Magnetotelluric(MT)three-dimensional inversion has the advantages of simple data preprocessing, the model is close to actual situation, and the inversion result is more reliable and stable. It is one of the most advanced research topics and would take the place of the dominant two-dimensional inversion definitely. With the improvement of computing capability of computers and the breakthrough in inversion methods, great progress was made in MT three-dimensional inversion in recent years, from the theoretical research and test of this method at the beginning to the current application to practical data interpretation. For the great computation amount of MT three-dimensional inversion, current MT three-dimensional inversion algorithm programs are all implemented in parallel way and it is recommended to do three-dimensional inversion calculations on supercomputing system to make better use of computing resources and improve the inversion efficiency.

Different from the MT three-dimensional inversion algorithm programs which have basically realized the utility function, the practical application of MT three-dimensional inversion is still in an early stage. Users should be familiar with the use of multiple software and fulfill the function manually with the help of the software as follows: generating the files required for the inversion program, connecting to the supercomputer to upload data, inputting the command to perform the inversion, etc. The process of manually connecting and operating calculations is the most primitive cloud computing. All processes need to be done manually, which would cause not only heavy workload and the complicated operation, but also the problems for the long-term effective preservation and management of complex inversion data.

To conquer this, we develop independently a three-dimensional magnetotelluric inversion cloud computing system, toPeak, using Delphi language. This paper introduces some main features of toPeak. To begin with, system design and analysis are carried out in combination with the current situation and system structure and functions are realized. The main idea is to realize a set of cloud computing system platform based on server-client(C/S), on the basis of perfect inversion data management, integrate the most advanced three-dimensional magnetotelluric inversion algorithm program in the cloud, and connect through the Internet to realize all the system functions of three-dimensional magnetotelluric inversion. Then, the different parts of toPeak are introduced separately, including design structures and designs. The server is deployed on the supercomputer system(supercomputing)to receive the data for inversion tasks, configure and manage the storage of the inversion result data. Combined with the Internet connection, the server and the Internet together constitute a computing cloud. The client is deployed on the users’ windows operating system, including Windows visual data integration processing software and Internet operation middleware. The client is designed on the basis of object-oriented programming ideas, with data as the core, using data engineering objects to encapsulate and store all MT data, process and interpret the results, realize data processing inversion and other operations around this data project, and display the process and results of these processing and inversion in graphics using visualization technology. Internet operation middleware connects the client and server based on the SSH protocol to realize data processing and inversion, transmission and command sending and receiving. Furthermore, the whole work flow of inversion using toPeak and parts of procedure of it are shown. At last, some inversion results from toPeak are displayed. toPeak has realized the full functions require for implementing three-dimensional inversion and can grid, process and select, inverse and explain the data. It is a good tool for the practical use of three-dimensional inversion.

Magnetotelluric(MT)is a method of detecting electrical structures. The natural field source signal is weak, and there are many factors that affect the impedance estimation results, such as dead band, near-field interference, and random noise, so it is difficult to obtain accurate electromagnetic response in strong interference area. The stable and reliable impedance estimation is the premise for the follow-up inversion and interpretation. In order to suppress noise and improve the accuracy of impedance estimation, researchers have proposed various new data processing methods. However, these data processing methods are not widely used due to insufficient stability and poor applicability. The classic remote-reference method and robust estimation method are still the most widely used methods. This paper analyzes the characteristics of the strong interference data and the applicable scope of various data processing methods, combined with the processing effect of the measured magnetotelluric data in the strong interference area in eastern China, and summarizes a set of data processing strategies suitable for the strong interference area.

The remote-reference method can effectively suppress coherent noise. It is essential in data processing in strong interference areas. Usually, the results will be improved after processing by remote reference. The remote-reference site should be selected at a place far enough away from the measuring point without interference.

Robust estimation can highlight high-coherence signals and suppress low-coherence signals. In the dead band, the coherence of the natural field signal is higher than that of the background noise signal, so the robust estimation processing can improve the data processing result of the dead band. The intensity and coherence of the long-lasting near-field interference signal is higher than that of the natural-field signal. The robust estimation process will treat the near-field interference as the desired signal and suppress the natural source signal. Therefore, data containing long-term strong near-field interference is not suitable for using robust estimation but non-robust estimation. For data that is not well processed by the two methods, we can try a combination of the two. By carefully selecting the power spectrum obtained by the two methods, it is possible to improve the processing result.

Increasing the number of data segments can provide more sets of power spectra for selection, and also increase the probability of obtaining higher quality power spectra. Through careful selection of multiple power spectra, it is more likely to obtain better processing results than when the number of segments is smaller.

During the day when there is a lot of human activity, the interference signal is strong. And at night, the interference signal is weak. The measured data well proves this point, so we should extend the acquisition time at night as much as possible, and the data processing should also focus on the night data.

In general, it is more likely to obtain better data with longer acquisition time. Research on synthetic data shows that the maximum valid period of magnetotelluric theoretical data is 1/8 of the data duration. The measured data results of Fengning Station also support this conclusion. The longer the data acquisition time is, the more effective power spectra can be obtained, and the more likely it is to select a better quality spectrum from them, and obtain a stable impedance estimation result. Therefore, the data collection time should be adjusted reasonably according to the interference situation during the observation to ensure the stability of the impedance estimation result.

