Transient Electromagnetic (TEM) methods are widely used in near-surface geophysical exploration. Traditional ground-based TEM systems utilize large transmitter loops (25x25 to 50x50 m²) to investigate depths between 200 and 300 meters, yielding 15-20 soundings per day. To enhance efficiency in shallower investigations (0 - 100 m), we introduce a compact TEM system with a small coil setup for rapid deployment and mobility, increasing data collection rates along extensive transects.

The novel system comprises a 3x3 m transmitter loop with two turns and an equivalently sized offset receiver loop with four turns, separated by 10 m to minimize coil coupling and ensure unbiased signals. Operating as a single-moment setup, it achieves a peak current of 5 Amp or 10 Amp, turning off in approximately 7 µs. Unbiased measurement begins at 10 µs post turn-off, extending to a late gate of 3 ms. Both transmitter and receiver are integrated into a portable unit powered by lightweight lithium-ion batteries. A dedicated mobile application for Android and iOS devices controls the system, facilitating real-time monitoring of data curves and system parameters like current and temperature.

With this system, two people can collect a 400 m profile in under 60 minutes, significantly faster than Electrical Resistivity Tomography (ERT) methods. The presentation will cover the system's layout, operational methodology, depth capabilities, and validation against the Danish National TEM Test Site. Comparative analyses will underscore its efficiency and effectiveness in aquifer layer mapping, offering a compelling alternative to traditional ground-based systems.

How to cite: Maurya, P., Auken, E., and Bording, T.: A small coil transient electromagnetic system for quick subsurface mapping , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-2809, https://doi.org/10.5194/egusphere-egu24-2809, 2024.

Coal-fired power plant requires huge storage of coal. During the storage of coal, heat is accumulated inside the stock piles and eventually results in the self-combustion or coal fire, which is a very serious problem in the fuel management and environmental aspect of the power plant facilities. To detect and forecast the coal fire, various methods had been suggested but there are no proven early warning technology until today. Since the resistivity of the coal is strongly affected by temperature, we suggested the ERT (Electrical Resistivity Tomography) monitoring technology to identify the heat accumulation inside the coal stock pile, which can eventually provide an early warning method of coal fire in the power plant facilities. To prove the technology, we prepared a small scale coal stock pile and electrodes were placed on the bottom of the stock pile. In the inside of the coal pile, temperature was continuously increased by using heating tools and ERT monitoring data were acquired for a few days until real coal fire take place on site. The whole ERT monitoring data were processed and we tried the 4D inversion to obtain changes of 3D resistivity distribution with temperature changes. In the 4D inversion results, we could identify the systematic change of resistivity values due to the heating process. Although resistivity is increased in the very early heating stage, increased resistivity is evident with the increase of coal temperature until self-combustion of coal. Therefore, we could prove that 4D ERT monitoring technology is a very promising method to detect and forecast the coal fire in the power plant facility.

How to cite: Yi, M. J., Jeong, S., and Shin, S.: Early Detection of Coal Fire inside the Coal Stock Pile by 4D ERT Monitoring, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-7179, https://doi.org/10.5194/egusphere-egu24-7179, 2024.

Controlled-source audio-frequency Magnetotellurics method (CSAMT) partially overcomes the drawbacks of weak natural field signals. However, substantial interference is an inevitable part of field surveys in practice, which negatively impacts signal quality. We require new denoising techniques since traditional techniques, such as Fourier transformation, which compute apparent resistivity directly from frequency-domain data, are insufficient in our situation. CSAMT denoising research is currently lacking, nevertheless. This research proposes the use of long short-term memory (LSTM) neural networks to denoise CSAMT signals in the time domain, given their good performance in processing Magnetotelluric (MT) data as shown by prior studies. We seek to directly extract the desired frequency signal for denoising from the time series data, in contrast to conventional denoising techniques. Since noise and target frequency signals are mixed together in MT data, the only way to suppress noise is to find the characteristics of the noise in the time series. CSAMT, on the other hand, differs from MT in that it uses an artificial transmitting source and fixes the valid signal frequency within a temporal window. This makes it possible to extract target frequency signals directly without taking into account the intricate properties of noise. In order to complete the noise suppression job, we created a neural network in this study that is based on bidirectional LSTM. This method was able to partially handle the difficulty of denoising when the data's SNR falls below 0 dB and, on average, enhance the signal-to-noise ratio (SNR) of CSAMT data by roughly 20 dB after executing both simulated and measured data testing.

How to cite: An, Z., Xu, B., Han, Y., and Ye, G.: Time-domain denoising of CSAMT data base on long short-term memory, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5979, https://doi.org/10.5194/egusphere-egu24-5979, 2024.

