Delineating Lake Bottom Structure by TEM on Water Surface

Durusu Lake is the biggest and most important freshwater source supplying drinking water to the European side of Istanbul. In this study, electrical resistivity tomography (ERT) and transient electromagnetic (TEM) measurements were applied to detect a possible salt water intrusion into the lake and to delineate the subsurface structure in the north of Durusu Lake. The ERT and TEM measurements were carried out along six parallel profiles extending from the sea coast to the lake shore on the dune barrier. TEM data were interpreted using different 1-D inversion methods such as Occam, Marquardt, and laterally constrained inversion (LCI). ERT data were interpreted using 2-D inversion techniques. The inversion results of ERT and TEM data were shown as resistivity depth sections including topography. The sand layer spreading over the basin has a resistivity of 150–400 Ωm with a thickness of 5–10 m. The sandy layer with clay, silt, and gravel has a resistivity of 15–100 Ωm and a thickness of 10–40 m followed by a clay layer of a resistivity below 10 Ωm. When the inversion of these data is interpreted along with the hydrogeology of the area, it is concluded that the salt water intrusion along the dune barrier is not common and occurs at a particular area where the distance between lake and sea is very close. Using information from boreholes around the lake, it was verified that the common conductive region at depths of 30 m or more consists of clay layers and clay lenses.

Very low-frequency (VLF) communication signals of 15–25 kHz contaminate the transient electromagnetic (TEM) data. The notch filter or Butterworth filter is commonly used to suppress VLF noise in addition to synchronous detection of TEM data such as gating and stacking, while also introducing the TEM signal distortion around the VLF band. We propose a B-spline gating, which suppresses the VLF noise while integrating TEM data in the gate interval without any extra filter. B-spline gating is designed to ensure that the zeros of its amplitude frequency response (AFR) are located around VLF frequencies by optimizing the knot vector of the B-spline. It is available for various VLF frequencies. We apply B-spline gating to synthetic TEM data compared with rectangular gating and full-tapered gating, the AFRs of B-spline and full-tapered gating are 25 dB lower at high frequencies than rectangular gating, and the amplitude of B-spline is more reduced at VLF frequency than full-tapered gating; the standard error of the B-spline gates is reduced by >15% compared to the full-tapered gates. The field example from a TEM survey in the Inner Mongolia Autonomous Region of China shows that the amplitude of B-spline gating frequency response in late gates is reduced by about 20 dB at 24.1 kHz VLF than that of full-tapered gating, and the standard error is lower. It is concluded that B-spline gating can suppress VLF noise while following the normal decay of TEM signal without distortion.

A fast approximate three dimensional inversion scheme for interpreting ground transient electromagnetic (TEM) data is presented. The scheme relies on linear superposition of magnetic moments. The moment transform of TEM data is a time-weighted integral of the impulse response. In effect, the moment transformation converts the 3-D TEM inversion problem into a 3-D magnetic inversion problem, with accuracy traded for speed. A starting model is generated from conductivity-depth sections, and the inversion is conditioned using standard potential field inversion devices such as depth weighting.

Transient electromagnetic (TEM) data can be seriously distorted by induced polarization (IP) phenomena when a polarizable body is present. The TEM field generated by a grounded-wire source contains transverse electric (TE) and transverse magnetic (TM) modes. The IP effect is most commonly studied with the TEM total field, rather than considering the difference between TE and TM fields. To investigate the effect of IP phenomena on the TE and TM fields, we have performed a detailed analysis on IP-distorted TEM data based on numerical and field examples. We first compare the IP effect on the TE and TM fields when polarizable bodies with different polarizable parameters are present. The TM field is more severely affected by the IP effect than the TE field. Compared to a single grounded-wire source, a double-line grounded-wire source can generate a larger TM field in the horizontal electric field. We compare the IP effect on TEM data from single- and double-line grounded-wire TEM configurations, and find that the data from the double-line configuration have a higher TM/TE ratio and are more severely affected by IP phenomena than in the single-line case. Thus, it would be easier to identify and extract the IP response from field data acquired with a double-line grounded-wire source configuration. These results have been verified by a field survey of the Kalatongke copper-nickel ore district, which has predominantly layered geology, in Xinjiang, China.

