-ascending-resuming process and independent of azimuths and mechanisms of earthquakes. The coseismic variations of groundwater temperature often occur at the time when seismic waves arrive and after the water level seismic wave starts. The recorded maximum coseismic variation of groundwater temperature can reach 0.097℃.
Further analysis suggests that the water body in the well will be excited by oscillation, resulting in accelerated convection and mixture, which is the main cause for coseismic descending of groundwater temperature in the Tayuan well. The water temperature sensor is commonly placed at a relatively deep section of the well with relatively high temperature in normal conditions. When the seismic waves come, the water body in the well will accelerate convection and deep hot water ascends, while the shallow cold water falls down, so the water temperature sensor will observe temperature descending firstly. With the water level seismic waves calming down, the well water temperature around the sensor will resume ascending gradually.
The amplitude of coseismic descent of groundwater temperature increases with the growing magnitudes of earthquakes and decreases with the increasing well-earthquake distance. The coseismic variations of groundwater temperature caused by individual earthquakes east of the well are smaller than theoretic values, which may be due to the influence of the Huangzhuang-Gaoliyin fault there.
Classification statistics of coseismic variations of groundwater temperature caused by Wenchuan Ms 8.0 earthquake (2008) is made. Among 96 observation wells that have coseismic effects of groundwater level and temperature, there are 66 wells that exhibit coseismic variations of groundwater temperature, in the 66 wells, 65 wells show corresponding coseismic variations of groundwater level, the well that no corresponding coseismic variations of groundwater level were observed may be caused by some special reasons. In the classification statistics of coseismic variations of groundwater level and temperature in same well, there are 77 wells that show coseismic ascending or descending steps of groundwater level, of which 31 wells show changes of groundwater level and temperature in same directions; 23 wells with changes in opposite directions; and 23 wells without variation of groundwater temperature. The wells with changes of groundwater level and temperature in same directions are predominant. Among the 18 wells with oscillation of groundwater level, 7 wells show no changes in groundwater temperature; 2 wells show temperature ascending; 9 wells show temperature descending; and descent dominates the groundwater temperature.
Among the 32 wells with a range of 1000 km centered by the epicenter of the Wenchuan Ms 8.0 event in 2008, the coseismic variations of groundwater temperature by this event and the Sumatra Ms 8.5 of 2007 are compared. It indicates that there are more fewer wells without such changes, with respect to the case of remote earthquakes. If the energy of seismic wave is strong enough, some wells without coseismic responses originally will show water temperature steps. Of course, the properties of most coseismic ascending or descending of groundwater temperature will not change with the variations in distances, magnitudes, mechanisms or azimuths of the earthquakes. The two wells that occurred character change in coseismic ascending or descending of groundwater temperature were caused by artesian flowing change or the coseismic variation of groundwater level changing from oscillating to step.
Case comparison of one earthquake-multi wells and nearby and remote earthquakes also indicate that coseismic variations of groundwater temperature occur on the premise of coseismic variations of groundwater level. The properties of coseismic ascending and descending of groundwater temperature will not change with different earthquakes. The amplitude of coseismic variation is related with such factors as magnitudes, well-earthquake distance, season, geothermal gradient, and location of temperature sensor.
3 Test of well water temperature sections and numerical modeling of dynamic relation between groundwater level and temperature
The Tayuan well in Beijing is selected as the test site for discussing influence factors of well-water temperature variation. With a new high-precision temperature meter, detail measurement of water temperature gradients and dynamic continuous measurements of groundwater temperature at different depths have been carried out. The dynamic data of groundwater temperature at different depths in the Tayuan well are obtained with new sensor that allow a comparative study with data obtained by intrinsic groundwater level and temperature meter (which was installed in July, 2001). At the same time, the observed data from the Tangshan mine well and Taihe well in Xichang are also referenced to analyze the influences of groundwater gradient on tide and postseismic effects.
From the detailed measurements of water temperature gradient, it exists a negative gradient section at 105-180 m away from the mouth of the well. the occurrence of negative gradient section may be caused by cold fluid existing at or flowing into local sections The test results at different depths indicate that water temperature gradients are large and the tidal effects are not evident at the shallow part of the wells. The groundwater level and temperature show change in same directions in the positive gradient section of deep groundwater temperature; and in opposite directions in the negative gradient section. The larger the groundwater gradient is, the bigger the amplitudes of tidal effects become. Differences of water temperature gradients around the sensor and its relative position to aquifer are the important factors that cause complex dynamic variations of water temperature.