Magnetotelluric data processing methods are not invariable, and different data processing methods should be adopted according to the actual situation. When the better data processing method is not yet mature, flexible application of existing method is the necessary means for magnetotelluric data processing.

The electromagnetic(EM)method using controlled-source extremely low-frequency(CSELF)waves is a new technology based on the large-power alternating electromagnetic field generated by an artificial procedure. The biggest advantage of this technology is that it has a long transmitting antenna(tens to hundreds of kilometers)and a large transmitting current(hundreds of amps)and can emit strong and stable electromagnetic waves, covering millions of square kilometers. It can be applied to earthquake monitoring, surveys for mineral resources and treatment of waste nuclear material as well as marine and land communication and detection to ionospheric structure in space. At present, domestic theoretical research on CSELF is not mature enough. This paper has carried out a more detailed study on the spatial propagation characteristics of the electromagnetic(EM)of controlled-source extremely low frequency(CSELF).

The large-power CSELF EM waves cover almost all sections of space which can be divided into near, far and waveguide zones according to their propagation characteristics. The propagation of electromagnetic waves in the near and far zone is mainly manifested as the distribution and induction of the conductive currents, and the displacement current and effects of the ionosphere and spheric structure of the Earth can be neglected. The propagation theory of CSELF EM wave is similar to CSAMT in the near and far zones, and it can be described by the theory of quasi-stable field which is analogous to that of the classical theory of EM sounding. In this paper, we collated and verified the field strength calculation formulas in the existing literature. While in the waveguide zone, EM waves appear mainly as the displacement current, and the displacement current and effects of the ionosphere and spheric structure of the Earth must also be considered. The electromagnetic field is mainly the radiation field, and it runs in a way completely different from what the classic theory describes. Using the achievements of communication technology for reference, this paper presents the approximate calculation formula of CSELF EM wave of the earth-air-ionosphere spherical cavity model. Based on the field strength calculation formulas of the three regions obtained above, this paper has designed a piece of visualized software for calculation of the CSELF EM field in three coordinate systems(Cartesian, cylindrical and spherical coordinates). Finally, according to the calculation results, the spatial propagation characteristics of CSELF in the near area, far area and waveguide area are analyzed.

The results show that the decay of CSELF EM field intensity is rapid in the near and far zone, but slightly slow in the far zone, which reflects the spatial distribution characteristics of the induced field in the lossy medium and the radiation field in the dielectric medium. The electric field enters the waveguide zone earlier than the magnetic field. Under the earth model, there is an increase in the field strength in the waveguide area near the antipole of the dipole source which shows completely different EM waves propagation characteristics in horizontal formation model. According to the calculation results of the CSELF EM field in near and far zones under the three coordinate systems, it is found that in the Cartesian coordinate system, the horizontal components have two zero lines and are distributed in four quadrants. While the vertical component field has only one zero line and are distributed in two half planes. In the cylindrical and spherical coordinate systems, all field components have merely one zero line and are characterized by half-plane distribution. The location of the zero line should be avoided as much as possible in the layout of field observation stations. We can choose different coordinate systems to solve this problem. In addition, it is also recognized that in the frequency domain EM sounding based on the horizontal electric dipole source, the far-field sounding mainly depends on the magnetic field rather than the electric field. Furthermore, it is recognized that in the frequency domain electromagnetic sounding method based on the horizontal electric dipole, the horizontal component of the electric field in the near zone is proportional to the resistivity of the medium, and has nothing to do with the frequency; the vertical component is proportional to the frequency and has nothing to do with the dielectric resistivity; the magnetic field has no relationship with the frequency and the dielectric conductivity. In the far zone, the horizontal component of the electric field is basically independent of frequency, and the vertical component of the electric field is related to both frequency and earth conductivity. However, due to the difficulty of observation, it is generally not used in the actual sounding. The three components of magnetic field in the far zone are all related to the frequency and the earth’s conductivity, so the far-field sounding mainly depends on the magnetic field rather than the electric field.

Since CSELF antennas are generally very long(tens to hundreds of kilometers), the antenna can no longer be regarded as an electric dipole when measuring in the near and far zones, but should be regarded as a long wire source composed of multiple electric dipoles. In this paper, the electric dipole theory is still used for analysis, which has certain limitations that need to be overcome by further in-depth research.

