带地形高频大地电磁法有限元数值模拟

随着中深度地球物理勘探的迫切需要,以美国EH-4电导率成像系统为代表的高频大地电磁法在我国地球物理勘探行业应用越来越广泛,其勘探深度在地下1000 m以浅,频率范围10 Hz～100 KHz.本文研究了带地形的高频率大地电磁法有限元高精度、快速数值模拟.首先利用广义变分原理推导出电磁场的有限元方程,编制了双线性和双二次插值有限元程序,对典型山谷和山脊地形的数值模拟结果进行了分析,得出了一些有意义的结论.     

基于南水北调西线工程岩性特征的CSAMT法有限元三维数值模拟研究

有限元法是地球物理数值模拟中常用的方法,本文采用三维可控源音频大地电磁法(CSAMT)有限元数值模拟的程序,根据南水北调西线工程岩性的地球物理特征,设置了不同的三维模型,并对其进行了有限元数值计算分析,从三维空间中模拟场的规律,探索了不同地质异常体的特征,为提高可控源音频大地电磁法在南水北调西线工程深埋藏隧洞探测地质异...     

基于非结构双网格的2D RMT双参数同步反演研究

Radio-magnetotelluric(RMT)是以无线电发射机为信号源的一种地球物理勘探方法,近年来被广泛应用于数米至数十米内的近地表工程和环境地球物理勘探.目前,各类电磁资料的反演均是以寻求满足目标拟合差的地下介质电阻率分布为目的.然而,对于勘探频率为10~300kHz的RMT数据,由介电常数所引起的波动场在总场中的比例可达20%以上,在这种情况下,忽略介电常数,仅通过电阻率参数的反演来进行数据拟合势必降低反演资料解释的准确性.为解决这一问题,本文研究了基于电阻率-介电常数的双参数同步反演算法.构建了一个全新的双参数目标函数,并推导了双参数反演迭代方程组;通过灵敏度分析,研究了电阻率和介电常数对正演响应的影响,并据此提出相对电导率的概念,统一了反演参数的灵敏度;通过理论模型分析了参考频率、双参数正则化因子对反演结果的影响,并给出了一般性的参数优选方案.此外,为了能够灵活处理复杂地形,本文采用非结构的正反演双网格进行模型离散,并通过局部加密技术保证反演的速度和精度.最后,对一带地形的理论模型分别进行了单参数和双参数反演,结果表明单参数反演无法正确反映出地电信息,而双参数反演能够准确得到异常的分布,验证了本文所开发的双参数反演程序的有效性.     

大亚湾中微子试验隧道工程不同物探方法探测结果综合分析

对不同埋深花岗岩采用重力法、电法和电磁法、地震法等地球物理勘探方法的结果差异性如何?本文以广东大亚湾花岗岩为例,先后开展了重力法、高密度电法、地震折射法、可控源音频大地电磁法(CSAMT)等4种不同方法,探测中微子场地断层破碎带、花岗岩与沉积岩接触带、风化壳厚度、地下水埋深等与隧道工程设计方案和围岩稳定性密切相关的地质条件.地球物理勘探以岩层物性(电阻率、波速和密度等)为基础,这些特征与岩性及地下水状态相关.同时相同物性特征对应不同岩层和含水状态,因而物探结果具有多解性.在工作中采用多种物探方法进行了相互补充验证.勘探结果比较后发现,对浅表50 m以内风化壳厚度,以人工重锤为激发震源的地震折射法效果较好,对风化壳分带和厚度达100多米的风化囊,高密度电法探测效果较好,而对于深大断裂或密闭断裂,重力法和电磁法效果较好.这四种不同物探方法结果的综合分析,对全面深入认识花岗岩地质条件提供了依据,其结果得到了钻探和隧道工程施工开挖揭露的验证.     