Control equations of fluid motion and heat exchange equations are introduced in this study. COMSOL Multiphysics is a professional FE numerical analysis software that is developed for describing and modeling varies of physical phenomena; it can model those physical procedures expressed with partial difference equations. Numerical modeling of coseismic variations of groundwater level and temperature is carried out with COMSOL 4.1 software. The ascending and descending or oscillating of groundwater level are realized with Arbitrary Lagrangian-Eulerian (ALE) method in COMSOL software.
A three-dimensional model is constructed to model and explain the phenomenon of water level oscillating and water temperature descending. The geometry of the model includes 0.25 m of well radius (r) and 100 m of well depth (H). The temperature difference between the top and bottom of the well is 3℃. Because the borehole and the water comprised in it construct an axial symmetry structure, the three-dimensional model is simplified to a two-dimensional axial symmetry model. Compressive waves composed of sine waves of 5 s and 20 s are exerted at the side wall of the bottom section of the well to simulate the effect of seismic waves on well groundwater. The results show that obvious groundwater level oscillation appears on the upper surface of well water; the groundwater temperature oscillates downward in the middle part of the well where compressive waves are exerted, which is coincided with many observations of groundwater level oscillations and coseismic descending of groundwater temperature. Because of the delayed effect of water temperature sensor and sample rate of one value per minute, it is difficult to observe the oscillation effects of groundwater temperature.
By systemic test observations and analysis of coseismic variations of groundwater level and temperature, this thesis has obtained some insights presented below.
(1) The coseismic ascending or descending of groundwater level and temperature at the same well is mainly controlled by local geohydrologic conditions, and will not change with the orientations, magnitudes, distances, and mechanisms of earthquakes. Those wells with groundwater ascending are always in an ascending state, while those with groundwater descending wells are always in a descending regime.
(2) The ascending or descending amplitude of groundwater level in a single well increases with growing magnitudes of earthquakes and decreases with increasing well-earthquake distances. There is a good correlation among them. Moreover, such factors as seismic waves crossing faults or coming from a special direction, and seasonal rainfall will cause the amplitude of coseismic variation deviate from general statistics relation. Earthquakes preparation near the well may probably reduce coseismic response capability.
(3) Coseismic variation of groundwater level is the necessary condition for coseismic variation of groundwater temperature at the same well. The coseismic response of groundwater temperature usually appears at the time when seismic waves arrive and coseismic variations of groundwater level begin. The amplitude of coseismic variation of groundwater temperature is affected by such factors as magnitude, well-earthquake distance, seasonal rainfall, geothermal gradient, locations of water temperature sensor.
(4) The coseismic variations of groundwater level caused by a same earthquake are mainly ascending, which is true for either nearby or remote earthquakes and may be caused by permeability decrease. Most wells with coseismic descending of groundwater level may penetrate fault fracture zones. The seismic waves will cause the faults to open and close or creep. The leakage of well water may cause coseismic descending of groundwater level.
(5) Compared with remote earthquakes, the number of wells with coseismic steps of groundwater level and temperature caused by nearby earthquakes is much larger, while the number of wells with oscillations or without changes is less. If the energy of seismic waves is strong enough, the groundwater level and temperature of some wells that only oscillate or show no coseismic responses originally will show steps, but the seismic waves cannot change the step character of groundwater level and temperature. The character change of coseismic ascending or descending of groundwater temperature in individual well was caused by variation in water artesian flowing or coseismic variation of groundwater level changing from oscillation to step.
(6) Differences in groundwater temperature gradients around the water temperature sensor and its relative position to aquifer are important factors causing complexity in dynamic variation of groundwater temperature. The groundwater level and temperature show syn-directiion change at deep sections of positive gradients but reverse- direction change at sections of negative gradients. The larger the groundwater gradient is, the bigger the amplitudes of tidal effects and coseismic variations become. The occurrence of negative gradient section in Tayuan well may be caused by cold fluid flowing into local sections of the well.
(7) In the same well, the relation of coseismic stop of groundwater level and temperature shows that changing in same direction is dominant to changeing in opposite directions. For wells with coseismic oscillations of groundwater level, the deep groundwater temperature variations are mainly of coseismic descending-resuming. The reason is that the water body in the well is triggered by oscillations so that water convection and mixture are accelerated
Because some objective factors such as limited time available, large amount of work, numerical modeling software that can be used only lately, and rebuilding of the test well (Tayuan well) , there are many problems that exist in the article or remain unsolved. For example, each well has its special characters; the detailed data of geohydrologic condition, structure of borehole, geothermal grident, etc. are needed to have thorough analysis of observations, but these data of many wells are incomplete. There are limited examples for the case studies of one well-multi earthquakes and one earthquake-multi wells. The movement of fluid is complex; some phenomena are hard to be modeled. These problems will be studied further in the future.
