Orals: Tue, 5 May, 10:45–12:30 | Room -2.20
Submarine hydrothermal vents associated with active volcanoes provide valuable insight into the shallow fluid migration pathways and volatile transport from magmatic systems. The distribution and activity of these vents are controlled by the subseafloor structure such as faulting and zones of hydrothermal alteration. These structural features govern fluid pathways and influence long-term volcanic behavior making the location, geometry, and variability of these structural and alteration features important information for volcanic hazard assessment. Additionally, hydrothermal vents provide sources of heat and chemically distinct substrates that support benthic ecosystems. Identifying and characterizing these features aids better understanding and management of these ecosystems.
Controlled-source electromagnetic (CSEM) methods are well suited for investigating these systems due to their sensitivity to changes in electrical resistivity associated with pore fluid composition and hydrothermal alteration. Here, we present results from a surface-towed CSEM survey conducted offshore Whakaari/White Island in the Taupō Volcanic Zone, New Zealand. The survey targeted regions containing previously mapped vent sites and was designed to image hydrothermal pathways within the upper few hundreds of meters of the seafloor. Additionally, survey lines extended from the island to the area surrounding Te Paepae o Aotea/Volkner Rocks, a site of inferred structural and magmatic connectivity, to capture active hydrothermal vent sites or shallow subseafloor alteration associated with prior venting activity.
Preliminary inversions reveal electrical resistivity signatures that we interpret as zones of volcanically altered material, fluid and/or gas flow along fault structures, and individual vent features. These results provide a detailed view of near-seafloor electrical structure associated with active and relict hydrothermal processes offshore Whakaari. Our inversions complement deeper constraints on magmatic systems from ocean-bottom EM surveys and regional airborne EM data by resolving shallow conductive and resistive features linked to fluid flow and alteration. These data improve characterization of shallow subsurface structure and hydrothermal pathways in an active volcanic setting and demonstrate the sensitivity of surface-towed CSEM to vent-related processes in the shallow seafloor.
How to cite: King, R., Miller, C., and Constable, S.: Imaging Submarine Faulting, Hydrothermal Alteration, and Vent Structures Offshore Whakaari/White Island Using Surface-Towed Controlled Source Electromagnetics, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5746, https://doi.org/10.5194/egusphere-egu26-5746, 2026.
The North China Craton (NCC), the largest and oldest craton in China, has experienced multiple significant crustal growth and evolutionary events. Its northern margin adjoins the Central Asian Orogenic Belt (CAOB), forming a complex tectonic transition zone. During the Late Permian to Triassic period, the subduction and subsequent closure of the Paleo-Asian Ocean led to the formation of the CAOB, a process widely believed to have altered the composition and properties of the lithospheric mantle at the northern margin of the NCC. This study aims to use the Magnetotelluric (MT) method to investigate the impact of the subduction and accretion of the Paleo-Asian Ocean on the northern margin of the NCC and to explore the global significance of continental reworking as a deep carbon cycling mechanism.
We utilized long-period MT data from the SinoProbe NCC MT array and two MT profiles from the CAOB, totaling 114 MT stations. After data processing and analysis, we performed 3D inversions using the ModEM package. We found that the lithospheric mantle of the Yinshan-Yanshan orogenic belt in the northern NCC mainly exhibits low resistivity, and a large-scale conductor exists from the lower crust to the upper mantle in the central and northern parts of the Ordos Block, indicating that the mantle may have undergone some degree of modification. These may be related to the collision-accretionary orogenesis of the CAOB, the sulfur- and carbon-bearing sediments brought in by the subduction plate separated from the sinking plate at depth to form sedimentary diapires. In this process, carbonate melting and carbon precipitation may have occurred.
These findings indicate that the crust at the northern margin of the NCC has experienced tectonic reactivation, with the stretching and extension of the lower crust facilitating the migration and mixing of deep melts. This process likely involved crust-mantle interaction, leading to partial modification of the composition of the ancient crustal basement. The complexity of crustal composition and structure within the CAOB reflects the characteristics of the young crust of the Phanerozoic accretionary orogenic belt. We believe that these electrical characteristics are all related to the deep carbon cycling processes induced by the subduction, accretion of the Paleo-Asian Ocean, and the reworking of the NCC continent, providing important constraints on the deep physical properties for understanding the compositional evolution mechanisms and characteristics of continental crust at different stages of growth.
*This research is funded by Deep Earth Probe and Mineral Resources Exploration - National Science and Technology Major Project (2024ZD1000100-06-04) and NSFC (42074089).
How to cite: Li, Y., Zhang, L., Jin, S., Becken, M., Wei, W., and Ye, G.: The Role of the Closure of the Paleo-Asian Ocean in Lithospheric Modification of the Western North China Craton: implications from magnetotelluric array data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-752, https://doi.org/10.5194/egusphere-egu26-752, 2026.
Sedimentary basins in Northern Ireland are increasingly recognised as promising targets for low- to medium-temperature geothermal energy, yet their subsurface architecture and reservoir properties remain poorly constrained. The Lough Neagh Basin exhibits elevated geothermal gradients relative to other regions of Ireland and represents a low-enthalpy, sedimentary-basin geothermal system. The basin is known to contain sedimentary rocks ranging from at least Permian to Upper Cretaceous in age, overlain by thick Paleogene basalts. Within this system, the electrically conductive Triassic Sherwood Sandstone Group is considered the principal geothermal reservoir, although deeper Permian sandstone units may also have reservoir potential.
In this study, we assess the geothermal potential of the Lough Neagh Basin using a broadband magnetotelluric (MT) dataset comprising more than 250 MT stations, including 118 newly acquired stations as part of the GEMINI (Geothermal Energy Momentum on the Island of Ireland) project. MT data were processed to derive impedance tensors and vertical magnetic transfer functions, followed by phase tensor and induction arrow analyses to characterise dimensionality and lateral resistivity variations. Three-dimensional (3-D) MT inversion was then applied to recover the subsurface resistivity structure.
The resulting 3-D resistivity structure reveals laterally extensive low-resistivity zones with a spatially variable upper boundary, typically initiating at depths of ~0.7–1.0 km and extending to ~2–2.5 km beneath a high-resistivity basalt sequence. These zones are interpreted as the Triassic Mercia Mudstone Group–Sherwood Sandstone Group succession with variable thickness, consistent with borehole constraints and regional geological understanding. While MT primarily constrains the geometry of this conductive package, integration with petrophysical measurements from selected rock samples and other geophysical datasets (i.e., gravity and passive seismics) aids discrimination between mudstone- and sandstone-dominated intervals, enabling first-order estimates of reservoir geometry and associated heat capacity within the Lough Neagh Basin geothermal system.