The first control source extremely low frequency(CSELF)electromagnetic observation network through the world, consisting of 30 fixed stations located in the Beijing captical circle region(15 staions)and the sourthern secton of the north-south earthquake belt(15 stations), China, has been established under the support of the wireless electromagnetic method(WEM)project, one of the national science and technology infrastructure construction projects during the 11th Five-year Plan period. As a subsystem of the WEM project, the CSELF network is mainly to study the relationship between elctromagnetic anomalies and mechanisms of earthquake, and further improve our ability to monitor and predict earthquakes by monitoring real-time dynamic changes in both electromagnetic fields and subsurface electric structure. Carrying out the detection of the underground background electric structure in the CSELF network area/station is an important part of this project and of great significance to play its role in the study of earthquake prediction and forecast. In this paper, we elaborate how to acquire the subsurface electric structure of the CSELF network in the Beijing captical circle region and make a simple explanation for the structure. Firstly, a short magnetotelluric(MT)profile, almostly perpendicular to the regional geological strike, was deployed at each station of the CSELF network in the capital circle region during the 2016 and a total of 60 broadband MT sites was collected using ADU -07e systems. Then, all the time series data were processed carefully using the robust method with remote reference technique to MT transfer functions. MT data quality was assessed using the D+algorithm. In general, data at most sites are of high quality as shown by the good consistency in the apparent resistivity and phase curves. Different impedance tensor decomposition methods including the phase tensor analysis, Groom and Bailey(GB)tensor decompositon, and statistical image method based on multi-site, multi-frequency tensor decompositon were used to analyze data dimensionality and directionality. For data inversion, on the one hand, one-dimensional(1-D)subsurface electrical resistivity structures at each station and MT site were derived from 1-D adaptive regularized MT inversion algorithm. On the other hand, we also imaged the 2-D electric structures along the short MT profile by the nonlinear conjugate gradients inversion algorithm at each station. Robustness of all 2-D structures along each short profile were verified by sensitivity tests. Although fixed stations and MT sites are limited and distributed unevenly, the 3-D inversion of 15 stations was also performed to produce a 3-D crustal electrical resistivity model for the entire network using the modular system for 3-D MT inverson: ModEM based on the nonlinear conjugate gradients algorithem. Intergrating 1-D, 2-D and 3-D inversion results, the resistivity structure beneath the CSELF network in captical circle region revealed some significant features: The crustal electrical structures are mainly characterized by high resistivity beneath the Yinshan-Yanshan orogenic belt in the northern margin of North China, the Taihangshan area in the middle, the Jiao-Liao block in the east, while the North China Plain and Shanxi depression areas have relatively lower resistivity in the crust; There are obvious electrical resistivity difference on both sides of the gravity gradient of Taihang Mountains and the Tanlu fault zone, which indicates they could be manifested as an electric structure boundary zone, respectively. Overall, the electric structure characteristics of the entire network area shows high correspondence with the regional geological structure and earthquake activity to some extent. In summary, implementing the detection of underground electrical resistivity structure in the CSELF network of the capital circle region will provide important foundations for the researches on the regional seismogenic environment, the generation mechanism of seismic electromagnetic anomaly signals, and earthquake prediction and forecast.

The East Kunlun Fault is a giant fault in northern Tibetan, extending eastward and a boundary between the Songpan-Ganzi block and the West Qinling orogenic zone. The East Kunlun Fault branches out into a horsetail structure which is formed by several branch faults. The 2017 Jiuzhaigou M_S7.0 earthquake occurred in the horsetail structure of the East Kunlun Fault and caused huge casualties. As one of several major faults that regulate the expansion of the Tibetan plateau, the complexity of the deep extension geometry of the East Kunlun Fault has also attracted a large number of geophysical exploration studies in this area, but only a few are across the Jiuzhaigou earthquake region. Changes in pressure or slip caused by the fluid can cause changes in fault activity. The presence of fluid can cause the conductivity of the rock mass inside the fault zone to increase significantly. MT method is the most sensitive geophysical method to reflect the conductivity of the rock mass. Thus MT is often used to study the segmented structure of active fault zones. In recent years MT exploration has been carried out in several earthquake regions and the results suggest that the location of main shock and aftershocks are controlled by the resistivity structure. In order to study the deep extension characteristics of the East Kunlun Fault and the distribution of the medium properties within the fault zone, we carried out a MT exploration study across the Tazang section of the East Kunlun Fault in 2016. The profile in this study crosses the Jiuzhaigou earthquake region. Other two MT profiles that cross the Maqu section of East Kunlun Fault performed by previous researches are also collected. Phase tensor decomposition is used in this paper to analyze the dimensionality and the change in resistivity with depth. The structure of Songpan-Ganzi block is simple from deep to shallow. The structure of West Qinlin orogenic zone is complex in the east and simple in the west. The structure near the East Kunlun Fault is complex. We use 3D inversion to image the three MT profiles and obtained 3D electrical structure along three profiles. The root-mean-square misfit of inversions is 2.60 and 2.70. Our results reveal that in the tightened northwest part of the horsetail structure, the East Kunlun Fault, the Bailongjiang Fault, and the Guanggaishan-Dieshan Fault are electrical boundaries that dip to the southwest. The three faults combine in the mid-lower crust to form a “flower structure”that expands from south to north. In the southeastward spreading part of the horsetail structure, the north section of the Huya Fault is an electrical boundary that extends deep. The Tazang Fault has obvious smaller scale than the Huya Fault. The Minjiang Fault is an electrical boundary in the upper crust. The Huya Fault and the Tazang Fault form a one-side flower structure. The Bailongjiang and the Guanggaishan-Dieshan Fault form a “flower structure”that expands from south to north too. The two “flower structures”combine in the high conductivity layer of mid-lower crust. In Songpan-Ganzi block, there is a three-layer structure where the second layer is a high conductivity layer. In the West Qinling orogenic zone, there is a similar structure with the Songpan-Ganzi block, but the high conductivity layer in the West Qinling orogenic zone is shallower than the high conductivity layer in the Songpan-Ganzi block. The hypocenter of 2017 M_S7.0 Jiuzhaigou earthquake is between the high and low resistivity bodies at the shallow northeastern boundary of the high conductivity layer. The low resistivity body is prone to move and deform. The high resistivity body blocked the movement of low resistivity body. Such a structure and the movement mode cause the uplift near the East Kunlun Fault. The electrical structure and rheological structure of Jiuzhaigou earthquake region suggest that the focal depth of the earthquake is less than 11km. The Huya Fault extends deeper than the Tazang Fault. The seismogenic fault of the 2017 Jiuzhaigou earthquake is the Huya Fault. The high conductivity layer is deep in the southwest and shallow in the northeast, which indicates that the northeast movement of Tibetan plateau is the cause of the 2017 Jiuzhaigou earthquake.