引文
Beresnev I.A. , Johnson P.A. 1994 . Elastic-wave stimulation of oil production: A review of methods and results[J].Geophysics,59(6):1 000-1 017
Brodsky E. E.,Roeloffs E. A.,Woodcock D,et al.2003.A mechanism for sustained groundwater pressure changes induced by distant earthquakes, J. Geophys. Res.,108(B8),2390,doi:10.1029/2002JB002321.
Carrigan C.R.,G.C.P. King, G. E. Barr,et al.1991. Potential for water-table excursions induced by seismic events at Yucca Mountain, Nevada[J].Geology,19(12): 1 157-1 160
David R Montgomery,Michael Manga.2003.Stream?ow and water well responses to earthquakes[J].Science,300(27):2047～2049
Fujio Akita, Norio Matsumoto. 2004. Hydrological responses induced by the Tokachi-oki earthquake in 2003 at hot spring wells in Hokkaido, Japan[J].Geophys Res Lett,31,L16603,doi:10.1029/2004GL020433
G.DE Vahl Davis,I.P.Jones.1983.Natural convection in a square cavity a comparison exercise[J].International Journal for Numerical Methods in Fluids,3:227-248
Hudnut K W ,Seeber L,Pacheco J.1989.Cross-fault triggering in the Novem ber 1987 Superstition Hills earthquake sequence,southern Califomia[J].Geophys Res Lett,16:199-202
Isao Shimizu, Hideaki Osawa, Toshihiro Seo,et al. Earthquake-related ground motion and groundwater pressure change at the Kamaishi Mine[J].Engineering Geology,43,107-118
JauméS C,Sykes L R.1992.Change in the state of stress on the southern San Andr eas fault resulting from the California earthquake sequence of April to June 1992[J].Science,258:l 325～1 328
King C Y,Azuma S,Igarashi G,et al.1999.Earthquake-related water-level changes at 16 closely clustered wells in Tono,central Japan[J].J Geophys Res,104(B6):13 073～13 082
Kitagawa Y, Koizumi N,Tsuskuda T.1996. Comparison of post-seismic groundwater temperature changes with earthquake-induced volumetric strain release: Yudani hot spring,Japan[J].Geophys Res Lett, 23(22):3147～3150
Koizumi N, W.C. Lai,Y Kitagawa et al.2004.Commnent on“Coseismic hydrological changes associated with dislocation of the September 21,1999 Chichi earthquake,Taiwan”by Min Lee et al[J]. Geophys. Res. Let,31,L13603,doi:10.1029/2004GL019897
Lee Min, Liu Tsung-Kwei, Ma Kuo-Fong.2002. Coseismic hydrological changes associated with dislocation of the September 21, 1999 Chichi earthquake, Taiwan[J]. Geophysical Research Letter , 29(17), 1824, doi:10.1029/2002GL015116
Li VC,Seale SH,Cao T.1987.Post seismic stress and pore pressure readjustment and aftershock distributions[J].Tectonophysics,144:37～54
Lomnitz C.1996.Search of a worldwide catalog for earthquakes triggered at intermediate distances[J].Bull Seism Soc Amer,88;293～298
Matsumoto N.1992.Regression analysis for anomalous changes of ground water level due to eathquake[J], Geophys Res Lett,19(12):1 193～1 196
Michael Manga, Emily E. Brodsky, Michael Boone.2003.Response of streamflow to multiple earthquakes. Geophys Res Lett ,30(5),1214,doi:10.1029/2002GL016618
Mogi k.,H Mochizuki,Y Kurokawa.1989. Temperature changes in an artesian spring at Usami in the Izu Peninsula(Japan) and their relation to earthquake[J].Tectonophysics,159,95-108
Naoji Koizumi,Wen-Chi Lai,Yuichi Kitagawa,et al.2004.Comment on “Coseismic hydrological changes associated with dislocation of the September 21,1999 ChiChi earthquake,Taiwan”by Min lee et al.
Geophysical research letter,VOL.31,L13603,doi:10.1029/2004GL019897 Nur A,Booker J R.1972.Aftershocks caused by pore fuid flow[J]. Science,175:885～887
Noir J,Jacques E,Békri S,et a1.1997.Fluid flow triggered migration of events in the 1989 Dobi earthquake sequence of central Afar[J].Geophys Res Let,24: 2 335～2 338
Parotidis M., S. A. Shapiro, E. Rothert.2005.Evidence for triggering of the Vogtland swarms 2000 by pore pressure diffusion, J. Geophys. Res., 110, B05S10, doi:10.1029/2004JB003267.