How to cite: Diba, D., Kiyan, D., Hogg, C., Maggio, G., Bean, C., Reay, D., MacKenzie, M., O'Grady, M., and Cowan, M.: Magnetotelluric imaging of a low-enthalpy geothermal system in the Lough Neagh Basin, Northern Ireland, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11412, https://doi.org/10.5194/egusphere-egu26-11412, 2026.
Hekla volcano, one of Iceland’s most active stratovolcanoes, is characterised by very short eruption precursory times and a complex tectonic setting at the intersection of the South Iceland Seismic Zone (SISZ) and the Eastern Volcanic Zone. Despite its frequent activity, the geometry and connectivity of the magmatic system beneath Hekla remain poorly constrained. In this study, we use three-dimensional magnetotelluric (MT) imaging to investigate the electrical resistivity structure beneath Hekla and its implications for magma storage and transport.
This study forms part of a newly funded project under the Research Ireland Frontiers for the Future Programme, with objectives to (i) characterise the present-day structure, depth extent, and geometry of the magmatic system beneath Hekla, (ii) identify low-resistivity zones that may serve as proxies for partial melt and magma migration pathways, and (iii) assess potential magmatic interconnection between Hekla and the adjacent Torfajökull volcanic system, including the possibility of a shared deep magma source. Preliminary 3D inversion results based on MT data from 20 broad-band stations reveal two principal conductive features. Shallow (<2 km) N–S-trending conductive anomalies are observed beneath and southeast of the central edifice. These features are interpreted as groundwater-saturated fracture networks associated with regional faulting within the SISZ. At greater depths (approximately 6–24 km), a pronounced NW–SE-oriented conductive body is imaged beneath Hekla, oriented obliquely relative to the ENE-trending surface fissure swarm. This deeper anomaly is interpreted as a magma storage zone and may play a key role in controlling Hekla’s eruptive behaviour and rapid unrest development. In 2025, the MT network was expanded with data from over 30 additional stations, significantly improving spatial coverage and resolution. This contribution presents the newly acquired data and integrated 3D inversion results combining the new and existing datasets, providing enhanced constraints on the crustal-scale magmatic architecture beneath Hekla and its relationship to regional tectonic structures.
How to cite: Kiyan, D., Gudmundsson, M. T., Arnason, K., Sigmarsson, O., Hersir, G. P., Hill, G., Grayver, A., Bean, C. J., Castro, C., Kovacikova, S., Hogg, C., Maggio, G., and Molodtsov, D.: Present-Day Architecture of the Magma Plumbing System beneath Hekla Volcano from Magnetotelluric Imaging, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11889, https://doi.org/10.5194/egusphere-egu26-11889, 2026.
The borehole transient electromagnetic (BTEM) method is a key technology for metallic mineral, groundwater, and related resource exploration, with measured responses exhibiting exponential decay over multiple orders of magnitude in dynamic range. However, late-time signals attenuate to extremely low amplitudes and become submerged in noise, creating a bottleneck for exploration depth. To address the limitations of conventional processing approaches under low signal-to-noise ratio conditions, we propose a reconstruction framework that exploits the non-stationary nature and time–frequency evolution of BTEM decay curves. The framework incorporates two complementary mechanisms. A time–frequency ultra-pyramid fusion module captures the evolving decay behavior in the time–frequency domain and enables robust separation of signal and noise. In parallel, a noise-aware gating mechanism learns point-wise reliability weights from feature statistics to regulate activations, suppressing noise-dominated late-time components while retaining informative signal content. Validation on synthetic and field datasets demonstrates that the proposed approach extends the effective observation window and reliably recovers weak signals from the noise floor, thereby enhancing the achievable exploration depth for BTEM.
How to cite: Ye, Y., Zhang, C., and Yu, N.: Weak-Signal Reconstruction under Strong Noise in Borehole Transient Electromagnetics, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5214, https://doi.org/10.5194/egusphere-egu26-5214, 2026.
The advantages of well-documented and modularly designed open-source projects starts to shine when they allow for the combination of different tools to create new possibilities. We have achieved this in the last few years within the electromagnetic (EM) geophysics community.
PyGIMLi is an open-source library for multi-method modelling and inversion in geophysics. It is particularly strong in electrical resistivity tomography, induced polarization, magnetics, and seismic refraction tomography, as well as in joint inversions.
SimPEG is an open-source Python package for simulation and gradient-based parameter estimation in geophysical applications. It provides strong capabilities, particularly for modelling gravity, magnetics, direct current resistivity, induced polarization, and frequency- and time-domain electromagnetic data. Additionally, it provides a joint inversion capability. However, the analytical 1D forward modelling is, currently, limited to loop-loop configurations. Furthermore, for 3D EM modelling, it uses a direct solver with a large memory requirement.
The emsig project contains a variety of codes. One of them is empymod, a semi-analytical electromagnetic code for layered media that can model any source-receiver configuration. Another one is emg3d, a three-dimensional modeller for EM diffusion. It provides a matrix-free multigrid solver, which means that it has a comparatively low memory footprint. However, both of these codes are purely forward modelling codes, and contain no possibility for inversions.
We will present how these codes can be combined to use the forward modelling capabilities of emsig, together with the inversion capabilities of SimPEG and pyGIMli. This not only elevates all codes to create new tools in the form of SimPEG(emsig) and pyGIMLi(emsig), but more importantly, it also allows for comparisons between different frameworks. While doing these exercises, we did encounter some struggles and concepts that need to be modularized better and be improved in the future. In particular, forward modelling codes should provide easy ways to obtain the forward response as well as the (adjoint-state or analytical) gradient. Inversion codes, on the other hand, should be able to run the inversion without knowledge of the survey configuration or any of the underlying method, just with the forward responses and the gradients. These are ideas that are often not thought of when starting a new project, but they would make life much easier if they were, which is why we offer guidelines for developers to improve the modularity of future forward modeling and inversion codes.
How to cite: Werthmüller, D., Kang, S., Günther, T., Deleersnyder, W., Carrizo Mascarell, M., and Aigner, L.: The power of modularity in open-source projects, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1578, https://doi.org/10.5194/egusphere-egu26-1578, 2026.
Magnetotelluric (MT) data inverse problem is inherently characterized by strong non-linearity. Consequently its solution using the conventional gradient-based algorithm is highly dependent on the initial model. Usually, conventional inversion schemes (e.g. Non-linear Conjugate Gradient, NLCG) employ L₂ norm to form the objective function, measuring the data misfit. However, this metric often fails to capture the phase shifts and global structural features of complex response curves, causing the inversion iteration to become easily trapped in local minima. To address this challenge, we propose a novel 2D MT inversion framework based on Optimal Transport (OT) theory, introducing the Wasserstein distance (W₂) as a robust misfit measure.