In 2011, the Summer of Applied Geophysical Experience (SAGE) students acquired transient electromagnetic (TEM) sounding data at 20 sites in the Caja del Rio (CDR) area, west of Santa Fe, New Mexico to assess groundwater conditions and the potential for future geophysical work. The CDR area is reported to have geothermal potential, so TEM data were collected to constrain the structure and water quality of the local aquifer. Unfortunately, the CDR survey area also has multiple noise sources present including a power transmission substation, a series of high‐voltage transmission lines, and gas and water pipelines. It is necessary to assess the impact of the noise sources on TEM data from study areas like the CDR before inverting and interpreting the data. We analyzed the maximum amplitude of background noise relative to the measured transient decay and used the deviation of the sounding response from a power‐law decay to empirically estimate the extent of electromagnetic (EM) coupling to anthropogenic structures at each site along the profile. The theoretical effects of EM coupling as a function of distance were also calculated for an infinite line source. A correlation between potential noise sources and measured background noise was found, but no evidence of EM coupling was observed in the data, which we attribute to the conductive environment and relatively small transmitter loop size. In a more resistive environment the background noise would have been more problematic and a larger transmitter moment would increase coupling with the infrastructure. Our analysis indicates that the TEM data are valid, and for this data set the estimated signal‐to‐noise ratios are sufficient for reliable interpretation of TEM inverse models. The electrical resistivity of the CDR subsurface was modeled to a depth of approximately 150 meters. Results of two‐ and three‐layered one‐dimensional TEM inverse models are in agreement with well‐log data in the CDR area. Porosity variations along the top of the aquifer were not resolvable due to high clay content in the sediments. The results of this study leave room for further investigations of the local aquifer structure and chemistry.

The transient electromagnetic (TEM) responses measured in areas with highly polarizable media commonly show induced polarization (IP) effects, which poses difficulties to the TEM data interpretation and inversion with conventional methods. When present, the IP effects break the monotony in TEM decay curves and can even cause sign reversals, and lead to the singularity and non-monotony of inversion method. It is therefore important to take the IP effects into account when processing those TEM data. In this paper, a new inversion method is developed to invert TEM data with IP effects. The Cole-Cole model was introduced to the forward modeling solver so that the IP effects can be accurately calculated. The regularization terms were adapted to the objective function and a global optimization method, the particle swarm optimization algorithm, was used to solve the optimization problem, which weakened the singularity and non-monotony of the inversion greatly. Field data were collected in Wulong gold mine where IP effects were detected. The data were inverted using the method developed in this paper, and the inversion results were able to identify the depth extension of NO. 163 ore-bearing quartz veins and their associated fine-grained diorite veins. The global inversion method in this paper primarily solved the inversion problem of IP-affected TEM responses, and recovered the polarization parameters, which can be used as an aid to the geological interpretation.

Data processing techniques in transient electromagnetic (TEM) methods face numerous challenges, such as the problem of 3-D imaging and the demand for new data processing techniques arising from the development of small-loop TEM devices. We have developed an imaging algorithm based on gradient features derived from TEM data to address these challenges. We found that normalising the gradient field of TEM responses under appropriate background conditions enables imaging of the target. Numerical experiments conducted under theoretical models have validated the feasibility of this approach and highlighted the significance of background fields. We further designed more complex models and achieved 3-D imaging of synthetic data within seconds, demonstrating the method's high efficiency and applicability to high-dimensional data. Subsequently, we applied the algorithm to challenging small-loop TEM data and demonstrated its superior performance compared to the inversion method, which presents a novel approach for processing data from small-loop TEM devices. This study demonstrates that the algorithm possesses characteristics of high efficiency and wide applicability, suitable for TEM data from various dimensions and types of devices. Furthermore, it holds promising potential in assisting the enhancement of the efficiency of the three-dimensional inversion of TEM data.