Many synthetic model studies suggested that the best way to obtain good 3D interpretation results is to distribute the MT sites at a 2D grid array with regular site spacing over the target area. However, MT 3D inversion was very difficult about 10 years ago. A lot of MT data were collected along one profile and then interpreted with 2D inversion. How to apply the state-of-the-art 3D inversion technique to interpret the accumulated mass MT profiles data is an important topic. Some studies on 3D inversion of measured MT profile data suggested that 2D inversions usually had higher resolution for the subsurface than 3D inversions. Meanwhile, they often made their interpretation based on 2D inversion results, and 3D inversion results were only used to evaluate whether the overall resistivity structures were correct. Some researchers thought that 3D inversions could not resolute the local structure well, while 2D inversion results could agree with the surface geologic features much well and interpret the geologic structures easily. But in the present paper, we find that the result of 3D inversion is better than that of 2D inversion in identifying the location of the two local faults, the Shade Fault(SDF)and the Yunongxi Fault(YNXF), and the deep structures.
In this paper, we first studied the electrical structure of SDF and YNXF based on a measured magnetotelluric(MT) profile data. Besides, from the point of identifying active faults, we compared the capacity of identifying deep existing faults between 2D inversion models and 3D models with different inversion parameters. The results show that both 2D and 3D inversion of the single-profile data could obtain reasonable and reliable electrical structures on a regional scale. Combining 2D and 3D models, and according to our present data, we find that both SDF and YNXF probably have cut completely the high resistivity layer in the upper crust and extended to the high conductivity layer in the middle crust. In terms of the deep geometry of the faults, at the profile's location, the SDF dips nearly vertically or dips southeast with high dip angle, and the YNXF dips southeast at depth. In addition, according to the results from our measured MT profile, we find that the 3D inversion of single-profile MT data has the capacity of identifying the location and deep geometry of local faults under present computing ability. Finally, this research suggests that appropriate cell size and reasonable smoothing parameters are important factors for the 3D inversion of single-profile MT data, more specifically, too coarse meshes or too large smoothing parameters on horizontal direction of 3D inversion may result in low resolution of 3D inversions that cannot identify the structure of faults. While, for vertical mesh size and data error thresholds, they have limited effect on identifying shallow tectonics as long as their changes are within a reasonable range. 3D inversion results also indicate that, to some extent, adding tippers to the 3D inversion of a MT profile can improve the model's constraint on the deep geometry of the outcropped faults.

On 31 July 1954, an M_S7.0 earthquake occurred southeast of Minqin, Gansu Province, northwestern China. Its epicenter was located at the edge of the Alxa block, subject to northeastward compression of the Tibetan plateau, resulting in active tectonics there. Because of few records and field investigations, the seismogenic fault and tectonic setting of this event remain unclear. To probe the deep structure of this region, magnetotelluric (MT) measurements have been carried out near the epicenter, and new data of 28 sites were collected. Using the methods including the remote reference, "robust" and phase tensor decomposition, these MT data were processed, followed by NLCG two-dimensional inversion of the data to reveal the deep electrical structure of the study area. Combining with previous studies, geologic interpretation of the MT survey suggests that the Minqin earthquake of 1954 may be related to the Hongyashan-Sidaoshan Fault, which is a high-angle thrust with left-slip component. It lies between the Tibetan plateau and the Alxa block, where substantial elastic strain has accumulated due to the northeastward extrusion of the plateau, leading to occurrences of several earthquakes greater than M_S5.0 in the history. Our electrical structure derived from the MT survey supports the following tectonic interpretations:The Tibetan plateau expands to the northeast in a flower-like style while the Alxa block subducts to southwest in a listric-shaped manner, which forms the northeastward growth pattern of the Tibetan plateau. The forefront of the plateau expansion is around the Hongyashan-Sidaoshan Fault, indicating that the extension of the plateau has surpassed the Hexi Corridor to the southern margin of the Alxa block. The deformation nearby the Hongyashan-Sidaoshan Fault could be linked to the northeastward propagating extrusion of the Tibetan plateau as a far-field dynamic effect of the India-Eurasia collision. The Tibetan plateau is continuing to grow northeastward, resulting in folds and thrusts in the Hexi Corridor, and even farther to the southern margin of the Alxa block.