We represent the MT dataset as a six-dimensional point cloud within a joint space-feature domain. In this framework, frequencies and station coordinates constitute the spatial dimensions, while the multi-modal responses—including apparent resistivity and phase for both TE and TM modes—form the feature dimensions. By incorporating coordinates and frequencies into the distance computation, the W₂ metric effectively constrains the overall morphological evolution of the response curves across both spatial and spectral domains, providing stronger geometric and topological constraints than the L₂ norm. We implemented the algorithm using the GeomLoss library and PyTorch, leveraging entropy-regularized Sinkhorn distances and Automatic Differentiation (AD) for efficient and precise gradient computation.
Numerical experiments on 2D synthetic models demonstrate that the OT-based inversion exhibits superior convergence stability compared to traditional methods, particularly under demanding conditions where the initial model significantly deviates from the ground truth. Furthermore, the proposed method maintains high resolution for deep conductive anomalies, even under significant noise levels. These results indicate that treating MT data as high-dimensional point clouds within an Optimal Transport framework provides a robust, geometry-sensitive, and innovative technical path for geophysical imaging.
This research was supported by grants from National Major Science and Technology Projects of China: Local Funds for the "Double First-Class" Initiative (924041), Deep Earth Probe and Mineral Resources Exploration (2024ZD1000200) and National Natural Science Foundation of China General Program (42474103).
How to cite: liu, X., Chen, X., Tang, Z., and Bo, Y.: Two-dimensional Magnetotelluric Inversion using Optimal Transport and Automatic Differentiation , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11849, https://doi.org/10.5194/egusphere-egu26-11849, 2026.
Magnetotelluric (MT) data inversion aims to recover subsurface resistivity model by minimizing an objective function, typically comprising a data misfit term and a regularization term (e.g., Tikhonov-style regularization). While gradients-based optimization is widely used, it is prone to local minima and highly sensitive to the initial model. In contrast, nonlinear stochastic algorithms explore a broader solution space but are computationally prohibitive for large-scale MT problems.
Deep learning (DL) has emerged as a powerful alternative. Unlike purely data-driven methods, physics-driven DL frameworks embed physical laws, such as wave propagation equation and Maxwell’s equation, as constraints, enhancing interpretability and reducing the need for massive datasets. Implicit neural representation (INR) is a novel physics-driven technique that represents physical properties as continuous functions of spatial coordinates. A key advantage of INR is its inherent ‘frequency principle’ (or spectral bias), where the network learns large-scale (low-frequency) structures before fine-tunning high-resolution (high-frequency) details. In MT inversion, this bias acts as an implicit regularization, improving stability without requiring manually tuned penalty terms.
In this paper, we propose an INR-based 2D MT inversion algorithm. Synthetic tests on block and layered models demonstrate that the proposed method recovers anomalous boundaries with higher resolution than traditional Occam-based inversions. Finally, application to the COPROD2 field dataset confirms the practical robustness of the approach and its potential for extension to three-dimensional, mesh-free inversions.
How to cite: Wang, Y., Garcia, X., Attias, E., Guo, Z., and Zhang, B.: Implicit neural representation for Magnetotelluric inversion , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15114, https://doi.org/10.5194/egusphere-egu26-15114, 2026.
The Ni–Cu sulphide mineralisation investigated in this study is hosted within a steeply dipping, differentiated ultramafic sill composed primarily of tremolite schist and lenticular bodies of harzburgite. The harzburgite contains net-textured sulphides and occasional massive sulphide veins, with mineralisation dominated by pyrrhotite, chalcopyrite, and pentlandite. Physical property measurements indicate that the mineralised sill is both more conductive and magnetically susceptible than the surrounding country rocks, making it a viable target for geophysical inversion.
This study integrates airborne magnetic data, ground and downhole electromagnetic (EM) surveys, and audio-frequency magnetotelluric (AMT) data to map discontinuous sulphide-rich harzburgite bodies. Magnetic and EM modelling techniques were used to delineate high-conductance zones, which were integrated with geological models to highlight potential extensions of mineralised pods. AMT inversion further confirmed the geometry of the mineralised intrusion. The results demonstrate the value of combining multiple geophysical methods to guide drilling and improve targeting of discontinuous sulphide-rich bodies within ultramafic systems.
How to cite: Mapuranga, V. and Ushendibaba, M.: Mapping discontinuous sulphide-rich harzburgite bodies using geophysical inversion techniques: A case study from a Ni–Cu deposit, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12604, https://doi.org/10.5194/egusphere-egu26-12604, 2026.
Magnetotellurics (MT) images electrical conductivity, a property influenced by fluids, alteration, graphite, and partial melt. Because many different subsurface configurations can explain the MT responses observed in the field, the preferred subsurface conductivity model estimated through inversion is often ambiguous. A central challenge is that a robust stochastic model exploration in 3D, while conceptually the best way to expose such ambiguity, is too computationally demanding to be used as a standard practice. Consequently, most studies report a single deterministic model, which obscures the range of alternative geological structures that are equally consistent with the data.
In this work we propose a practical approach to stochastic model exploration in 3D MT inversion. Drawing inspiration from annealing methods, we adopt a Very Fast Simulated Annealing (VFSA) framework, with sparse parameterization that makes the optimisation feasible for large-scale problems. Our inversion algorithm is built around the widely used ModEM forward solver, ensuring compatibility with existing workflows. Instead of producing one definitive model, the workflow can generate a set of plausible models that explain the data equally well. From this ensemble one can compute statistical summaries: a mean model that captures the most consistent structures and quantitative measures of variability that highlight where geometry, depth, or connectivity remain uncertain. This representation enables geologists to make interpretations while being explicitly aware of the uncertainty inherent in the inversion. We demonstrate that the approach works at regional scale using a large-area subset of USArray MT data from Cascadia. This dataset has been extensively studied and previously modelled with deterministic 3D inversion, which allows us to benchmark our results.
How to cite: Mishra, P. K., Kamm, J., Patzer, C., Autio, U., and Sen, M. K.: Building uncertainty-aware subsurface models with 3D magnetotelluric inversion, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4367, https://doi.org/10.5194/egusphere-egu26-4367, 2026.
Posters on site: Wed, 6 May, 08:30–10:15 | Hall X2
With rising sea levels and frequent heavy rainfall, water retention structures such as dikes become increasingly susceptible to failure. Therefore, there is a need to monitor such structures in a time and cost-efficient manner, and sufficient resolution. Electromagnetic geophysics could potentially be used for this purpose, since instability features such as water underflow or animal paths change the electrical conductivity inside the dike. In this project, we study if the Radiomagnetotelluric method can be used to investigate dikes for instability features. First, we perform forward and inverse simulations with a synthetic dike model to verify the feasibility of the method for the purpose of detecting dike instabilities. Next, we will define suitable frequencies and measurement setups through a synthetic survey design study. Subsequently, we will apply the newly acquired knowledge from the synthetic experiments to conduct a Radiomagnetotelluric survey on a dike in the Netherlands. Dikes are often situated in populated areas, which means that significant cultural noise must be accounted for during the data processing. Through 2D and 3D inversions we will generate resistivity models of the Radiomagnetotelluric data and compare them to a resistivity model obtained from ERT data to validate the results. The final objective is to evaluate the potential of the method for practical dike monitoring and give recommendations for its implementation.