浅部频率域电磁勘探方法综述

适用于近地表(2000m以内)勘探的频率域电磁法主要有音频大地电磁法(audio-frequency magnetotellurics,AMT),无线电大地电磁法(radio-magnetotellurics,RMT),可控源音频大地电磁法(controlled source audiofrequency magnetotellurics,CSAMT),广域电磁法(Wide Field Electromagnetic Method,WFEM).本文拟从最新的数据采集技术、数据处理技术、正反演算法、实例等四个方面,论述适用于浅部勘探的AMT,RMT,CSAMT和WFEM方法的国内外最新进展,总结目前AMT,RMT,CSAMT和WFEM方法遇到的困难,对潜在的发展方向提出建议.综述表明:(1)张量测量、多站阵列、多站叠加可提高AMT、RMT和CSAMT数据的质量.利用近区数据WFEM法可获得良好的效果.国产与国外仪器在质量方面的差距正在逐步缩小.(2)数学形态滤波技术、Hilbert-Huang变换等可有效分离出有用的数据,局部畸变仍然是亟待解决的难题,需要更为深入的研究.(3)矢量有限元与非结构网格的出现大幅度提高了有限元处理复杂电磁问题模拟的精度与应用范围,成为目前电磁正演的首选工具.完全非线性反演算法仍然局限于1D、2D问题,共轭梯度法和高斯牛顿算法等为解决3D问题的发展趋势.地质约束的引入和多数据联合反演可以减小反演的非唯一性.各向异性的反演为目前反演研究的热点之一.(4)野外数据解释的正确性严重依赖于对地下结构先期的维性判别,在2D特性不明显、3D特性明显时,需要采用3D进行反演解释.     

音频大地电磁与浅层地震在隧道地质勘查中的应用

针对山区长埋深隧洞勘测技术的研究,是近年来我国隧洞地质勘查中的研究热点,然而传统的钻探等手段在多山区域有很大的局限性,因此,近些年来地球物理勘探方法逐渐面向工程勘探,而单一的物探方法往往在异常判断方面存在局限性,鉴于此在工程勘探中往往运用多种地球物探方法相结合,可以在一定程度上减小了单一物探方法的多解性,提高物探解释的质量以及精度.本文以南水北调某工程为例,在某受水段分别布置了大地电磁法、地震浅层反射物探方法,其中大地电磁法主要用于辅助查明隧洞覆盖层厚度、地层岩性、隐伏构造、地下水位等工程地质条件,地震浅层反射用于查明小规模的地质构造(小断层、溶洞等)的勘查,经过这两种方法的结合,有效的查明了工程受水区的基本地质情况.     

可控源音频大地电磁三维共轭梯度反演研究

可控源音频大地电磁法在资源勘探等领域中发挥着重要的作用.我们把有限差分数值模拟方法用于可控源音频大地电磁三维正演,结合正则化反演方案和共轭梯度反演的思路,将反演中的雅可比矩阵计算问题转为求解两次"拟正演"问题,得到模型参数的更新步长,形成反演迭代,实现了可控源音频大地电磁三维共轭梯度反演算法.该反演算法可用于对有限长度电偶源激发下采集到的可控源音频大地电磁全区(近区、过渡区和远区)视电阻率和相位资料进行三维反演定量解释,获得地下三维模型的电阻率结构.理论模型合成数据的反演算例验证了所实现的可控源音频大地电磁三维共轭梯度反演算法的有效性和稳定性.     

聚类分析算法在大地电磁三维解释中的应用

为了提高对大地电磁三维反演结果的分析和解释能力,完成构造界面识别、异常构造刻画等地质和地球物理解释,本文提出使用非监督的聚类方法分析大地电磁反演结果.根据三维反演模型电阻率值的分布和各种聚类分析方法的特点,选择使用K均值聚类方法对反演模型电阻率值进行聚类分析.在K均值聚类分析过程中,本文采用了RS指数指导选择聚类数目, Kaufman法进行中心初始化.通过对东北地区大地电磁数据三维反演结果使用K均值聚类分析方法,得到了东北地区电性岩石圈的厚度估计,结果表明东北地区岩石圈底部聚类电阻率大约为339Ω·m,其中松辽盆地岩石圈最薄,约为60 km;大兴安岭地区最厚,约为150 km;佳木斯地体厚度约为100 km;而长白山地区岩石圈厚度不易确定,可能受新生代构造活动影响,电阻率明显减小.聚类方法能够有效地帮助对大地电磁三维反演结果中的地质构造进行识别和归类.     