Quilty E G,Roeloffs E A.1997.Water level changes in response to theDecember 20,1994 M 4.7 earthquake near Parkfield,California[J].Bull Seism Soc Amer,87:310～317
Rexin E E,Oliver J,Prentiss D.1962.Seismically-Induced Fluctuation of the water level in the Nunn-Bush Well in Milwaukee[J]. Bull Seism Soc Amer,52(1):17-25
Roeloffs E A.1998. Persistent water level changes in a well near Parkfield California,due to local and distant earthquakes[J].J Geophys Res,103(B1): 869～889
Roeloffs E A,Sneed M,Galloway D L,et al.2003.Water-level changes induced by local and distant earthquakes at Long Valley caldera,California[J].J Volcanol Geotherm Res,127:269～303
Timothy L Masterlark,Herbert F wang,Lung S Chan,et al.1999.Coseismic fluid-pressure response estimated from Prediction-error filtering of tidal-band loading. Bull Seism Soc Amer,89(6):1439-1446
Wakita H.1975.Water wells as possible indicators of tectonic strain[J].Science,189:553～555
Wang C.Y. , L. H. Cheng, C. V. chin, et al.2001.Coseismic hydrological changes associated with dislocation of the September 21,1999 Chichi earthquake,Taiwan[J]. Geology ,29,831-834
Wang Chi-yuen,Dreger D S,Wang Chung-Ho.2003.Field relations among coseismic ground motion, water level change and liquefaction for the 1999 Chi-Chi (Mw = 7.5) earthquake, Taiwan[J].Geophysical Research Letter, 30(17), 1890, doi:10.1029/2003GL017601.
Wang Chi-yuen,Chia Yeeping.2008.Mechanism of water level changes during earthquakes: Near fieldversus intermediate field. Geophysical Research Letter, 35, L12402, doi:10.1029/2008GL034227
William S.D. , Wilcock . 2004 . Physical response of mid-ocean ridge hydrothermal system to local earthquakes[J]. Geochemistry Geophysics Geosystems,VOL.5,Q11009,doi:10.1029/2004GC000701
北京市地质局水文地质工程地质大队.1980.平原区第四系含水层分布图.见:北京市水文地质图集[M].上海:中华印刷厂,25
边立恩,贺振华,黄德济.2008.饱含流体介质的地震波场特征及频率分布[J].岩性油气藏,20(3):74～78
蔡义汉.2004.地热直接利用[M].天津:天津大学出版社,20
车用太,鱼金子,刘春国.1996.我国地震地下水温度动态观测与研究[J].水文地质工程地质,4:34～37
车用太,刘喜兰,姚宝树,等.2003.首都圈地区井水温度的动态类型及其成因分析[J].地震地质,25(3):404～419
车用太,鱼金子.2004.塔院井水温的大幅度奇异变化[A].地下流体典型异常的调查与研究[C].北京:气象出版社,414～418
陈建国,陈海昕,章光华,等.2006.轴向浮力驱动下的旋转圆筒热对流稳定性[J].36(3):263～273
崔勇,黄辅琼,李新勇.2006.新04井(泉)流体测项长程趋势变化分析[J].内陆地震,20(1):81～87
丁仁杰.1990.井孔水位记震能力与地震关系探讨[A].见:国家地震局监测司编.地震预报方法实用化研究文集.水位水化专辑[C].北京:地震出版社,186～193
付子忠.1989.地热动态观测与地热前兆[A].见:国家地震局地壳应力所编.地壳构造与地壳应力文集[C].北京:地震出版社
付虹,刘丽芳,王世芹,等.2002.地方震及近震地下水同震震后效应研究[J].地震,22(4):55～66
付虹,刘强,邬成栋,等.2007.印尼8.7、8.5级远震对云南地区小震的激发和地下流体异常响应初探[J].地震研究,30(1):8～13
黄辅琼,迟恭财,徐桂明,等.2000.大陆地下流体对台湾南投7.6级地震的响应研究[J].地震,20(增刊):119～125
高小其,许秋龙,兰陵,等.2002.新04号井、新10号井水位和新10号泉流量地震前兆特征的对比分析[J].内陆地震,16(4):346～354
谷元珠,车用太,鱼金子.等.2003.塔院井水温微动态研究[J].地震,23(1):102～108
顾申宜,李志雄,张慧.2010.海南地区5口井水位对汶川地震的同震响应及其频谱分析[J].地震研究,33(1):35～42
虢顺民,汪洋,计凤桔.1999.云南思茅-普洱地区中强震群发生的构造机制[J].地震研究,22(2):105～115
胡敦宽,李淑芳,刘永铭.1996.白家疃观测井的地热干扰排除及机理探讨[J]地震学报,18(1):128～131
李有才.1995.四川深井水位同震阶变异常与地震预报[J].四川地震,3:32～35
林学钰,廖资生,赵勇胜,等.2005.现代水文地质学[M].北京:地质出版社,19～24
林松,敬成君,刘旭良.2006.热对流稳定性机理研究新方法初探[J].热科学与技术,5(3):236～241
刘全,水鸿寿,张晓轶.2002.数值模拟界面流方法进展[J].力学进展,32(2):259～274
刘耀炜,杨选辉,刘用铭.等.2005.地下流体苏门答腊8.7级地震的响应特征[A].见:中国地震局监测预报司编:2004年印度尼西亚苏门答腊8.7大地震及其对中国大陆地区的影响[C].北京:地震出版社,131～258
戚筱俊.1997.工程地质及水文地质[M].北京:中国水利水电出版社.