How to cite: Vanoli, R., Werthmüller, D., Weckmann, U., and Rulff, P.: The Potential of Radiomagnetotellurics for Detecting Dike Instabilities, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-396, https://doi.org/10.5194/egusphere-egu26-396, 2026.
A High Temperature Aquifer Thermal Energy Storage system will be installed on the TU Delft campus, The Netherlands, utilising an aquifer at 120-180 m depth to store hot water (>60 °C). For optimal recovery efficiency, it is essential that the injected water stays near the injector well until it can be extracted, and that there is no mixing with cold water. By monitoring the spread of injected water using Controlled-Source Electromagnetics, the heat transport in this subsurface storage system could be better understood. For this, a surface-to-borehole survey design was proposed, sensitive to the resistivity at the depth of the aquifer through the vertical component of the electric field. 3D time-lapse data will be acquired using receivers inside a monitoring borehole and horizontal electric dipole sources surrounding the site. Optimal source positions will be determined through a feasibility study, using forward-modelling and inversion of synthetic data. By relating the resistivity of the water to its temperature, this data can be used to image hot plume propagation over time. However, the relationship between conductivity and temperature is expected to vary between sites and for large temperature ranges. Laboratory measurements will therefore be performed, uncovering site-specific resistivity-temperature relationships to improve the interpretation of Controlled-Source Electromagnetic data. Lastly, a future research interest lies in the combination of monitoring data with thermodynamic modelling. Through workflows of data-assimilation or process-based inversion, these methods could complement each other, leading to an enhanced understanding of heat flow in the aquifer.
How to cite: van Noordt, S., Drijkoningen, G., Daniilidis, A., and Rulff, P.: Surface-to-borehole Controlled-source Electromagnetics for monitoring temperature changes in a geothermal aquifer, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-785, https://doi.org/10.5194/egusphere-egu26-785, 2026.
Southeastern Korea is transected by the Yangsan Fault System, a major fault network comprising numerous faults developed around the Yangsan Fault. In September 2016, a ML 5.8 earthquake occurred in the Gyeongju area along the Yangsan Fault, with a focal depth of ~12–16 km. This event represents the largest instrumentally recorded earthquake in South Korea, which is located in an intraplate tectonic setting. Given the complex fault geometry and stratigraphy of the Gyeongju area, imaging the deep subsurface structure is essential for understanding the seismotectonic framework.
We conducted three-dimensional (3D) inversion of magnetotelluric (MT) data from 120 sites in the epicentral area and interpreted the resulting geoelectrical structure. The resistivity model delineates high-resistivity zones corresponding to granitoids and volcaniclastic rocks, whereas low-resistivity zones are consistent with sedimentary rocks and Quaternary alluvial deposits. This correspondence indicates that the geoelectrical structure reflects regional lithologic variations. At depths corresponding to the hypocentral range, a pronounced low-resistivity anomaly is resolved, and the 2016 Gyeongju earthquake is interpreted to have occurred beneath this conductive body. These results indicate that the deep geoelectrical structure beneath the Gyeongju area is characterized by distinct conductive and resistive features. In future studies, a multidisciplinary approach is needed for a reliable, integrated interpretation of the seismotectonic framework.
How to cite: Kim, K., Oh, S., Kwon, H.-S., Lee, S. K., and Ryu, K.: Geoelectrical structure beneath the epicentral area of the 2016 Gyeongju earthquake, southeastern Korea, from magnetotelluric data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4681, https://doi.org/10.5194/egusphere-egu26-4681, 2026.
This work concerns the construction and testing of a 3D numerical model suitable for the simulation of magnetotelluric measurements using COMSOL Multiphysics. The aim of the study is to inspect the effect of varying relative magnetic permeability of rocks (different from the general approximation of 1) and integrate it to geology based numerical models.
Experiences from field magnetotelluric (MT) measurements indicate that in the vicinity of igneous rocks field MT data shows increased resistivities at lower frequencies. This effect may be connected to the rocks’ magnetic susceptibility/permeability which differ from these parameters of vacuum and are generally neglected (e.g. Li & Cao 2005, Kiss et al. 2023). The presence of media with relative magnetic permeability higher than 1 may affect the depth estimation during data processing and interpretation (Kiss et al. 2023). In this study, we investigate these effects in detail applying a model to a real geological research area in Hungary to find possible explanations of the behaviour of measured MT data.
First, 1-, 2-, and 3-layer synthetic models were created to test several model parameters and verify numerical results with the analytical solutions of the respective models. Parameter tests included the size of the model, the frequency range, the resistivity and thickness of the “air” domain, the resistivity contrast of the layers, and the mesh resolution. As a result, numerically accurate models were gained: the difference between the numerical results and the analytical ones were less than 0.05% for the apparent resistivity and less than 0.3% for the phase.
Second, the effect of the relative magnetic permeability was studied. Values were chosen to range from 1 to 10. Larger values allowed us to examine the effect of the physical phenomenon, while smaller values provided information on the potential extent of the effect in reality. In the model 11 simulated measurement points were distributed along a line, 1 km apart from each other. Several models were created where the boundary of the regions with different permeability values was perpendicular to the “measurement line” or was located at an angle to the line.
Third, a model was built based on the real geology of a study area near Székesfehérvár, Hungary, where the Pre-Cenozoic basement consists of granitoid plutons and siliciclastic and carbonate formations covered by Miocene sedimentary rocks. Preliminary results of real magnetotelluric field measurements and the modelled ones were compared. It was established that the numerical model results harmonize with field observations, but further refinements are needed.
References:
X.M. Li & J.X. Cao (2005): A study on the influence of magnetic susceptibility on MT response. Chinese Journal of Geophysics, 48(4):1017-1021
Kiss, L. Szarka & E. Prácser (2023): Magnetic distortions in magnetotellurics: Predictable distortions in classical processing MT procedures in presence of a magnetic medium based on 2D direct modelling results (in Hungarian). Hungarian Geophysics, 64(1):43-57
How to cite: Szebenyi, R., Galsa, A., and Kiss, J.: 3D numerical model for investigating the effect of relative magnetic permeability in magnetotellurics, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5546, https://doi.org/10.5194/egusphere-egu26-5546, 2026.