In 1631, an earthquake of M_S6 3/4 occurred in the Taiyangshan uplift about 10km north of Changde City, Hunan Province, which is the largest destructive temblor documented in history of South China. With the economic and social development of Changde City and the expansion of the urban, it is necessary to conduct assessment of seismic hazard, including probing the deep structure beneath the region around this historical event. To this end, three magnetotelluric(MT) profiles have been carried out across the Taiyangshan area with 76 sites in 2014. Remote reference, "robust", and phase tensor decomposition techniques were used to process the acquired MT data, and the NLCG two-dimensional inversion was made to image the deep electrical structure in combination with relevant geological and geophysical data available. The images of 3 MT profiles permit to delineate the deep extension of major faults and the deep structural features of the tectonic units in the study area. The largest fault, the Xiaowupu fault shows a steep southwest-dipping with extension of tens of kilometers from the surface to the subsurface. The Shichaipo Fault presents a low-resistivity body around a depth of about 5km. The Huanxian and Dongting Lake Basins show a low-resistivity characteristic from the ground to a depth more than 10km, good-electricity layering, meaning tectonic stability, and corresponding to extensive Cretaceous and Cenozoic strata. The electrical structure of Taiyangshan uplift overall presents a high-resistivity characteristic from the surface to a depth of about 20km, which is the widest in the central Taiyang Mountains. The deep electrical structure of 3 profiles together reveal that the contact between the Dongting Lake Basin and Taiyang Mountains is obviously segmented in NS direction. It is inferred that the Xiaowupu fault is probably the causative feature of the 1631 Changde M_S6 3/4 earthquake. The deep electrical structure nearby the epicenter appears to be complex with alternating high and low resistivity, and the epicenter is located in the high resistivity zone. The low-resistance decoupling in proximity of the fault is likely responsible for the earthquake generation. The Taiyangshan uplift resides in the southwest corner of the Jianghan-Dongtinghu Basin, where differential up and down activity during Quaternary was most intense resulting in big landform contrast, forming the tectonic setting of medium-sized earthquakes in this region.

In this paper,we propose a method of seismic prediction using the geo-electric resistivity shifting self-correlation (SSC),and a numerical test is carried out using random time series analysis to verify the validity of the method.The SSC method is applied to the actual observation data of three geo-electric resistivity stations,and results are obtained as follows:(1) SSC coefficient changes in Ganzi and Shandan stations have good correspondence to earthquake,which is represented mainly by the phased increase of correlation coefficient appearing six months to a year before the earthquake.At the same time,the correlation coefficient anomalies of the two stations also exhibit strong anisotropy.(2) Although Chengdu geo-electric resistivity station had suffered serious disturbance,the correlation coefficient anomaly also has a good correspondence with earthquake.In addition to the validity of the SSC method,it may also be attributed to the magnitude of the earthquake event,the smaller distance of epicenter,and the time of the earthquake.Anisotropy also exists in the anomaly at Chengdu station.(3) By comparing the characteristics of different magnitudes of earthquakes,the results are obtained that,when the magnitude of the selected characteristic earthquake is relatively small,the amplitude of the anomaly before earthquake is different,but when the magnitude is larger,for example M_S ≥ 5.0,the impact on the results of this study is very limited.In addition,we briefly discussed the anisotropy of seismic geoelectrical resistivity anomalies and the selection of the characteristic earthquake.

Magnetotelluric data are collected along a NW-SE trending and about 900km long profile within northeastern boundary areas of the North China craton(NCC). This profile extends from the Hegenshan belt within the Central Asian orogenic belt(CAOB), across the Baolidao arc, Solonker-Linxi suture zone, Ondor Sum accretion complex, Bainaimiao arc, Inner Mongolia paleo-uplift, Yanshan belt, and ends on the Liaohe depression of the NCC. Impedance tensor decomposition methods are used to study the dimensionality and geo-electric strike of MT data of the region. Two-dimension (2D) analysis is appropriate for this profile. The 2-D subsurface electrical resistivity structure along profile is obtained using the non-linear conjugate gradient (NLCG) algorithm. The electrical resistivity structure is characterized by lateral segmentation, and divided into high resistive, low resistive, and high resistive areas; The lateral variation of electrical resistivity is significant within the CAOB, but it is smooth in the NCC; The extensive high conductive body(HRB)is observed in the mid-low crust beneath the Solonker-Linxi suture zone and Inner Mongolia paleo-uplift, respectively; The low resistivity could be due to the partial melts and crustal flows. Based on our electrical resistivity structure and other geological, geophysical observations, we speculate that (1)the final suturing of the Siberian craton to the NCC could be along the areas between Xilinhot Fault and Xar Moron Fault; (2)the relatively thick high resistive body beneath the Yanshan belt may serve as a tectonic barrier separating the on-craton and off-craton regions into different upper mantle convection system, and lower the effect of tectonic evolution of CAOB on the destruction to NCC.