石耀霖,曹建玲,马丽,等.2007.唐山井水温的同震变化及其物理解释[J].地震学报,29(3):265～273
舒优良,张世民.2003.周至深井水震波数字化记录与地震波记录的对比研究地震地磁观测与研究[J].24(5):26～31
陶文铨.2009.数值传热学[M].西安:西安交通大学出版社
唐小勇,王新刚,黄建明,等.2005.乌鲁木齐台2004年印尼苏门答腊特大地震钻孔应变观测报告[J].地壳构造与地壳应力,2:14～15
田竹君,车用太,朱清钟,等.1983.塔院井孔水位动态实验站建设[M].北京:国家地震局地质所,1～10
万永革,吴忠良,周公威,等.2002.地震应力触发研究[J].地震学报,24(5):533～551
万登堡.1992.井孔水位记震能力变化与近大地震关系研究[J].地震研究,15(4):381～391
汪成民,车用太,万迪堃，等．1988.地下水微动态研究[M].北京:地震出版社,81～88
汪成民,李宣瑚,王铁城,等.1990.中国地震地下水动态观测网[M].北京:地震出版社,81～88
汪成民.2001.后效预报[A].见:地震预报引论.地下水微动态与地震预报[C]北京:地震出版社:165
王道.2000.天山地震活动区地下水同震、震后效应的研究[J].内陆地震,14(3):252～259
王勇,张国红,代建华,等.2000.昌平地震台地温前兆特征分析[J].西北地震学报,22(3):236～240
汪一鹏,沈军.2000.天山北麓活动构造基本特征[J].新疆地质,18(3):203～210
王大纯,张人权,史毅虹,等.1994.水文地质学基础[M].北京:地质出版社
伍致中,郭富贤.1991.再论博格达推覆构造与油气[J].新疆地质,9(1):40～49
徐锡伟,闻学泽,叶建青,等.2008.汶川Ms8.0地震地表破裂带及其发震构造[J] .地震地质,30(3) :597—629
徐江荣,李超.2008.湍流特征频率的模式理论计算[J].杭州电子科技大学学报,28(4):73～76
杨期隆,辛奎德. 1991.西藏羊八井地热田简介[J].地质论评,37(3):283～287
杨竹转,邓志辉,赵云旭,等.2005.云南思茅大寨井水位同震阶变的初步研究[J].地震学报,27(5):569～574
杨竹转,邓志辉,陶京玲,等.2007.北京塔院井数字化观测水温的同震效应研究[J].地震学报,29(2):203～213
徐世光,郭远生.2009.地热学基础[M].北京:科学出版社,20～22
鱼金子,车用太,王爱英,等.1995.大同-阳高地震前静乐井水位记震能力的异常及其意义[J].山西地震,1:50～53
鱼金子,车用太,刘伍洲.1997.井水温度微动态形成的水动力学机制研究[J].地震,17(4):389～396
张素欣,杨卫东、张子广.2007.唐山矿井模拟与数字水位的记震能力对比分析[J].西北地震学报,29(2):170～173
张昭栋、迟镇乐、陈会民,等.2000.井水位的振荡与地震波[J].地震研究,23(4):418—425
张昭栋,赵淑平,董传富,等.1994.井水位阶变与含水层所受体应力之间的定量关系[J].地球物理学报,37(1):222～229
张子广,张素欣,李薇,等.2007.昌黎井水温潮汐形成机理分析[J].地震,27(3):34～40
张子广,万迪堃，董守玉．1998.水震波与地震面波的对比研究及其应用[J].地震,18(4):399～404
庄礼贤,尹协远,马晖扬.1997.流体力学[M].安徽:中国科学技术大学出版社
邹剑峰,郜冶.2007.含加热圆管方腔内自然对流的数值研究[J].热科学与技术,6(4):326～330
庄光国.1999.台湾海峡7.3级地震水震波特征研究[J].地震,19(2),199～203
Brodsky E. E.,Roeloffs E. A.,Woodcock D,et al.2003.A mechanism for sustained groundwater pressure changes induced by distant earthquakes, J. Geophys. Res.,108(B8),2390,doi:10.1029/2002JB002321.