Geophysical inversion algorithms can be categorized into gradient-based quasi-linear inversion and stochastic-search-based fully nonlinear inversion. Quasi-linear methods, such as gradient descent and conjugate gradient, rely on the gradient information of the objective function with respect to the model parameters for updates. These methods are prone to getting trapped in local minima and lack quantitative methods for evaluating inversion results. Fully nonlinear inversion methods, such as genetic algorithms and Monte Carlo methods, search the solution space through various stochastic processes, offering the advantage of avoiding local minima. However, they incur excessively high computational costs in 3D scenarios, making them difficult to apply to field data at present. We propose the Levy Gradient Descent (L-GD) method based on the Levy flight process and explore its performance in 3D magnetotelluric (MT) inversion.
The Levy flight process, derived from the Levy distribution introduced by the French mathematician Paul Lévy, is a random walk process that combines high-frequency small steps with low-frequency large steps. The heavy-tailed characteristic of Levy flights demonstrates unique advantages in escaping local minima and extensively exploring the solution space, making it highly suitable for solving large-scale complex optimization problems. This makes the L-GD algorithm a highly promising semi-stochastic search-driven algorithm. We have implemented the 3D MT L-GD inversion algorithm based on the open-source ModEM framework. Building upon this existing codebase, we integrated an inversion module based on the L-GD algorithm. Furthermore, we accelerated the forward modeling process using the cuBiCG solver developed by Dong et al. (2024) on GPU, transforming it into a practical 3D MT inversion method. Additionally, traditional methods typically rely solely on the RMS data misfit to evaluate the final inversion result. To provide model evaluation criteria beyond data misfit, we calculate statistical information for models along the search path, obtaining the mean and standard deviation of all models during the inversion iterations. This statistical information is essentially a weighted combination of gradients, reflecting the characteristics of the marginal distribution for each parameter in the high-dimensional solution space, thereby providing crucial indirect information for the reliability analysis of the optimal model.
We conducted a series of synthetic and field data tests on the L-GD algorithm. The results indicate that this algorithm can achieve better data misfit compared to the NLCG algorithm, and the model statistical information provides intuitive reference for evaluating the optimal model. Furthermore, our experiments demonstrate that for semi-stochastic inversion algorithms like L-GD, the traditional cooling method—which gradually reduces the regularization factor of the model constraint term during iterations—is not conducive to obtaining better inversion results. Instead, fixing the regularization factor at a small, low value proves to be a superior strategy.
This research was supported by grants from National Major Science and Technology Projects of China: Deep Earth Probe and Mineral Resources Exploration (2024ZD1000200) and National Natural Science Foundation of China General Program (42474103).
How to cite: Tang, Z., Yang, B., and Zhang, Y.: 3D Magnetotelluric Inversion Using Levy Gradient Descent Scheme, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6947, https://doi.org/10.5194/egusphere-egu26-6947, 2026.
The Aeolian Archipelago (southern Italy) hosts several active volcanic islands characterised by widespread hydrothermal manifestations and represents a strategic natural laboratory for assessing geothermal resources in insular volcanic settings. In such environments, Audio-Magnetotelluric (AMT) surveys are particularly effective for imaging shallow geothermal systems, as electrical resistivity is highly sensitive to fluid content, temperature, and hydrothermal alteration. Moreover, AMT data provide key constraints on major structural discontinuities that control fluid circulation at depth.
This study presents the results of a large-scale geophysical investigation conducted across the islands of Lipari, Salina, Vulcano, and Panarea, with the aim of providing a geophysically constrained assessment of geothermal resources at the scale of the entire archipelago. The focus is on the application of three-dimensional AMT resistivity imaging, which allows detailed characterization of subsurface electrical properties and offers insights into hydrothermal alteration patterns, fluid pathways, and the structural framework relevant for geothermal exploration.
The resulting 3D resistivity models reveal marked lateral and vertical variability in subsurface electrical structure among the investigated islands. These differences reflect contrasting degrees of geothermal system development, variations in fluid circulation regimes, and the role of island-specific structural controls. Rather than indicating a uniform geothermal architecture across the Aeolian Archipelago, the models highlight distinct resistivity patterns and geothermal signatures for each island, emphasizing the heterogeneity of volcanic and hydrothermal processes at the regional scale.
Overall, this study demonstrates the effectiveness of AMT resistivity imaging in discriminating between different geothermal settings and in identifying structural controls on fluid circulation in complex volcanic island environments. The results provide a robust geophysical contribution to the construction of an archipelago-scale inventory of geothermal resources and establish a solid basis for future integration with geological, geochemical, and complementary geophysical datasets. This integrated approach is essential for informed evaluation and modelling of sustainable geothermal exploitation scenarios in the Aeolian Islands
How to cite: Troiano, A., Di Giuseppe, M. G., Isaia, R., Boni, P., De Paola, C., Fedele, A., Gonzalez, G., Pagiara, F., and Sposato, M.: 3D Audio-Magnetotelluric Imaging of the Aeolian Islands for Geothermal Resource Assessment, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17202, https://doi.org/10.5194/egusphere-egu26-17202, 2026.
MARE2DEM (Modeling with Adaptively Refined Elements for 2-D EM) is a freely available 2-D inversion code originally developed by Dr. Kerry Key for marine electromagnetic (EM) surveys. Since the code is open-source, other researchers have extended it to new domains. This demonstrates the flexibility of MARE2DEM for different survey environments beyond its initial marine focus.
In this study, we adapted MARE2DEM to invert semi-airborne and full airborne EM data. The semi-airborne configuration consists of a grounded electric or magnetic dipole transmitter on land with magnetic field receivers flown in air (operating in the ~1–12 kHz band). The inversion results for the semi-airborne case successfully imaged all the detailed structures in the synthetic resistivity models when the recording locations are within 15 m height, confirming that MARE2DEM performs robustly for this new application. Such semi-airborne surveys are well-suited for our targets, such as quick clay deposits and mineral deposits.
By contrast, the full airborne EM case proved challenging. We applied MARE2DEM to a helicopter-borne frequency-domain system (NGU’s “Hummingbird” system) operating at five frequencies (approximately 880 Hz, 6.6 kHz, 34 kHz in horizontal coplanar, and 980 Hz, 7 kHz in vertical coaxial mode) with 5-6 m transmitter–receiver separations. In this scenario, where both transmitter and receivers are airborne and moving, we encountered inaccuracies in the forward modelling. High-frequency airborne EM data are particularly numerically demanding to model it accurately, as the free-space transmitter geometry and high frequency range require very fine discretization. We found that the standard adaptive meshing in MARE2DEM needed further refinement to capture the decaying fields in air. To improve the forward accuracy, we tested a range of strategies, including using more finely discretized meshes and carefully tuning the wavenumber sampling for the 2.5D solver. These measures reduced the modelling errors, but we still did not reach the same level of accuracy for full airborne modelling as in the marine or semi-airborne cases. The results indicate that additional developments are required for full airborne EM data modelling and inversion. In summary, our extension of MARE2DEM works well for semi-airborne EM surveys, achieving resolution comparable to the original marine applications, whereas the full airborne case remains problematic in forward modelling. Further improvements are being explored to enable reliable forward modelling and inversion of full airborne EM datasets.