With the growing number of observational data, increasingly complex geoelectric model, and high-dimensional magnetotelluric method, the efficient and stable forward and inversion technologies become more and more required. Multigrid method enables the asymptotically optimal approximate numerical solution to elliptic partial differential equations including the MT 2D forward modeling problem. Currently, the coarsening and refining operations are based on regular grid which is easy to understand and implement. However, when the geometry of a region is complex or the region needs local refinements, regular grid becomes not well applicable. Unstructured triangular grid has better geometrical adaptability than structured grid. The Delaunay triangulation ensures that each triangle is generated to be Delaunay. This property is particularly suitable for numerical interpolation and FEM calculation. Besides, the Delaunay triangulation has a strong mathematical theory foundation which makes it very convenient for local refinement and sparseness. The key to solve the complicated geoelectric model using multigrid method is the automatic generation of unstructured multi-level grids. In this paper, we present an algorithm for generating unstructured grid for Multigrid method based on Delaunay triangulation. The algorithm will automatically generate the coarse and fine grid for complex input region, and all the triangle elements abide by the Delaunay criteria which ensure numerical accuracy and high convergence rate.

On July 22, 2013, an M_S6.6 earthquake occurred at the junction of Minxian and Zhangxian. After the earthquake, magnetotelluric(MT)measurement was carried out at 45 sites along the NE-oriented profile across the West Qinling orogen(the west segment)and the earthquake area. Remote reference, "robust", and phase tensor decomposition techniques were used to process the MT data, and the NLCG two-dimensional inversion method was adopted to get the deep electrical structures. The deep electrical structure images indicate that there exists an inverted trapezoidal high-resistivity layer in the West Qinling orogenic belt(west segment)at the depth from the surface to about 20km deep, which is shallow in the northeast and southwest and deep in the middle. Under the high-resistivity layer is a low-resistivity layer, and they conjoin each other. There is a low-resistivity layer in the Songpan-Ganzi block(north part)at the southwest side of West Qinling orogenic belt(west segment)under the depth of 20km in the lower crust, which is shallow in the northeast and deep in the southwest, and the Longxi Basin at its northeast has a stable layered structure, suggesting that West Qinling orogenic belt(west segment)is being subject to the northward extrusion of the Songpan-Ganzi block and southward resistance of the Longxi Basin. The East Kunlun Fault(Tazang segment)faulted the low-resistivity layer in the lower crust of Songpan-Ganzi block. The Diebu-Bailongjiang Fault and Guangaishan-Dieshan Fault zone extend to a shallow depth and merge into the East Kunlun Fault(Tazang segment)in the deep part. The characteristic of low-resistivity of the media in the deep-seated structures in the East Kunlun Fault(Tazang segment)is the underlying cause for the gradual decrease of horizontal slip rate and gradual increase of vertical movement of the Tazang segment. The West Qinling Fault is a main geoelectric boundary zone, which extends through the Moho; Lintan-Tanchang Fault zone behaves as a low-resistivity layer with a certain width, which extends into the low-resistivity layer in the mid to lower crust. The source region of Minxian-Zhangxian M_S6.6 earthquake locates in the core of inverted "trapezoid" of the low-resistivity layer in the West Qinling orogenic belt(west segment), that is, in the contact area between the high to low resistivity layers, and also in the low-resistivity fractured zone near the Lintan-Tanchang Fault. The interaction of southwest-northeast pushing from Songpan-Ganzi block and resistance of Longxi Basin block at its northeast is external dynamics of the Minxian-Zhangxian M_S6.6 earthquake, and the high- and low-resistivity medium property and their contact relation in the seismic source region of the earthquake are the internal factor to generate this earthquake.

The electrical conductivity of biotite-plagioclase gneiss was investigated at pressure of 1GPa and temperatures from 400 to 1073K by impedance spectroscopy within a frequency range of 10^-1 to 10⁶Hz. The electrical conductivity of the samples increases with increasing temperature, which can be described approximately as the Arrhenius equation. The logarithm of electrical conductivity varies from -6 to -0.5S/m at the range of 400 to 1073K. The measurement results reveal that the conductivities of samples with different orientations differ by an order of magnitude at the same temperature. The electrical conductivity parallel to foliation shows abrupt change in the temperature range of 881～1040K in the third heating and cooling cycle, which may be associated with the dehydration of biotite. The calculated activation enthalpies are 0.49eV(parallel to foliation)and 0.43eV(perpendicular to foliation)for low temperatures and 1.53eV(parallel)and 3.40eV (perpendicular)for high temperatures, respectively. The experiment results are compared with the magnetotelluric observations of the middle and lower crust in East China, our model is consistent with the electrical conductivity structures derived from geophysical observations. Our results indicate that the biotite-plagioclase gneiss may be one of the candidate rocks in this region.

On 20 April 2013, the Lushan M_S7.0 earthquake happened in the southwestern segment of the Longmen Shan Fault zone, which is a shock only five years after the devastating 2008 Wenchuan M_S8.0(M_W7.9) earthquake that ruptured the northeastern section of the same fault zone. At least seven years before the Wenchuan event, the most parts of this tectonic zone looked seismically quiet, without M>4.0 quakes, apparently consistent with the low slip rates along the fault evidenced in geological research and GPS measurements. The occurrence of these two major shocks has stimulated researchers to explore what is the cause, particularly the deep conditions, that made the Longmen Shan Fault zone wreak subsequent havoc within five years.To probe the crustal structure beneath Longmen Shan, a magnetotelluric (MT) profile LS6 was deployed for data collection, which runs through the southwestern fault trace and facilitates comparison with another MT profile LS04 that crosses through the Lushan seismic area to the northeast. Using the advanced methods, MT data processing and 2D inversion have been carried out on both the two profiles. The comparison shows that the profile LS6 seems to reveal a more complex electric structure of crust with respect to that from the profile LS04, though both have similarities to some extent. The formation of such a deep structure is associated with the southeastward motion of the Songpan-Ganzi block in the northwest, which is hampered by the Longmen Shan Fault zone and by the rigid Sichuan Basin farther to the southeast. It is also influenced by the motion of the Chuandian block as well as active faults, such as the Xianshuihe Fault in the southwest. The occurrence of the 2013 Lushan earthquake implies a subsequent release of the elastic strain accumulated within the Longmen Shan Fault zone, which is an independent event to the 2008 Wenchuan earthquake, though both have some relationship between each other.