Carrigan C.R.,G.C.P. King, G. E. Barr,et al.1991. Potential for water-table excursions induced by seismic events at Yucca Mountain, Nevada[J].Geology,19(12): 1 157-1 160
David R Montgomery,Michael Manga.2003.Stream?ow and water well responses to earthquakes[J].Science,300(27):2047～2049
Fujio Akita, Norio Matsumoto. 2004. Hydrological responses induced by the Tokachi-oki earthquake in 2003 at hot spring wells in Hokkaido, Japan[J].Geophys Res Lett,31,L16603,doi:10.1029/2004GL020433
G.DE Vahl Davis,I.P.Jones.1983.Natural convection in a square cavity a comparison exercise[J].International Journal for Numerical Methods in Fluids,3:227-248
Hudnut K W ,Seeber L,Pacheco J.1989.Cross-fault triggering in the Novem ber 1987 Superstition Hills earthquake sequence,southern Califomia[J].Geophys Res Lett,16:199-202
Isao Shimizu, Hideaki Osawa, Toshihiro Seo,et al. Earthquake-related ground motion and groundwater pressure change at the Kamaishi Mine[J].Engineering Geology,43,107-118
JauméS C,Sykes L R.1992.Change in the state of stress on the southern San Andr eas fault resulting from the California earthquake sequence of April to June 1992[J].Science,258:l 325～1 328
King C Y,Azuma S,Igarashi G,et al.1999.Earthquake-related water-level changes at 16 closely clustered wells in Tono,central Japan[J].J Geophys Res,104(B6):13 073～13 082
Kitagawa Y, Koizumi N,Tsuskuda T.1996. Comparison of post-seismic groundwater temperature changes with earthquake-induced volumetric strain release: Yudani hot spring,Japan[J].Geophys Res Lett, 23(22):3147～3150
Koizumi N, W.C. Lai,Y Kitagawa et al.2004.Commnent on“Coseismic hydrological changes associated with dislocation of the September 21,1999 Chichi earthquake,Taiwan”by Min Lee et al[J]. Geophys. Res. Let,31,L13603,doi:10.1029/2004GL019897
Lee Min, Liu Tsung-Kwei, Ma Kuo-Fong.2002. Coseismic hydrological changes associated with dislocation of the September 21, 1999 Chichi earthquake, Taiwan[J]. Geophysical Research Letter , 29(17), 1824, doi:10.1029/2002GL015116
Li VC,Seale SH,Cao T.1987.Post seismic stress and pore pressure readjustment and aftershock distributions[J].Tectonophysics,144:37～54
Lomnitz C.1996.Search of a worldwide catalog for earthquakes triggered at intermediate distances[J].Bull Seism Soc Amer,88;293～298
Matsumoto N.1992.Regression analysis for anomalous changes of ground water level due to eathquake[J], Geophys Res Lett,19(12):1 193～1 196
Michael Manga, Emily E. Brodsky, Michael Boone.2003.Response of streamflow to multiple earthquakes. Geophys Res Lett ,30(5),1214,doi:10.1029/2002GL016618
Mogi k.,H Mochizuki,Y Kurokawa.1989. Temperature changes in an artesian spring at Usami in the Izu Peninsula(Japan) and their relation to earthquake[J].Tectonophysics,159,95-108
Naoji Koizumi,Wen-Chi Lai,Yuichi Kitagawa,et al.2004.Comment on “Coseismic hydrological changes associated with dislocation of the September 21,1999 ChiChi earthquake,Taiwan”by Min lee et al.
Geophysical research letter,VOL.31,L13603,doi:10.1029/2004GL019897 Nur A,Booker J R.1972.Aftershocks caused by pore fuid flow[J]. Science,175:885～887
Noir J,Jacques E,Békri S,et a1.1997.Fluid flow triggered migration of events in the 1989 Dobi earthquake sequence of central Afar[J].Geophys Res Let,24: 2 335～2 338
Parotidis M., S. A. Shapiro, E. Rothert.2005.Evidence for triggering of the Vogtland swarms 2000 by pore pressure diffusion, J. Geophys. Res., 110, B05S10, doi:10.1029/2004JB003267.