How to cite: Wang, S., Yanev, M., and Baranwal, V. C.: Extending MARE2DEM to Semi-Airborne and Full Airborne EM Inversion: Synthetic Validation , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21012, https://doi.org/10.5194/egusphere-egu26-21012, 2026.
The Seival Mines comprise a group of open-pit copper mines that have been inactive for more than 60 years, located in the municipality of Caçapava do Sul, southern Brazil. In the region, volcanic, subvolcanic, and volcaniclastic rocks of the Hilário Formation crop out; this formation belongs to the Bom Jardim Group, a unit of the Camaquã Basin (450–620 Ma). The Seival area hosts hydrothermal copper sulfide deposits within volcanic rocks of shoshonitic affinity, emplaced in brittle zones controlled by structures with predominant N–NE and NW trends. Enrichment in Cu, Au, and Ag within andesitic dikes suggests that intrusions acted as important conduits for the hydrothermal processes responsible for metallogenesis. Despite Brazil’s high potential for the discovery of new mineral deposits, mineral exploration faces increasing challenges related to the growing scarcity of economically viable outcropping deposits. This context justifies the development of more detailed geological and geophysical studies, combined with the adoption of new strategies and subsurface investigation technologies. In this study, the Transient Electromagnetic (TEM) geophysical method was applied using the WALKTEM 2 system (ABEM). Surveys were carried out in a centre-loop configuration, with a 100 × 100 m transmitter loop and an RC-200 receiver antenna (10 × 10 m), operating in dual-moment mode. Three data profiles were acquired in the study area, totaling 14 electromagnetic soundings spaced at 250 m intervals. Data processing and inversion were performed using the SPIA and Workbench software packages. The resulting electrical resistivity sections reveal significant lateral and vertical variations, with resistivity values ranging approximately from 10 to 2000 Ω·m. In general, an upper unit is characterized by resistivity values exceeding 1000 Ω·m, associated with volcanic rocks of the Hilário Formation, whose thickness varies along the profiles. At depths of around 400 m, subhorizontal layers with resistivity values below 60 Ω·m and thicknesses of up to 200 m are observed. This horizon is possibly related to sandstones and conglomerates of the Maricá Group, the basal unit of the Camaquã Basin. The low-resistivity layer at this interface is a clear indication of the presence of conductive materials, which in the regional context are represented by sulfides hosted within a porous interface during the ascent of hydrothermal fluids responsible for the genesis of the Seival deposits. The identification of this potentially mineralized interface highlights the regional potential for deep copper deposits, based on a relatively rapid and versatile geophysical diagnosis.
Keywords: copper, mineral deposit, transient electromagnetics, resistivity
How to cite: Barros, M., Moreira, C., Masquelin, H., Ilha, L., Kumaira, S., and Kuranaka, D.: Prospecting Deep Targets Potentially Mineralized with Copper South of the Seival Mines, Camaquã Basin, Southern Brazil, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2811, https://doi.org/10.5194/egusphere-egu26-2811, 2026.
We present a three-dimensional (3D) geological model of the south-eastern Gangwon Province, Korea, using integrated geophysical datasets. The study area has historically hosted numerous coal and metal mines, many of which are now abandoned. Recently, renewed interest in critical minerals has driven active mineral exploration. In addition, the area has been selected as a potential site for underground research laboratory for high-level radioactive waste disposal and cosmic particle research.
In this study, audio-magnetotelluric (AMT) data acquired at grid-distributed measurement stations were analyzed. A total of 49 AMT survey stations were deployed in a generally non-uniform grid configuration. Measurement stations were densely distributed at approximately 100 m intervals around the candidate site for underground research facility, while station spacing increased to approximately 250–350 m toward the outer areas. For remote reference measurements, a remote reference station was operated at a distance of approximately 12–14 km from the survey area.
Three north–south profiles, three east–west profiles, and additional virtual two-dimensional profiles oriented in the northeast–southwest and northwest–southeast directions were defined. Two-dimensional inversion was performed along these profiles to obtain a preliminary geological model for subsequent 3D inversion. These models, together with two-dimensional inversion results derived from electrical resistivity survey lines, were used as initial models to facilitate faster convergence in 3D inversion. Furthermore, near-surface electrical resistivity obtained from inversion of electrical resistivity data was incorporated into the initial model to minimize static effects caused by by near-surface resistivity inhomogeneities.
The geological model derived from the 3D inversion was validated through comparison with airborne magnetic survey data and available borehole information from the study area. Continuity of geological features, including coal seams and lithological boundaries between limestone and granite, was confirmed on individual two-dimensional sections, leading to the construction of a new 3D geological model. The developed 3D resistivity model provides a reliable geophysical framework for mineral exploration and for site characterization of underground research facilities in this region.
How to cite: Lee, S. K., Kwon, H.-S., Kim, K., Oh, S., Han, M., and Jeong, S.: Three-dimensional resistivity model of the southeastern part of Gangwon Province, Korea, using integrated AMT and electrical resistivity surveys, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6356, https://doi.org/10.5194/egusphere-egu26-6356, 2026.
The Xiaojiang fault is a large active fault and a strong earthquake zone with strong tectonic motion. Located near the middle zone of the Emeishan Large Igneous Province and as the southeastern boundary of the Tibetan Plateau, it plays an indispensable role in the lateral escape mechanism of plateau material and regional tectonic evolution. Previous studies have shown that the Xiaojiang fault has low velocity, high Poisson's ratio, low electrical resistivity, high heat flow, and strong attenuation, while the Emeishan Large Igneous Province inner zone has high velocity, high electrical resistivity, and low heat flow. The location and morphology of channel flow are still inconclusive. The magnetic susceptibility structure obtained from regional magnetic anomalies can directly reveal the deep magnetic structure, while the thermal structure is direct evidence to support the material transport. This report aims to study the thermal structure of the Xiaojiang fault and its surrounding areas. Combined with the previous research to identify heat flow channels, determine the possible places of crust-mantle material exchange, clarify the material transport mechanism, and explore the regional seismic hazard. This report aims to provide a diversified supplement to the deep material structure and material transport mechanism in this region, and to understand the deeper geoscientific significance of the Xiaojiang fault and its surrounding area.