某新建铁路复杂长隧道岩层地质与CSAMT特征

分析了多种地球物理勘探方法在复杂长隧道探测中的优缺点,指出了CSAMT（可控源音频大地电磁）法在工作条件极其困难,地形复杂,交通不便,要求勘探深度大的隧洞勘察中,是一种行之有效的物探手段.以较高的显示度列举了CSAMT法在本区探测成功的实例.就隧道主要特殊岩层膏溶角砾岩的地球物理异常特征进行了详尽分析,给出了CSAMT法在本区隧道结构面综合地质解释结果.     

瞬变电磁法拟地震成像研究进展

瞬变电磁法是工程地球物理勘察中应用较多的一种勘探方法之一,瞬变电磁法拟地震成像方法研究是当前电磁探测理论与应用研究的热门.文中详细叙述了瞬变电磁法拟地震成像的两种不同思路,一种是建立时间—频率转换经验公式,把TEM数据转换成平面波数据,借鉴大地电磁法拟地震解释思路进行成像,另一种是通过积分变换,把扩散场变成虚拟波场,把电磁衰减信号变成拟地震子波,然后按照地震解释法进行解释,并分析了各自存在的的问题.指出下一步研究的方向.     

2-D Versus 3-D Magnetotelluric Data Interpretation

In recent years, the number of publications dealing with the mathematical and physical 3-D aspects of the magnetotelluric method has increased drastically. However, field experiments on a grid are often impractical and surveys are frequently restricted to single or widely separated profiles. So, in many cases we find ourselves with the following question: is the applicability of the 2-D hypothesis valid to extract geoelectric and geological information from real 3-D environments? The aim of this paper is to explore a few instructive but general situations to understand the basics of a 2-D interpretation of 3-D magnetotelluric data and to determine which data subset (TE-mode or TM-mode) is best for obtaining the electrical conductivity distribution of the subsurface using 2-D techniques. A review of the mathematical and physical fundamentals of the electromagnetic fields generated by a simple 3-D structure allows us to prioritise the choice of modes in a 2-D interpretation of responses influenced by 3-D structures. This analysis is corroborated by numerical results from synthetic models and by real data acquired by other authors. One important result of this analysis is that the mode most unaffected by 3-D effects depends on the position of the 3-D structure with respect to the regional 2-D strike direction. When the 3-D body is normal to the regional strike, the TE-mode is affected mainly by galvanic effects, while the TM-mode is affected by galvanic and inductive effects. In this case, a 2-D interpretation of the TM-mode is prone to error. When the 3-D body is parallel to the regional 2-D strike the TE-mode is affected by galvanic and inductive effects and the TM-mode is affected mainly by galvanic effects, making it more suitable for 2-D interpretation. In general, a wise 2-D interpretation of 3-D magnetotelluric data can be a guide to a reasonable geological interpretation.     

CSRMT measurements in the frequency range of 1–250 kHz to map a normal fault in the Volvi basin, Greece

In order to gain a better understanding of the geometry of surface faults, five Controlled Source/Radio Magnetotelluric (CSRMT) profiles were measured across the Volvi basin, 45 km northeast of the city of Thessaloniki in Greece. The data were collected in two frequency ranges: a) 1–12.5 kHz using a remotely controlled double horizontal magnetic dipole transmitter (CSAMT measurements), and b) 15–250 kHz using the signal from distant radio transmitters (RMT measurements). The transition from the RMT band to the CSAMT band was smooth and continuous allowing us to combine both datasets for plane-wave modeling. The surface geology shows a predominantly 2D structure, and therefore we planned the survey into profiles perpendicular to the geological strike. We have used a 2D interpretation tool to model the data in TE, TM, TE + TM and determinant modes. Using a 4% error floor on the impedance, 2D resistivity models from inversion of the determinant data provide lower RMS data fits (4.2 and 1.2 for resistivity and phase, respectively) compared to the combined TE + TM data (4.4, 2.8, overall resistivity and phase, respectively). 2D inversion of the measured tensor data shows a sharp change in the depth to the top of resistive gneiss–schist basement that is overlain by a less resistive overburden at southern basin flanks. The change in depth to the bedrock is clearly seen in all 2D models along the measured profiles suggesting the existence of normal faults with strike directions of NE–SW to E–W. The 2D electrical resistivity models suggest that the bedrock deepens towards south-west. The resistivity models are also compared with the existing borehole information in the area and show a reasonable correlation. For example the sharp change of depth to the bedrock towards the center of the basin as seen in the resistivity models are also confirmed by the borehole data.     