Quilty E G,Roeloffs E A.1997.Water level changes in response to theDecember 20,1994 M 4.7 earthquake near Parkfield,California[J].Bull Seism Soc Amer,87:310～317
Rexin E E,Oliver J,Prentiss D.1962.Seismically-Induced Fluctuation of the water level in the Nunn-Bush Well in Milwaukee[J]. Bull Seism Soc Amer,52(1):17-25
Roeloffs E A.1998. Persistent water level changes in a well near Parkfield California,due to local and distant earthquakes[J].J Geophys Res,103(B1): 869～889
Roeloffs E A,Sneed M,Galloway D L,et al.2003.Water-level changes induced by local and distant earthquakes at Long Valley caldera,California[J].J Volcanol Geotherm Res,127:269～303
Timothy L Masterlark,Herbert F wang,Lung S Chan,et al.1999.Coseismic fluid-pressure response estimated from Prediction-error filtering of tidal-band loading. Bull Seism Soc Amer,89(6):1439-1446
Wakita H.1975.Water wells as possible indicators of tectonic strain[J].Science,189:553～555
Wang C.Y. , L. H. Cheng, C. V. chin, et al.2001.Coseismic hydrological changes associated with dislocation of the September 21,1999 Chichi earthquake,Taiwan[J]. Geology ,29,831-834
Wang Chi-yuen,Dreger D S,Wang Chung-Ho.2003.Field relations among coseismic ground motion, water level change and liquefaction for the 1999 Chi-Chi (Mw = 7.5) earthquake, Taiwan[J].Geophysical Research Letter, 30(17), 1890, doi:10.1029/2003GL017601.
Wang Chi-yuen,Chia Yeeping.2008.Mechanism of water level changes during earthquakes: Near fieldversus intermediate field. Geophysical Research Letter, 35, L12402, doi:10.1029/2008GL034227
William S.D. , Wilcock . 2004 . Physical response of mid-ocean ridge hydrothermal system to local earthquakes[J]. Geochemistry Geophysics Geosystems,VOL.5,Q11009,doi:10.1029/2004GC000701
北京市地质局水文地质工程地质大队.1980.平原区第四系含水层分布图.见:北京市水文地质图集[M].上海:中华印刷厂,25
边立恩,贺振华,黄德济.2008.饱含流体介质的地震波场特征及频率分布[J].岩性油气藏,20(3):74～78
蔡义汉.2004.地热直接利用[M].天津:天津大学出版社,20
车用太,鱼金子,刘春国.1996.我国地震地下水温度动态观测与研究[J].水文地质工程地质,4:34～37
车用太,刘喜兰,姚宝树,等.2003.首都圈地区井水温度的动态类型及其成因分析[J].地震地质,25(3):404～419
车用太,鱼金子.2004.塔院井水温的大幅度奇异变化[A].地下流体典型异常的调查与研究[C].北京:气象出版社,414～418
陈建国,陈海昕,章光华,等.2006.轴向浮力驱动下的旋转圆筒热对流稳定性[J].36(3):263～273
崔勇,黄辅琼,李新勇.2006.新04井(泉)流体测项长程趋势变化分析[J].内陆地震,20(1):81～87
丁仁杰.1990.井孔水位记震能力与地震关系探讨[A].见:国家地震局监测司编.地震预报方法实用化研究文集.水位水化专辑[C].北京:地震出版社,186～193
付子忠.1989.地热动态观测与地热前兆[A].见:国家地震局地壳应力所编.地壳构造与地壳应力文集[C].北京:地震出版社
付虹,刘丽芳,王世芹,等.2002.地方震及近震地下水同震震后效应研究[J].地震,22(4):55～66
付虹,刘强,邬成栋,等.2007.印尼8.7、8.5级远震对云南地区小震的激发和地下流体异常响应初探[J].地震研究,30(1):8～13
黄辅琼,迟恭财,徐桂明,等.2000.大陆地下流体对台湾南投7.6级地震的响应研究[J].地震,20(增刊):119～125
高小其,许秋龙,兰陵,等.2002.新04号井、新10号井水位和新10号泉流量地震前兆特征的对比分析[J].内陆地震,16(4):346～354
谷元珠,车用太,鱼金子.等.2003.塔院井水温微动态研究[J].地震,23(1):102～108
顾申宜,李志雄,张慧.2010.海南地区5口井水位对汶川地震的同震响应及其频谱分析[J].地震研究,33(1):35～42
虢顺民,汪洋,计凤桔.1999.云南思茅-普洱地区中强震群发生的构造机制[J].地震研究,22(2):105～115
胡敦宽,李淑芳,刘永铭.1996.白家疃观测井的地热干扰排除及机理探讨[J]地震学报,18(1):128～131
李有才.1995.四川深井水位同震阶变异常与地震预报[J].四川地震,3:32～35
林学钰,廖资生,赵勇胜,等.2005.现代水文地质学[M].北京:地质出版社,19～24
林松,敬成君,刘旭良.2006.热对流稳定性机理研究新方法初探[J].热科学与技术,5(3):236～241
刘全,水鸿寿,张晓轶.2002.数值模拟界面流方法进展[J].力学进展,32(2):259～274
刘耀炜,杨选辉,刘用铭.等.2005.地下流体苏门答腊8.7级地震的响应特征[A].见:中国地震局监测预报司编:2004年印度尼西亚苏门答腊8.7大地震及其对中国大陆地区的影响[C].北京:地震出版社,131～258
戚筱俊.1997.工程地质及水文地质[M].北京:中国水利水电出版社.