How to cite: Dong, C.: The Deep Material Structure of the Xiaojiang Fault and Its Surrounding Areas: Based on Magnetic Anomaly Model and Magnetotelluric Data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8629, https://doi.org/10.5194/egusphere-egu26-8629, 2026.
The horizontal magnetic inter-station transfer function (hereinafter referred to as M) establishes the relationship between horizontal magnetic fields at the local site and the reference site (Hl = MHr). 3D numerical simulation (Wen et al., in revision) indicates that M can effectively mitigate the "speckle-shaped artifacts" in shallow structures often observed in the inversion of long-period sparse arrays, and improve the recovery effect of the lateral boundaries of high-conductivity anomalies. Incorporating M data from a synchronized sub-array into the inversion, alongside the traditional impedance tensor (Z) and tipper (T), holds significant potential for enhancing the resolution of shallow electrical structures.
To verify the above conclusions and evaluate the practical application of M in complex geological environments, we conducted imaging of field data containing M data in the Bayan Obo area, Inner Mongolia, China. This region is located in the transition zone between the Yinshan Orogenic Belt and the North China Craton, characterized by significant lateral heterogeneity, making it an ideal site for testing resolution characteristics. We deployed an MT array with varying station spacings. While constructing a high-resolution reference model using the inversion of the dense station dataset, we also conducted inversion experiments on the sparse station data from multiple subsets of this dataset. By comparing the results of Z+T inversion with Z+M joint inversion, we verified and quantified the resolution capability of M for shallow electrical structures. Furthermore, by selecting sites with different electrical structures within the survey area as reference stations, we tested the impact of the subsurface electrical structure beneath the reference station on the resolution capability of M.
This research was supported by grants from National Major Science and Technology Projects of China: Deep Earth Probe and Mineral Resources Exploration (2024ZD1000200) and National Natural Science Foundation of China General Program (42474103).
How to cite: Chen, D., Yang, B., Wen, G., and Wu, X.: Three-dimensional magnetotelluric imaging including inter-station transfer function in Bayan Obo region, Inner Mongolia, China, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7264, https://doi.org/10.5194/egusphere-egu26-7264, 2026.
The Bakony-Balaton Highland Volcanic Field (BBHVF) is one of the most geodynamically complex regions within the Pannonian Basin. Based on legacy telluric and magnetotelluric surveys, it has been hypothesized that regions of high electrical conductivity dominate at crustal depths beneath this Neogene volcanic field. Several theories have been proposed to explain their existence, ranging from the localized presence of highly conductive graphite to the role of deep fluid upwelling. The objective of this study is to analyze newly available long-period magnetotelluric datasets using 2D and 3D inversion techniques and to model the Bouguer gravity anomalies of the area by 3D forward density modeling method. Furthermore, we analyze the information directly derivable from MT response functions to draw fundamental geological and petrophysical conclusions. This investigation aims to fill a gap by reviewing earlier geophysical interpretations that are still debated. The goal is to see which geological explanation is more likely, the presence of a specific solid component with high conductivity or the flow of deep fluids linked to the Neogene volcanic field.
How to cite: Rubóczki, T., Vozár, J., Pánisová, J., Lange, T., Enikő, B., and Berkesi, M.: Analysis of Magnetotelluric and Gravity Anomaly Datasets in the Bakony-Balaton Highland Volcanic Field, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10787, https://doi.org/10.5194/egusphere-egu26-10787, 2026.
Low-frequency electromagnetic sounding, particularly magnetotellurics (MT) and audio-magnetotellurics (AMT), is widely applied to characterize subsurface resistivity. In operational settings, however, impedance estimation is frequently constrained by cultural noise, power-line harmonics, grounding artifacts, and instrument drift, which can substantially degrade data reliability. Under such conditions, automated selection of usable time windows and principled weighting of frequency bands become critical determinants of estimation stability. This literature review examines methodological approaches to window selection and frequency-band weighting in MT/AMT processing, focusing on coherence- and stability-based quality metrics and decision rules that span hard rejection and soft weighting. The review also discusses how these quality control procedures interact with transfer-function estimation workflows and reporting practices. Emphasis is placed on reproducibility and the transparent specification of quality control criteria, while summarizing commonly reported applicability conditions, limitations, and failure modes in the existing literature.
How to cite: Liu, C.-Y.: Automated Window Selection and Frequency-Band Weighting for Reliable Magnetotelluric and Audio-Magnetotelluric Impedance Estimation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2095, https://doi.org/10.5194/egusphere-egu26-2095, 2026.
Identifying geomagnetic pulsations is crucial for monitoring and issuing early warnings of space weather hazards that threaten modern technological infrastructure. This study examines the relationship between variations in magnetotelluric data quality and geomagnetic disturbances, aiming to develop an early space weather warning system utilizing existing magnetotelluric station networks. Magnetotelluric surveys measure natural electromagnetic signals to study subsurface structures, but signal quality in the 0.1–10 Hz frequency range—known as the dead band—is typically poor during geomagnetically quiet periods due to weak source fields. Geomagnetic disturbances significantly enhance signal strength in this band, markedly improving measurement quality. We developed a parameter called the linearity ratio, based on electromagnetic field linearity, to automatically detect geomagnetic disturbances. Additional metrics including phase differences and energy are used to comprehensively monitor data quality changes. Compared to traditional methods, this approach requires minimal computational resources, enables real-time monitoring, directly utilizes existing magnetotelluric infrastructure, and provides physically meaningful indicators. Analysis of two representative events using data from the Memambetsu station in Japan demonstrates that this method reliably identifies geomagnetic activity with minute-scale temporal resolution. Geomagnetic storms comprise initial, main, and recovery phases, with the main phase posing the greatest threat to critical infrastructure. Our approach complements existing geomagnetic storm forecasts by detecting the initial phase, providing early warning before the most hazardous conditions develop. This offers a cost-effective space weather monitoring solution by repurposing established ground-based electromagnetic observation networks.
How to cite: Chen, H.: Identifying Pc2 Geomagnetic Pulsation through Magnetotelluric Data Quality Analysis in Dead Band Frequencies, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1474, https://doi.org/10.5194/egusphere-egu26-1474, 2026.
The Eastern Himalayan Syntaxis (EHS) has undergone intense crustal shortening, giving rise to spectacular topography in the Tibetan Plateau. Several tectonic models have been proposed to explain its deformation, including crustal-scale folding, indenter corner dynamics, and tectonic aneurysm mechanisms. Understanding the crustal architecture and present-day state of the EHS is essential for deciphering its tectonic evolution. We deployed a 100-km-long magnetotelluric profile across the Eastern Himalayan Syntaxis and performed three-dimensional inversion of the acquired data, yielding a high-resolution 3D electrical model that finely constrains the crustal structure and material state. Based on this model, we analyzed the crustal material conditions, estimated partial melt fractions, and assessed crustal rheology. Integrating our results with complementary geophysical and petrological evidence, we found that crustal materials on the northwestern side of the EHS show extensive partial melting, indicative of channel flow or upward migration of hot material. In contrast, the southeastern side exhibits pronounced strike-slip characteristics. Thus, the rapid uplift of the EHS appears to result primarily from a combination of crustal partial melting within the syntaxis and intense erosion by the Yarlung Tsangpo Gorge.
