When the controlled-source electromagnetic (CSEM) data are contaminated by intense cultural noise and signal-to-noise ratio (S/N) is lower than 0 dB, existing denoising methods can hardly achieve good results. To overcome problem, a new strong-noise elimination method called inception-temporal convolutional network-shift-invariant sparse coding (IncepTCN-SISC) developed based on deep learning dictionary learning. First, novel neural network model IncepTCN created inception block temporal...

Inverting potential field data presents a significant challenge due to its ill-posed nature, often leading nonunique model solutions. Addressing this, our work focuses on developing robust joint inversion method for data, aiming achieve more accurate density and magnetic susceptibility distributions. Unlike most previous that used regular meshes, approach adopts an adaptive, unstructured tetrahedral mesh, offering enhanced capabilities in handling the inverse problem of methods. During...

Abstract Gravity surveys in regional geophysical research can be used to estimate the depth of sediment‐basement interface. In this study, we investigate a novel method using convolutional neural network (CNN) for depth‐to‐basement inversion directly from gravity data. Based on Random‐Midpoint‐Displacement (RMD) and features observed data, generate large set realistic interface models. This new model generation significantly reduce size training data sets which is usually considerably train...

The magnetotelluric (MT) signals are susceptible to anthropogenic noise and the existing denoising methods have significant shortcomings in low-frequency situations. To address problem, we propose an innovative approach. It is different from that attempt achieve signal-noise separation through one step. process divided into two steps proposed effective dominant component high-frequency sequentially extracted deep learning dictionary learning. We a new network named DnCNN-GRU which combines...

One of the major problems in modeling and inversion marine controlled-source electromagnetic (CSEM) data is related to need for accurate representation very complex geoelectrical models typical environment. At same time, corresponding forward-modeling algorithms should be powerful fast enough suitable repeated use hundreds iterations multiple transmitter/receiver positions. To this end, we have developed a novel 3D approach, which combines advantages finite-difference (FD) integral-equation...

In controlled-source electromagnetic (CSEM) inversion with conventional regularization, the reconstructed conductivity image is usually blurry and only has limited resolution. To effectively obtain more compact models, we apply concept of multinary transformation to CSEM based on finite-element method an unstructured tetrahedral mesh. Within framework inversion, model conductivities are allowed be clustered within designed values, which obtained from other a priori information or inversion....

One of the most important applications gravity surveys in regional geophysical studies is determining depth to basement. Conventional methods solving this problem are based on spectrum and/or Euler deconvolution analysis field and parameterization earth’s subsurface into prismatic cells. We have developed a new method 3D Cauchy-type integral representation potential fields. Traditionally, fields been calculated using volume integrals over domains occupied by anomalous masses subdivided This...

The correct interpretation of time-domain or transient electromagnetic (TEM) data relies on effective inversion techniques. Herein, we have developed an 3D-TEM method based the vector finite-element method. domain was discretized with unstructured tetrahedra to simulate complex geologic structures. We develop a second-order backward-Euler scheme, rigorously nonuniform time-step sizes, approximate time derivative electric field. waveform directly simulated and considered in forward modeling...

In induced-polarization (IP) surveys, the raw data are usually distorted significantly by presence of electromagnetic (EM) interferences, including cultural noise. Several methods have been proposed to improve signal-to-noise ratio these data. However, signal processing in an electromagnetically noisy environment is still a challenging problem. We determined new and simple technique based on analysis correlation between measured potential injected primary current signals. This applied...

Presently, the 3-D inversion technique has started playing a more important role in controlled-source electromagnetic (CSEM) data interpretation. With development of hardware and computation algorithm, developed rapidly during past decades. In this article, we present newly parallelized algorithm frequency domain with hexahedral discretization. Within framework approach, use finite-element method (FEM) forward modeling Gauss-Newton optimization inversion. We solve adjoint problem efficiently...

Abstract Gravity inversion quantitatively provides a 3D model of density contrasts, significantly enhancing the information extracted from acquired data. However, inherent non-uniqueness poses challenges in precisely determining boundaries anomalous bodies. We have developed an iterative algorithm gravity that reconstructs geometric features bodies by discretizing interpretation with vertical and juxtaposed prism cells. These prisms incorporate sheet-like initial models which are typically...

Funding: This research is funded by the National Natural Science Foundation of China (42274085).Abstract: The Yilgarn Craton in Western Australia, one world's oldest cratons, rich mineral resources and provides significant opportunities for into geothermal energy, crustal dynamics, exploration. To investigate electrical structure southwestern we interpret magnetotelluric data using a finite element-based inversion algorithm developed, complemented Bouguer gravity anomaly data, perform...

Funding: This research is funded by the National Natural Science Foundation of China (42274085).Abstract：Gravity and magnetotelluric methods are pivotal geophysical techniques used to study distribution density electrical conductivity within Earth's interior. These have been widely in multi-scale explorations for various engineering academic applications. Considering varying resolution capabilities different delineating near-surface geological structures, we propose a...

Summary In transient electromagnetic (TEM) methods, the presence of chargeable properties in subsurface introduces parasitic induced polarization (IP) effects, primarily caused by inherent characteristics medium under influence vortex current. TEM responses with IP effects exhibit complex behaviors, especially evident time derivative magnetic field, compared to without effects. By considering current, which includes both and currents, along their derivatives, we aim untangle intricate nature...

Abstract Deep learning methodologies can significantly accelerate the interpretation of airborne transient electromagnetic (ATEM) data. Nevertheless, it remains challenging for deep methods to deal with data vectors missing values. This study introduces innovative processing techniques data, enabling trained neural network effectively manage Furthermore, presents a comprehensive analysis within Yellowstone National Park area, comparing performance networks on real field sets and synthetic in...

Recent advancements highlight the benefits of unstructured tetrahedral meshes in addressing magnetotelluric modeling and inversion challenges. However, geo-electromagnetic inverse problems continue to face significant obstacles, including differing resolution requirements between forward meshes, as well blurring conductivity boundaries results. To address these issues, we propose an innovative mesh decoupling strategy for meshes. The sub-domain decomposition method is used decompose model...

Abstract The complex mechanism by which the conducting phase in pore space influences electrical conductivity of rocks has been a critical focus geophysical exploration. In this study, Pore‐Solid Fractal model is used to accurately describe fluid distribution within space, and resistivity index follows form analogous Archie equation, being expressed terms tortuosity fractal dimension, saturation. physically based parameters then determine saturation exponent. We find reasonable prediction...

In this study, we present an efficient and memory saving 3D inversion algorithm for interpreting controlled-source electromagnetic (CSEM) data using the total electric field formulation. To tackle CSEM problems involving complex geometries, discretize study domain both forward unstructured tetrahedral elements. Our scheme combines parallelized finite element (FE) method with Gauss-Newton optimal strategy. Additionally, transform conventional model space into inversion, significantly reducing...

It is well known that both gravity and magnetolluric (MT) methods can be used for the depth-to-basement estimation due to density conductivity contrast between sedimentary basin underlaid basement rocks. In this case, primary targets are interface sedimenary rocks as physical properties of (density conductivity). The solution inverse problem typically nonunique unstable, especially inversion. order overcome difficulty provide a more robust solution, we have developed method joint inversion...

Over several decades, much research has been done to develop 3D electromagnetic inversion algorithms. Due the computational complexity and memory requirements for time-domain (TEM) algorithms, many real-world surveys are inverted within one dimension. To speed up calculations manage inversions of TEM data, we have developed an approach using three uncoupled meshes: mesh, a forward-model mesh Jacobian calculations. The is coarse regular structured such that constraints easily enforced between...