A radio magnetotelluric study to evaluate the extents of a limestone quarry in Estonia

Electromagnetic signals from distant radio transmitters in the frequency range 15–250 kHz were measured to model an electrical resistivity structure beneath 7 profiles in the vicinity of the Karinu limestone quarry in Estonia with the aim to map the extent of the economically exploitable limestone. The resistivity models from a 2D inversion of determinant resistivity and phase values using an Occam type of regularization contained reasonably accurate information about the geometry, namely depth to the top and the bottom of the target high‐resistivity limestone. The resistivity models correlated well with existing geological evidences as well as information from closely located boreholes. However, the sharp lithological boundaries seen in the boreholes were not resolved exactly in the resistivity models. This is probably because of the smoothing regularization used in the inversion process. Combined use of borehole data together with resistivity models resulted in two major geological interpretations; a) towards the western part of the existing limestone quarry there is a NNW to NS striking fault, covered by post‐glacial sediments, b) a potential cost‐effective exploitable area containing high quality highly resistive limestone is located south of the existing quarry. This case study shows the applicability of the reasonably fast radio magnetotelluric (RMT) method for the exploration of near‐surface resources.     

THE IMPORTANCE OF TOPOGRAPHIC CORRECTIONS ON MAGNETOTELLURIC RESPONSE DATA FROM RUGGED REGIONS OF ANATOLIA

Topographic irregularities cause some distortions of magnetotelluric (MT) fields. In the vicinity of a topographic feature, the TM-mode distortion increases with the height and inclination of the slope. It is well-known that TM-mode ( E _⊥) topographic effects are much greater than TE-mode ( E _ı) distortions.   We have made a study of MT anomalies in TM-mode due to two-dimensional topography. In order to reduce these effects, the distortion tensor stripping technique was used. After corrections, the resulting data can be interpreted as if they were obtained over a flat surface and depend only on the subsurface structure. However, this technique sometimes causes some geometrical distortions of the real subsurface structure.   One of our aims is to overcome this failure. We have modified the correction coefficients by considering the actual one-dimensional geology. Model studies showed that our approach is especially useful in removing the terrain effects on complex 2D subsurface structures.   The other purpose of this study is to emphasize the importance of a proper terrain correction for data from sites having mountainous topography over complex geology, e.g. strike-slip faults, suture zones and rift valleys. Some examples of MT data sets collected from the North Anatolian Fault Zone and from the thrust regions of the Western Taurides will be presented.     

CSAMT单分量数据解释方法

可控源音频大地电磁法(CSAMT)一直沿用大地电磁法(MT)的办法,通过计算电场分量与磁场分量的比值,求取卡尼亚视电阻率.而CSAMT场源已知,电场分量和磁场分量都与地下电阻率存在一定的关系,可以单独采用CSAMT电场分量或者磁场分量提取地下介质的视电阻率.本文通过分析电场分量与磁场分量的数据特性,提出利用CSAMT电场单分量数据进行视电阻率的计算,用改进的广义逆矩阵反演方法,使初始模型中的地电层数等于频道个数,克服了以往反演计算中层数较少的问题;实现全场区电场分量视电阻率曲线的拟合反演.同时对单分量视相位计算方法进行分析,结合山西大同地区积水采空区探测及数据解释结果,论证本文提出的单分量解释方法的有效性.     