Control Source Extremely Low Frequency(CSELF)electromagnetic technique is a new technique used in the earthquake monitoring. The signals transmitted by powerful generator propagate in "conductance" and cover the regions distant to thousands of kilometers. The signal frequency band contains the ELF and adjacent frequencies which were not used in earthquake monitoring by previous electric and magnetic observations and were understood for their sensitivity to the electromagnetic anomaly relative to events. The CSELF can monitor both the space electromagnetic fields and the sub-surface resistivity structures with their variations,which is favorable for four-dimensional monitoring. In the past experiments,the electromagnetic anomalies before shocks were measured twice. The experimental observation using observatory network-like stations shown that the data of artificial signals are quite stable and have high S/N ratio. A new observatory network with a certain scale is building for earthquake monitoring.

The Daixian Basin,a typical asymmetric half-graben basin,lies in the northern Shanxi Fault zone.It is bounded by the northeast trending Wutaishan northern piedmont fault zone and Hengshan southern piedmont fault zone.To investigate the shallow electric structures beneath this basin and the boundary faults,a CSAMT profile was applied which runs northwesterly through the basin.CSAMT data were recorded at 246 locations along the 12.55km-long profile.At each location data were recorded from 8533.333 to 1.333333Hz and processed to give estimates of apparent resistivity and phase.The profile is through the Wutaishan Fault in the southeast and arrives at the Hengshan piedmont loess hilly region in the northwest.Some data are of low signal-to-noise ratio due to ractive noises from power lines,pipelines,railways,highways and power substations.But useful data can be acquired 200meters away from the interference sources.The curves of apparent resistivity and phase have an obvious segmentation character.The shape of the curve of each section is consistent to the adjacent section and the segmentation shows good correlation with geomorphologic features.NLCG inversion was performed on the CSAMT data along this profile to reveal the electric structure within the depth of 1km from the surface in the basin.The result shows that the Wutaishan northern piedmont fault zone consists of a set of north-dipping normal faults which decline down stepwise,verging toward the Daixian Basin.Because the CSAMT profile does not run through the whole Hengshan southern piedmont fault zone,only partial features of it are revealed,i.e.the southeast-dipping high-resistivity foot wall bedrock and low-resistivity hanging wall alluvial deposits are underlain by thin-bed low-resistivity flood deposits.Besides,it is found that the Daixian Basin is laterally inhomogeneous,corresponding to its topography,where the tilting alluvial-flood plain is underlain by low-resistivity layer and the resistivity below the alluvial plain is slightly higher.The application of CSAMT to the Daixian Basin indicates that this method can be used to determine the location,dip and size of faults,and provide reliable basic data for survey of active faults.

Because the study in line source 1-D forward is few until now,in this paper,all-time apparent resistivity is attained by horizontal component in electrical field.Theoretical calculations indicate that the horizontal electrical field component is one-to-one correspondence with resistivity,which made the definition of all-time apparent resistivity unique and reliable.The paper uses horizontal electrical field component to define the apparent resistivity and calculate with dichotomy.The calculation example shows that the apparent resistivity defined by this method can accurately reflect the information of the model.In the following research,for time domain with line source we found that at near area there are bad results from later period data.Because of the quick electromagnetic wave spread and decay,we have bad results from early stage data in far area also,and the detection depth increases with the transmitter-receiver distance.Finally,we obtain the relationship between maximum detection depth and transmitter-receiver distance by adding 5% random noise and comparing with the theoretical results.

The preliminary research result of electromagnetic prospecting of the near field(transition region) of long grounded wire in frequency domain is presented in this paper.And the full-region apparent resistivity is defined and calculated from some modeling results including the response of half space and the response of stratified space.From the results,we get the conclusion that the full-region apparent resistivity in the above situation defined from horizontal electric field and vertical magnetic field can yield a good outcome of underground electric structure.The full-region apparent resistivity calculated by using dichotomy after joining horizontal magnetic field and horizontal electrical field together can also reflect the change of electric structure underground.The full-region apparent resistivity defined by using a single component in the near field matches well with the single dimensional MT curve,therefore,mature MT inversion technology can be used to do the inversion of the near field,thus transforming the function with source to non-source types.It is available to prospect in the near field(transition region) with long grounded wire,even there are still some pending problems.