How to cite: Wang, G., Fang, H., and Pei, F.: The Uplift Mechanism of the Eastern Himalayan Syntaxis, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15506, https://doi.org/10.5194/egusphere-egu26-15506, 2026.
To address the challenge of assessing the reliability of three-dimensional (3D) magnetotelluric (MT) inversion results, we have developed a variational inference (VI) inversion framework (VI-MT) based on the Stein Variational Gradient Descent (SVGD) method. Through parallel particle optimization, we efficiently approximate the model posterior distribution, overcoming the computational limitation of traditional Markov Chain Monte Carlo (MCMC) methods. Synthetic test shows the VI-MT inversion results align with deterministic solutions while better recovering resistivity amplitudes. In comparison, the VI-MT inversion can effectively identify large model uncertainties in the boundary region by the associated multimodal posterior characteristics. Furthermore, the VI-MT inversion is applied to the field MT data collected at the Weishan volcano in northeast China, with the posterior mean model consistent with the deterministic inversion model. Depth-dependent model uncertainties indicate strong data constraints in the upper crust. A clear low resistivity body of ~5 Ω·m with small uncertainties is imaged at depths of ~2-6 km beneath the volcanic crater, suggesting the existence of a shallow magma chamber. Our study shows that the VI-MT inversion based on the SVGD method can efficiently solve the 3D MT Bayesian inversion, providing reliable model uncertainties.
How to cite: Liao, Z., Yang, H., Zhang, X., Gao, J., and Zhang, H.: Three-dimensional magnetotelluric Bayesian inversion based on Stein variational gradient descent, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8800, https://doi.org/10.5194/egusphere-egu26-8800, 2026.
Posters virtual: Mon, 4 May, 14:00–18:00 | vPoster spot 1a
EGU26-21147 | ECS | Posters virtual | VPS29
Dimensionality Analysis of the Iberian Pyrite Belt Lithosphere derived from the Magnetotelluric Impedance Tensor.Mon, 04 May, 14:06–14:09 (CEST) vPoster spot 1a
The Iberian Pyrite Belt (IPB) hosts one of the largest concentrations of massive sulfide deposits in Europe, yet its lithosphere architecture remains incompletely understood. In this study, we employ magnetotelluric (MT) impedance tensor data to investigate the dimensionality and structural characteristics of the IPB crust. The analysis combines two complementary approaches: WAL invariants, computed from the MT impedance tensor using the Waldim code (Martí et al., 2013), which are scalar, rotation invariant quantities providing a robust, frequency dependent measure of three-dimensionality and highlighting anisotropic features in the conductivity distribution; and the Phase Tensor, following the methodology of Caldwell et al. (2004), which offers distortion free insights into the orientation and geometry of regional conductive structures. Integrating these methods enables a systematic dimensional analysis of the impedance tensor, revealing lateral heterogeneities, preferred orientations of conductive features, and depth dependent variations in lithospheric responses.
The results demonstrate that WAL invariants and Phase Tensor analysis together allow the separation of near surface distortions from deeper geoelectric structures, providing a robust framework for characterizing the lithospheric architecture of the IPB. This study highlights the enhanced resolution and robustness achieved by complementing the tensor based analysis of MT data with invariant derived quantities that provide rotationally independent measures of three-dimensionality and anisotropy.
Consequently, this dimensional and structural assessment constitutes a critical prerequisite for subsequent MT data inversion, as it provides essential constraints on model dimensionality, structural orientation, and the treatment of near surface distortion. By supporting the choice between 2D and 3D inversion strategies, the proposed framework enhances the stability of the inversion process, increases the reliability of the conductivity distributions, and ensures greater geological consistency of the resulting models.
Acknowledgment
This work is supported by FCT, I.P./MCTES through national funds (PIDDAC): LA/P/0068/2020, https://doi.org/10.54499/LA/P/0068/2020,UID/50019/2025, https://doi.org/10.54499/UID/PRR/50019/2025, UID/PRR2/50019/2025
References
Caldwell, T.G., Bibby, H.M. and Brown, C. (2004). The magnetotelluric phase tensor. Geophysical Journal International, 158: 457-469. https://doi.org/10.1111/j.1365-246X.2004.02281.x
Castro, C., Hering, P., Junge, A. (2020). FFMT: a MATLAB-based toolbox for Magnetotellurics (MT). 10.13140/RG.2.2.12465.92007.
F. E. M. Lilley. (1998). Magnetotelluric tensor decomposition; Part, Theory for a basic procedure. Geophysics; 63 (6): 1885–1897. doi: https://doi.org/10.1190/1.1444481.
Martí, A., Queralt, P., Ledo, J., Farquharson, C. (2010). Dimensionality imprint of electrical anisotropy in magnetotelluric responses, Physics of the Earth and Planetary Interiors, Volume 182, Issues 3–4, 2010, Pages 139-151, ISSN 0031-9201. https://doi.org/10.1016/j.pepi.2010.07.007.
Martí, A., Queralt, P., Ledo, J. (2013). WALDIM: A code for the dimensionality analysis of magnetotelluric data using the rotational invariants of the magnetotelluric tensor. Computers & Geosciences. 2295-2303. 10.1016/j.cageo.2009.03.004
Miensopust, M. P. (2017). Application of 3-D electromagnetic inversion in practice: Challenges, pitfalls and solution approaches. Surveys in Geophysics, 38(5), 869–933. https://doi.org/10.1007/s10712-017-9435-1.
Vozoff, K. (1991). The magnetotelluric method: Electromagnetic methods. In M. N. Nabighian (Ed.), Applied Geophysics (pp. 641–712).
Kelbert, A., Meqbel, N., Egbert, G. D., & Tandon, K. (2014). ModEM: A modular system for inversion of electromagnetic geophysical data. Computers & Geosciences, 66, 40–53. https://doi.org/10.1016/j.cageo.2014.01.010.
How to cite: Baltazar-Soares, P., Martinéz-Moreno, F. J., Gonzaléz-Castillo, L., Galindo-Zaldívar, J., Monteiro Santos, F., Mateus, A., and Matias, L.: Dimensionality Analysis of the Iberian Pyrite Belt Lithosphere derived from the Magnetotelluric Impedance Tensor., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21147, https://doi.org/10.5194/egusphere-egu26-21147, 2026.