Machine learning inversion (ML-inversion) has been widely used in the interpretation of geophysical data. However, because the transient electromagnetic (TEM) is not sensitive to the response of the small-size or large-depth abnormal body, there are some problems such as poor imaging accuracy and low-resolution when directly processing raw TEM data with ML method. To solve this problem, we propose to use the pseudo seismic wavelet (PSW) data obtained from TEM wave field transformation to perform TEM ML-inversion. Simulation and example application verify that compared with the TEM data, the PSW data has higher recognition sensitivity to the electrical changes of underground anomalies, and the ML-inversion based on the PSW data can significantly improve the TEM imaging accuracy. Our study demonstrates that the PSW data can be used to achieve TEM ML-inversion, which may provide a new way for the further combination of electromagnetic data processing and ML technology.

ABSTRACTWe investigate the early time induced polarization (IP) phenomenon in frozen unconsolidated rocks and its association with transient electromagnetic (TEM) signals measured in northern regions. The distinguishing feature of these signals is the distortion of the monotony or sign reversals in the time range from a few tens to a few hundreds of microseconds. In simulating TEM data, the IP effects in frozen ground were attributed to the dielectric relaxation phenomenon rather than to the frequency‐dependent conductivity. This enabled us to use laboratory experimental data available in the literature on dielectric spectroscopy of frozen rocks. In our studies we focused on simulating the transient response of a coincident‐loop configuration in three simple models: (i) a homogeneous frozen earth (half‐space); (ii) a two‐layered earth with the upper layer frozen; (iii) a two‐layered earth with the upper layer unfrozen. The conductivities of both frozen and unfrozen ground were assumed to exhibit no frequency dispersion, whereas the dielectric permittivity of frozen ground was assumed to be described by the Debye model. To simplify the presentation and the comparison analysis of the synthetic data, the TEM response of a frozen polarizable earth was normalized to that of a non‐polarizable earth having the same structure and resistivities as the polarizable earth. The effect of the dielectric relaxation on a TEM signal is marked by a clearly defined minimum. Its time coordinate tmin is approximately three times larger than the dielectric relaxation time constant τ. This suggests the use of tmin for direct estimation of τ, which, in turn, is closely associated with the temperature of frozen unconsolidated rock. The ordinate of the minimum is directly proportional to the static dielectric permittivity of frozen earth. Increasing the resistivity of a frozen earth and/or decreasing the loop size results in a progressively stronger effect of the dielectric relaxation on the TEM signal. In the case of unfrozen earth, seasonal freezing is not likely to have an appreciable effect on the TEM signal. However, for the frozen earth, seasonal thawing of a near‐surface layer may result in a noticeable attenuation of the TEM signal features associated with dielectric relaxation in a frozen half‐space. Forward calculations show that the dielectric relaxation of frozen unconsolidated rocks may significantly affect the transient response of a horizontal loop laid on the ground. This conclusion is in agreement with a practical example of inversion of the TEM data measured over the permafrost.

The use of ground-based electromagnetic methods (transient electromagnetics (TEM) and radiomagnetotellurics (RMT)) is increased in last decades for hydrogeological purposes. However, airborne electromagnetic methods were also proposed as a possible tool for near surface investigations due to the capability to cover large areas in short time. By jointly inverting different kinds of geophysical measurements at a site, the ambiguity inherent in different geophysical methods can be avoided. In order to couple spatial data from helicopter-borne electromagnetics (HEM) with ground-based electromagnetics, a common 1D joint inversion algorithm for HEM, TEM and RMT data is developed. The depth of investigation of HEM data is rather limited compared to time-domain EM sounding methods. In order to improve the accuracy of model parameters of shallow as well as of the deeper subsurface, the HEM, TEM, and RMT measurements are combined using joint inversion methodology. The 1D joint inversion algorithm is verified for the synthetic HEM, TEM and RMT data. The proposed concept of joint inversion takes advantage of each single method, which provides the capability to resolve near surface (RMT and HEM) and deeper electrical structures (TEM). Furthermore, the joint inversion is realized for the field data (HEM and TEM) from Cuxhaven area, Germany.