The Badain Jaran Desert, located in the western part of the Inner Mongolia Autonomous Region in China, distinguishes itself from typical deserts by its abundance of lakes and rich groundwater reserves. At the Badain Jaran Desert, 153 magnetotelluric sounding stations have been explored to perform one-dimensional and three-dimensional inversion analyses of the collected magnetotelluric dataset. The results of one-dimensional inversion at each sounding station, where the top interface of the first underground low-resistivity layer is less than 400 meters, were used to build a map of the potentiometric surface level of the study area. This map aligns closely with the findings from hydrological surveys. The three-dimensional resistivity model indicates the existence of a conductive layer at the deep of 2-3 km, interpreted as a sandstone-confined aquifer, located between the mountain areas surrounding the Badain Jaran Desert and its clusters of lakes. 
Moreover, there is an almost vertical conductive zone underneath the lake cluster, which is interpreted as the discharge area of the confined aquifer. This zone is related to the upward flow of deep groundwater through fractures, replenishing both the lakes and the shallow groundwater, while the surrounding mountainous regions of the desert act as the recharge areas for this confined aquifer. Finally, an estimation of the volumetric percentage of saline fluid in the confined aquifer was derived based on the electrical conductivity model of pore-fluid saturated sandstone, yielding the saline fluid content that meets the resistivity/conductivity range conditions of the confined aquifer.

How to cite: Yi, X., Ye*, G., Jin, S., Wei, W., and Zhang, Y.: Groundwater Recharge Mechanisms in the Lake Clusters of the Badain Jaran Desert and the Salinity of Confined Aquifers Based on the Electrical Conductivity Model of Pore-fluid Saturated Sandstone, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15176, https://doi.org/10.5194/egusphere-egu24-15176, 2024.

The region Lusatia in northeastern Germany, which is located about 100 km south of Berlin, is strongly affected by over a century of both former and on-going opencast lignite mining. Although, there is an abundance of borehole data from former excavation surveys both varying data quality and heterogeneous coverage is a challenge for deriving spatially continuous subsurface properties. To overcome these obstacles we combined airborne geophysical investigations with borehole data. Different machine learning-algorithms (Random Forest and K-means) are used to determine spatially and depth-related insights into the variability of geological and hydrogeological characteristics. An aeroelectromagnetic (AEM) survey was carried out in summer 2021 using BGR’s (German Federal Institute for Geosciences and Natural Resources) helicopter, which covered flight lines of 1680 km in an area of about 200 km². First results show that the machine learning approach can predict fine-grained sediments (clay and silt) in untrained areas and can distinguish between clusters of mining-affected regions and undisturbed ones. The results of the study will be further used to improve the parameterization of existing regional groundwater flow models to address challenges of water allocation in the region of Lusatia.

How to cite: Schönfeldt, E., Cortes Arroyo, O., Fahle, M., Siemon, B., Janetz, S., and Nixdorf, E.: Insights into geological and hydrogeological characteristics using airborne geophysical investigations of former opencast lignite mining areas, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15575, https://doi.org/10.5194/egusphere-egu24-15575, 2024.

At GTK we are currently developing the entire workflow of the controlled source EM (CSEM) method, ranging from data acquisition to time series processing to modelling and inversion. Part of this work is the development of a 3D modelling and inversion framework, which is mostly done within the DroneSOM project. The flexible implementation allows not only for modelling and inversion of semi-airborne drone EM data, but also land-based CSEM/MT data. The forward problem is solved using the finite element method on hexahedral meshes. We separate forward and inverse mesh using octree mesh refinement. This helps in solving the trade-off between the required accuracy in the forward modelling and computational cost. It is also a great tool to combine different multiresolution EM data (e.g., CSEM and MT) in a single comprehensive inversion framework. This work will focus on first applications of land-based CSEM and (CS)MT.

In 2022 we collected controlled source MT data using a grounded electric dipole transmitter along the Koillismaa ultra-mafic intrusion in North-Eastern Finland. Despite transmitter receiver offsets of 3-5 km far field condition does not apply for frequencies below 4kHz, which permits the use of standard MT inversion. Here we show first inversion results of these data using our new EM inversion routine taking the transmitter position into account. In addition to the active source MT data, we also collected conventional MT data on a larger scale crossing the Koillismaa intrusion. Our inversion routine also allows the inversion of MT data. We are thus showing first inversion results of joint inversion of both datasets. 

How to cite: Patzer, C., Xiao, L., and Kamm, J.: The first application of a new 3D octree finite element inversion framework to CS/MT data, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15611, https://doi.org/10.5194/egusphere-egu24-15611, 2024.

In Italy, the MARGE initiative, an abbreviation for Geoelectromagnetic Risk Map for Central Italy, strives to create an extensive map of subsurface electrical conductivity by analyzing natural electric and magnetic fields.

Led by the National Institute of Geophysics and Volcanology in collaboration with the University of Bari and the Institute of Environmental Analysis Methodologies at CNR, his project involves establishing measurement points distributed on a grid spaced approximately 50 km apart.

However, this endeavor faces significant challenges in the central region of the Italian peninsula due to extensive urbanization, numerous electromagnetic disturbances from railways and high-voltage power lines, and challenging topography, making finding suitable land parcels a complex task.

The MARGE project aims to gather broad-band and long-term magnetotelluric data, focusing on two primary objectives: utilizing magnetotelluric data to outline large-scale lithospheric structures in the Central Apennines and developing maps of the geoelectric field in Central Italy to support Space Weather modeling and critical infrastructure vulnerability analysis.