Magnetotelluric sounding method(MT),the most important geophysical means of detecting the crust and upper mantle electrical structure,is more and more widely used.The study of the inversion depth of MT(the effective detection depth) is relatively rare,and the previous view believed that the depth of effective detection was only in association with frequency.With a large number of the inversion results of two-dimensional vertical fault models,we find that the length of the inversion profiles also has an important impact on the result.In this paper we designed a set of vertical fault models with different fault slips.Focusing on the TM polarization mode,we studied the relationship between the length of the inversion profile and the result in detail.The results show that,for a certain depth of the target area,when the bandwidth of frequency is wide enough and the length of the density of the site is appropriate,the length of the inversion profile directly influences the quality of inversion results.If the length of the MT inversion profile is short,the results cannot respond to target depth correctly,and if the length is appropriate,the results may be good.The study also found that,with the fault slip increasing,the length of the inversion profile also needs to increase,and there is a linear relationship between them.This conclusion has great significance both for the field observation and theoretical studies.

The Ordos fault block is a stable block since the Cenozoic time.Whether low-resistivity layers exist in the middle and lower crust of this kind block is an open question.This work attempts to reveal the entire crustal structure of the Ordos fault block based on interpretation of magnetotelluric data collected along the profile across this region.The result shows that a layered structure characterizes the crust of the Ordos fault block,with a low-resistivity layer at depth of about 20km,presumably associated with fluids there.On contrast,in the areas of active tectonics in the east and west of the fault block,there is no such layered electric structure in crust,while the low-resistivity zones may be related with the detachments(or ductile-shear zones) in crust.The difference in electric structure of crust between the Ordos fault block and neighboring areas is of significance to the analysis of activity and deformation in varied blocks.

The EM measurement of earthquake monitoring for M_S 6.4 Ning'er aftershock series has been done at a continuous observation site after the earthquake on June 3,2007.The data was observed for 14 days.Very strong coseismic signals of aftershocks were found in the magnetic and electric fields' time series data.Electric and magnetic auto power spectrums have been obtained by using FFT in the bihourly data.The spectrum variations for the same frequency were compared.It was found that the peak of all components bounced before and after earthquakes of M_S≥3.0.This phenomenon may be corresponding to the coseismic signal and the electromagnetic precursor before and after the earthquake.We also discussed the mechanism in the paper and the seismo-dynamo effect seems to be a plausible mechanism.

The measurement of electromagnetic field for monitoring aftershock series has been done by continuous observation after the M_S8.0 Wenchuan earthquake on May 12,2008 in Hanwang Observatory,Longnan,Gansu Province and vicinity for 22 days since May 22,2008.Two V5-2000 instruments for MT measurement made by Phoenix Company were set up 2.5km away from each other near Hanwang Observatory where a geophone for strong earthquake recording was set up.The coseismic signals of aftershocks exist in all components of the magnetic and electric fields of the two electromagnetic sites.Comparing with the seismic data at Hanwang station,the signals arrive simultaneously with the seismic waves and do not at the origin time of the earthquake.The main frequency of both seismic waves and electromagnetic signals is almost the same.The signal produced at the origin time of the earthquake seems apparent in the EM data,but the signal amplitude is much smaller than those of seismic wave arrival.

This paper focuses on the calculation of the auxiliary field-another important factor related the precision of MT 2D forward modeling.The purpose is to raise the precision of forward calculation and provide MT inversion with more sufficient supports.It suggests constructing a quadratic interpolation shape function in the surface grid for calculation of the auxiliary field,when employing the current available method of linear interpolation for calculating the principal field.In this work,the Direct Iteration Finite Element Algorithm(DIFE)is utilized to solve the calculations of interpolation nodes.This new algorithm is tested through model calculation.Comparing with the 2D forward modeling of the auxiliary field by the linear interpolation,the quadratic interpolation does not increase much amount of computation,but raises the precision of MD forward calculation.We also discussed the reason from theory,and on the basis of the result,the current design criterion for 2D forward grid has been improved.The element size can be enlarged to one skin effect depth.So the new algorithm has special meaning to the study of complex topography forward problem.We will further discuss this question and compare the results.Besides,the calculation of auxiliary field in 3D forward modeling is also very interesting.

Magnetotelluric data have recently been obtained at 9 sites along a profile of about 8km long and orientated approximately E-W in the Ning'er earthquake area which lies in southwestern Yunnan province.From these data,the high-resolution apparent resistivity,impedance phase curves,skewnesses and regional strike were calculated by using the remote reference technique and robust data processing technique,and then analyzed.The 2-D inversion method NLCG was used for data interpretation.According to the 2-D electrical structure of Ning'er seismic area,there may be an extended fault because of an obvious different electrical structure between the left & right side of the site 12.On longitudinal distribution,the crust of this area can be divided into four electrical layers:1)Mesozoic and Cenozoic sediments,about 1km,2)Paleozoic sediment,about 5km,3)high conductive layers in crust,and 4)middle and lower crust.A high conductive layer exists in the crust of the area.Between the site 12 and 13,there might be a decollement structure on top of high conductive layer under them,because the buried depth of the layer is quite different between the left & right side of the site 12.Focal depth of the Ning'er earthquake is about 5km,the decollement structure presumed in this research lies in this layer,and its strike is NW,same as that of the major axis direction of the isoseismal curve of this earthquake.So we conclude that the decollement structure is the seismogenic structure.