The transient electromagnetic (TEM) method is widely used in shallow surface engineering geological surveys due to its advantages such as light weight, high efficiency, and strong resolution. However, interpretation and inversion of TEM data is a complicated process. The traditional algorithm of TEM inversion employs the “smoke ring” fast imaging method, which can only reflect the approximate morphology of the stratigraphic model, and the inversion accuracy is low. Therefore, this method cannot meet the requirements of high-precision inversion. In this article, we present the particle swarm optimization (PSO) algorithm for TEM inversion. First of all, the response of the rectangular loop source TEM based on electric dipole integration was calculated and compared with the analytical solution results of the rectangular loop source and the accuracy of the algorithm was verified. Then, we introduced the solution process of the particle swarm optimization algorithm and analyzed the influence of particle swarm optimization algorithm parameter selection on the accuracy of the inversion result and the convergence speed. Next, a layered medium model was established. The particle swarm optimization algorithm and “smoke ring” fast imaging method were used to perform inversion calculation. The results show that the PSO algorithm has the advantages of high efficiency and accuracy. Finally, we examined the effectiveness of the particle swarm optimization algorithm for TEM data processing by inverting survey data from an Air-raid shelter on the campus of Chongqing University in China and comparing the results with those from the “smoke ring” fast imaging. The research works in this article provide new methods and techniques for TEM data processing.

As the previous study known, based on surfacelevel observational analysis using GPS data from Light Detection and Ranging (LiDAR) satellite imagery with a resolution of 0.9 m and Interferometric Synthetic Aperture Radar (IFSAR), it is presumed that The Lembang Fault is an active sinistral strike-slip fault located approximately $\mathbf{1 0 ~ k m}$ north of Bandung, extends for about 29 km and is divided into six segments. Previous palaeoseismological studies indicate a slip rate of $3-5.5 \mathrm{~mm} /$ year, with the potential to generate earthquakes of magnitude 6.5-7 on the Richter Scale. This highlights the importance of detailed investigations to support disaster mitigation strategies. One approach involves examining subsurface geometric patterns and resistivity distributions using Electrical Resistivity Tomography (ERT), Vertical Electrical Sounding (VES), and Transient Electromagnetic (TEM) methods. Four sites suspected of active tectonic deformation were analyzed: Batu Lonceng (Line 1), Gunung Batu (Line 2), Jambudipa (Line 3), and Cimeta (Line 4). Refers to Geological map and the Resistivity table by Telford, at Batu Lonceng, an SW-NE ERT profile revealed a SE-NW thrust fault caused by volcanic uplift ($2.195-13.611$ ohm.m-1 at 32.5 m depth). At Gunung Batu, VES data indicated volcanic uplift with high resistivity values ($857-5.143 \mathrm{ohm}. \mathrm{m}-1$) and $\mathrm{SE}-\mathrm{NW}$ thrust fault at $\mathbf{1 5 ~ m}$ depth, while TEM analysis suggested an anticline of alluvial and sandy deposits, with resistivity values of 300-700 ohm.m-1 at $\mathbf{5 0 ~ m}$ depth. In Jambudipa, interpolation of an ERT West-East profile and TEM data along the North-South direction revealed a SE-NW thrust fault in Muril Hamlet at 40 m depth ($300-400$ ohm.m-1) within volcanic lithology. Finally, at Cimeta, a North-South cross-section identified a NW-SE trending normal fault at 10 m depth ($54-113$ ohm.m-1) composed by volcanic sandstones and limestones. Overall, inversion results across all sites achieved RMS errors below 5%, demonstrating consistency with geological mapping and confirming the reliability of the integrated ERT, VES, and TEM approaches in characterizing active tectonic structures along the Lembang Fault.