石耀霖,曹建玲,马丽,等.2007.唐山井水温的同震变化及其物理解释[J].地震学报,29(3):265～273
舒优良,张世民.2003.周至深井水震波数字化记录与地震波记录的对比研究地震地磁观测与研究[J].24(5):26～31
陶文铨.2009.数值传热学[M].西安:西安交通大学出版社
唐小勇,王新刚,黄建明,等.2005.乌鲁木齐台2004年印尼苏门答腊特大地震钻孔应变观测报告[J].地壳构造与地壳应力,2:14～15
田竹君,车用太,朱清钟,等.1983.塔院井孔水位动态实验站建设[M].北京:国家地震局地质所,1～10
万永革,吴忠良,周公威,等.2002.地震应力触发研究[J].地震学报,24(5):533～551
万登堡.1992.井孔水位记震能力变化与近大地震关系研究[J].地震研究,15(4):381～391
汪成民,车用太,万迪堃，等．1988.地下水微动态研究[M].北京:地震出版社,81～88
汪成民,李宣瑚,王铁城,等.1990.中国地震地下水动态观测网[M].北京:地震出版社,81～88
汪成民.2001.后效预报[A].见:地震预报引论.地下水微动态与地震预报[C]北京:地震出版社:165
王道.2000.天山地震活动区地下水同震、震后效应的研究[J].内陆地震,14(3):252～259
王勇,张国红,代建华,等.2000.昌平地震台地温前兆特征分析[J].西北地震学报,22(3):236～240
汪一鹏,沈军.2000.天山北麓活动构造基本特征[J].新疆地质,18(3):203～210
王大纯,张人权,史毅虹,等.1994.水文地质学基础[M].北京:地质出版社
伍致中,郭富贤.1991.再论博格达推覆构造与油气[J].新疆地质,9(1):40～49
徐锡伟,闻学泽,叶建青,等.2008.汶川Ms8.0地震地表破裂带及其发震构造[J] .地震地质,30(3) :597—629
徐江荣,李超.2008.湍流特征频率的模式理论计算[J].杭州电子科技大学学报,28(4):73～76
杨期隆,辛奎德. 1991.西藏羊八井地热田简介[J].地质论评,37(3):283～287
杨竹转,邓志辉,赵云旭,等.2005.云南思茅大寨井水位同震阶变的初步研究[J].地震学报,27(5):569～574
杨竹转,邓志辉,陶京玲,等.2007.北京塔院井数字化观测水温的同震效应研究[J].地震学报,29(2):203～213
徐世光,郭远生.2009.地热学基础[M].北京:科学出版社,20～22
鱼金子,车用太,王爱英,等.1995.大同-阳高地震前静乐井水位记震能力的异常及其意义[J].山西地震,1:50～53
鱼金子,车用太,刘伍洲.1997.井水温度微动态形成的水动力学机制研究[J].地震,17(4):389～396
张素欣,杨卫东、张子广.2007.唐山矿井模拟与数字水位的记震能力对比分析[J].西北地震学报,29(2):170～173
张昭栋、迟镇乐、陈会民,等.2000.井水位的振荡与地震波[J].地震研究,23(4):418—425
张昭栋,赵淑平,董传富,等.1994.井水位阶变与含水层所受体应力之间的定量关系[J].地球物理学报,37(1):222～229
张子广,张素欣,李薇,等.2007.昌黎井水温潮汐形成机理分析[J].地震,27(3):34～40
张子广,万迪堃，董守玉．1998.水震波与地震面波的对比研究及其应用[J].地震,18(4):399～404
庄礼贤,尹协远,马晖扬.1997.流体力学[M].安徽:中国科学技术大学出版社
邹剑峰,郜冶.2007.含加热圆管方腔内自然对流的数值研究[J].热科学与技术,6(4):326～330
庄光国.1999.台湾海峡7.3级地震水震波特征研究[J].地震,19(2),199～203