Presenting our initial findings, we discuss encountered challenges and potential solutions identified in this ongoing project.

How to cite: Pignatiello, G., Coco, I., De Girolamo, M., Di Persio, M., Giannattasio, F., Materni, V., Miconi, L., Miconi, M., Piangiamore, G. L., Romano, G., Romano, V., Santarelli, L., Sapia, V., Spadoni, S., Tripaldi, S., Tozzi, R., Siniscalchi, A., and De Michelis, P.: Mapping Subsurface Conductivity: Challenges and Progress in Italy's MARGE Project, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18165, https://doi.org/10.5194/egusphere-egu24-18165, 2024.

The uranium deposits of the Athabasca basin (Canada) represent one of the world’s highest-grade uranium resources. They are unconformity-related type at the base of relatively flat-lying sequences, where faults acted as circulation paths for hydrothermal fluids. The fault zones often contain graphitic mineralization and hence represent a valuable exploration guide of small lateral extension but detectable by electromagnetic (EM) surveys. Time-domain EM (TEM) is the method of choice for uranium exploration in the Athabasca, and taking into account the frequencies involved we can approximate the graphitization along the fault as a conductive thin plate.

To better determine the geometry of the deposit, it might be crucial to recover the subsurface resistivity and the geometric parameters of the plate (position, dip, depth, azimuth etc.). Moreover, the assessment of the uncertainty associated to the parameters can help to evaluate the reliability of geological models and to guide the subsequent drilling activities.

A quantitative approach consists of employing Bayesian inversion algorithms, which allows to exploit the prior information available. Indeed, Bayesian inversion algorithms aim to solve the inverse problem statistically returning the posterior probability density function (ppdf). In particular, they are based on the Bayes theorem, which relates the prior information (e.g. from geological and petrophysical models) with the likelihood function to assess the posterior probability density function and thus the uncertainty. We implement the Differential Evolution Markov Chain algorithm (DEMC), a multi-chain approach that integrates the Metropolis selection rule with population evolution to sample the ppdf. The chains run in parallel and each current model is updated drawing two other chains and exploiting the models at the previous iteration. After an initial stage of burn-in, the algorithm reaches the stationary regime where the chains start sampling the ppdf resulting at the end in an ensemble of models. From these models the moments of first and second order (mean and variance) are computed obtaining the uncertainty of the inverse problem solution. As forward operator we employ the LEROI forward code developed by CSIRO (AMIRA), which computes the TEM response of one or more conductive 3D thin plates embedded in a horizontally layered earth.

In this work we propose the DEMC inversion of TEM data as a tool to assess thin plates parameters and uncertainty in the context of uranium exploration.

How to cite: Vinciguerra, A., Marquis, G., Girard, J.-F., Harrison, G., and Williard, E.: Uncertainty estimation of conductive thin plates parameters through a Bayesian approach, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-19540, https://doi.org/10.5194/egusphere-egu24-19540, 2024.

In this study, we conducted an extensive geophysical survey to explore the potential of electrical resistivity methods in delineating deep ore deposits within between Koillismaa Intrusion and Näränkävaara intrusion, northeastern Finland. Preliminary investigations in 2022, including magnetic, gravity and audio-magnetotelluric (AMT) methods, along with drilling, uncovered significant anomalous structures in the survey area. Subsequent drilling of an exploration well provided positive lithological indications of a ultramafic igneous rock at more than 1.5 km depth, which are very likely of the same age as the layered intrusions in the area. Borehole data indeed revealed that the Archaean basement gneiss extends down to approximately 510 m, underlain by a granite dyke with interspersed thin layers of pyroxenite and peridotite. Notably, peridotite layers around 1500 m depth exhibited distinct magnetic and IP responses in core data.
We employed electrical methods at the site, including electrical resistivity tomography (ERT) and induced polarization (IP). To cover a large-scale area, 25 transmitter dipoles, each 1 km long and using three different transmitter systems, were deployed and data were recorded at 119 receiver stations. This work presents the acquisition and preliminary results from the ERT-IP surveys. During the processing of ERT and IP data, we utilized full time-series data recorded across the four lowest main frequencies (from 0.0625 Hz to 8Hz) to capture voltage data in a steady state. Apparent resistivity data were derived from the stacked voltage data, while IP data were initially extracted from these decay curves of these stacked voltage data and subsequently processed in the frequency domain (outphasing). Analysis of the resistivity and IP responses revealed notable IP signals at depths exceeding 1.5 km. Meanwhile, the resistivity data indicated generally very high values, around 10,000 ohm-m, with complex variations observed near the surface. This study demonstrates the efficacy of ERT and IP methods in delineating deep-seated mineral deposits, with the deep-depths IP responses being particularly noteworthy.

How to cite: Kim, B., Deparis, J., Darnet, M., Bretaudeau, F., Vedrine, S., Gance, J., Kamm, J., Autio, U., Patzer, C., and Heinonen, S.: Measuring Induced Polarization signals from deep seated magma chamber – preliminary results from a pilot survey in Finland, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-21924, https://doi.org/10.5194/egusphere-egu24-21924, 2024.