Addressing problems related to water quality and quantity (e.g. irrigation, contaminant transport and vulnerability to climate change), requires detailed information about the hydrogeological framework over large areas. Airborne transient electromagnetic (TEM) is a geophysical technique capable of constraining subsurface resistivity to several hundred meters depth; the resulting sections can be interpreted in terms of the hydrogeologic framework which constrains groundwater flow. The aim of this research is to define the hydrostratigraphy of the High Plains Aquifer over a range of different geological settings, in Nebraska, USA. A challenge is to accurately characterize the airborne TEM systems in order to recover subsurface resistivity models consistent with independent ground truth information, especially in areas where targets exceed 100m depth. Several airborne TEM systems are being evaluated for their applicability to these problems. Processed data are inverted in 1D with lateral constraints to obtain a pseudo‐2D resistivity section. Recently developed model assessment techniques are applied and extended to investigate regions of the model that are most sensitive to the data. These tools provide estimates of the depth of investigation (DOI), the top of investigation (TOI) and determine the uncertainty of the resulting model parameters. The DOI, TOI and start model selection problems are solved using a sensitivity‐matrix analysis method. DOI and TOI information is important because it is possible that the measured data do not have sensitivity to certain depth intervals within the target aquifer. One result is a map showing the configuration of the top‐ and base‐of‐aquifer surfaces, in relation to DOI and TOI surfaces derived from the TEM data, together with estimates from model parameter uncertainty. Results are verified with ground‐based TEM soundings and lithologic/geophysical bore hole logs.

Summary Paper shows results of shallow and deep 3D Transient electromagnetic (TEM) surveys which were performed in 2016 in the West Siberian basin covering a depth range from 10 m to 5000 m. These electromagnetic investigations are a part of integrated research of the area that includes also 3D seismic surveys and geochemical studies. From shallow TEM studies the high-quality high-resolution data on the near-surface structure (within 500 m) was derived. Was shown that resistivity pattern has implications for the location of free gas and gas hydrates. The 3D TEM data were used to compile prediction maps of Cenomanian and Neocomian gas reservoirs, as well as deeper reservoirs in the complexly structured Achimov sequence. Obviously the next stage of research is integration of TEM, seismic and geochemical data for building a comprehensive 3D model of investigated area.

Understanding subsurface properties within urban areas is critical for an adequate management of groundwater, for instance to delineate the migration of pollutants, artificial recharge systems or geothermal collectors. Information is available from extraction wells, yet the resolution of the information is limited to the locations where wells are available. Geophysical methods offer an alternative to gain subsurface information. However, asphalted roads and limited accessibility might reduce the applicability of electrical methods for investigations beyond a few meters, whereas vibrations due to traffic and railroads might hinder the application of seismic methods. In this work, we investigate the use of the transient electromagnetic (TEM) method to resolve the geometry of aquifers in urban areas. We propose the use of relative small loops to gain separation from buried structures and increase data quality in late times as required to reach a depth of investigations of ca. 40 m. Measurements were conducted in gardens located within cities deploying single-loop as well as in-loop geometries using two different instruments. Additionally, we evaluate our small loop configuration in a quasi noise-free site through comparison to larger loops and electrical methods. Analysis of the data demonstrate that relative small loops (12.5 m x 12.5 m) may be a possible solution to gain information in urban areas down to a depth of 30 m, yet a minimal separation to anthropogenic structures of ca. 5 m is required. Information at such depth can not be easily gain with refraction seismic or electrical resistivity tomographic measurements in such small areas. Moreover, our results reveal the possibility to gain similar information with smaller loops (6.25 m x 6.25 m), offering the possibility to increase the separation to sources of noise (i.e., buried infrastructure) and increase the data quality. The inversion of TEM measurements collected along a 100 m profile permitted to obtain vertical and lateral variations in aquifer geometry with a maximal depth of investigation of ca. 40 m, while DC-resistivity measurements in the same profile were limited to less than 10 m depth. Stochastic inversion of the data permitted to investigate the uncertainty in the obtained model parameters (resistivity and thickness of the resolved layers, i.e., aquifer).