3D inversion of DC data using artificial neural networks

In this paper, we investigate the applicability of artificial neural networks in inverting three-dimensional DC resistivity imaging data. The model used to produce synthetic data for training the artificial neural network (ANN) system was a homogeneous medium of resistivity 100 Ωm with an embedded anomalous body of resistivity 1000 Ωm. The different sizes for anomalous body were selected and their location was changed to different positions within the homogeneous model mesh elements. The 3D data set was generated using a finite element forward modeling code through standard 3D modeling software. We investigated different learning paradigms in the training process of the neural network. Resilient propagation was more efficient than any other paradigm. We studied the effect of the data type used on neural network inversion and found that the use of location and the apparent resistivity of data points as the input and corresponding true resistivity as the output of networks produces satisfactory results. We also investigated the effect of the training data pool volume on the inversion properties. We created several synthetic data sets to study the interpolation and extrapolation properties of the ANN. The range of 100–1000 Ωm was divided into six resistivity values as the background resistivity and different resistivity values were also used for the anomalous body. Results from numerous neural network tests indicate that the neural network possesses sufficient interpolation and extrapolation abilities with the selected volume of training data. The trained network was also applied on a real field dataset, collected by a pole-pole array using a square grid (8 ×8) with a 2-m electrode spacing. The inversion results demonstrate that the trained network was able to invert three-dimensional electrical resistivity imaging data. The interpreted results of neural network also agree with the known information about the investigation area.     

Near surface geophysical surveys with a high frequency mutual impedance measuring system

Relationships for the mutual impedance of horizontal coplanar, vertical coplanar, perpendicular and vertical coaxial loop antenna arrays operating at a high frequency, located over a layered conducting medium, have been derived. The relationships take into account the occurrence of displacement currents in both the free space and the conducting medium. The relations can be used for calculating the earth parameters in geophysical prospecting. The use of high frequency antenna array makes it possible to determine distributions of apparent resistivity as well as apparent electrical permittivity. The potential application of the measurements of mutual impedance of loop antennas operating at a high frequency to the geophysical examinations is presented. The measuring system developed for this purpose is described. A simple method of interpreting the mutual impedance measurements to obtain the traces of apparent resistivity and apparent permittivity of the investigated earth is given.     

EM AND IP RESPONSE OF A STEEL WELL CASING FOR A FOUR-ELECTRODE SURFACE ARRAY. PART II: NUMERICAL RESULTS*

Some numerical results for the apparent resistivity of a homogeneous half space containing a vertical steel well casing as measured by a dipole-dipole array are presented. The basic model includes the electromagnetic and induced polarization effects by allowing for frequency dependent eddy currents and an electrochemical boundary of the well casing. It is shown that the well casing produces an apparent complex resistivity response that could be mistaken for an induced polarization anomaly existing above a hydrocarbon deposit. This response of the well casing is strongly dependent on frequency.     

Resolution of airborne VLF data

The interpretation of airborne VLF data represents an important aspect of geophysical mapping of the upper few hundred meters of the Earth's crust, especially in areas with crystalline rocks. We have examined the ability of the single frequency VLF method to provide quantitative subsurface resistivity information using two generic models and standard airborne parameters with a flight altitude of 70 m and a frequency of 16 kHz. The models are long thin conductor (10 m thick, 10 Ω m resistivity and 1 km long) and a wider buried conductive dike (100 Ω m resistivity and 500 m wide). Using standard regularized inversion it turned out that for both models the conductivity of the conductors are underestimated and the vertical resolution is rather poor. The lateral positions of the minimum of the resistivity distributions coincide well with the true positions of the shallow conductors. For deeper conductors the position of the minimum resistivity moves from the edges of the conductor into the conductor. The depth to the minimum of the resistivity anomalies correlates well with the true depth to the top of the conductors although the latter is always smaller than the former.Interpretation of field airborne data collected at 70 m flight height resolved both small scale and large scale near surface conductors (conductance ∼1 S). Deeper conductors show up in the VLF data as very long wavelength anomalies that are particularly powerful in delineating the lateral boundaries of the conductors. Many of the VLF anomalies in the Stockholm area are dominated by these deep conductor responses with some near surface conductors superimposed. The deep conductors often follow topographic lows coinciding with metasediments. We interpret the frequent absence of near surface responses at 70 m flight height as a result of weak coupling between the primary VLF wave and the small scale (in all three dimensions) near-surface conductors.Radio magnetotelluric (RMT) ground measurements were carried out along a short profile coinciding with part of an airborne profile. Using data at 9 frequencies (14–250 kHz) small scale conductors in the upper few tens of meters, not identified from the airborne data, could be well resolved. Large scale deeper conductors could be identified by both methods at nearly the same positions.