The importance of graphite is increasing with the global shift toward carbon neutrality and clean energy transitions. As a critical mineral for battery anodes, graphite demand is projected to grow fourfold by 2030 compared to 2023 levels, with overall demand doubling during the same period. However, graphite is classified as a high-risk mineral due to its heavily concentrated supply chain, which poses significant geopolitical risks. The globalization of manufacturing and trade has further exacerbated supply chain vulnerabilities by centralizing production in low-cost regions, deepening dependence on specific countries. This letter provides a comprehensive overview of the graphite supply chain and its associated challenges while offering strategic insights for stabilization and diversification.

The Yeoncheon titanomagnetite deposit formed by Precambrian magma differentiation is located in Gyeonggi-do, South Korea. Our team conducted an airborne magnetic survey for multiscale mineral exploration and then selected a promising survey area. An electrical resistivity survey was carried out in the potential area to image subsurface structure. Because ore minerals are mainly distributed in gabbro monzodiorite rather than quartz monzodiorite, we applied three-dimensional inversion of electrical resistivity tomography (ERT) data to identify lithology boundaries related to magma differentiation. The resistivity criterion distinguishing the lithologies of gabbro and quartz monzodiorite was determined from laboratory resistivity experimental results performed on drilling cores. The selected region for gabbro monzodiorite extends to the northeast direction, which is consistent with the geology map, magnetic anomaly, and drilling data. The inversion results of ERT can help in selecting the location of geophysical survey or drilling.

As the potential locations of undiscovered ore deposits become deeper, a technique for predicting promising areas in the subsurface media has become necessary. Geoscience data on a wide range of underground media can be obtained through geophysical field exploration, but integration and interpretation of multi-geophysical data are difficult because of differences in spatial resolution. We developed a rock classifier that can predict promising vanadiferous titanomagnetite deposits from multi-geophysical data using supervised machine learning. Vanadiferous titanomagnetite ores are the main source of vanadium, which can be used as a large-scale energy storage system. Model training was conducted using rock samples from drilling cores, and the density of rock samples was used as a criterion for data labeling. We employed the support vector machine, random forest, extreme gradient boosting, LightGBM, and deep neural network for supervised learning, and the accuracy of all methods was 0.95 or greater. We applied trained models to three-dimensional geophysical field data to predict ore body locations. These candidate regions were distributed in the northeast of the geophysical survey area, and some classified areas were verified using a geological map.

Carbon capture and storage (CCS) is a key technology that directly removes industry driven CO₂ to achieve carbon neutrality. In the process of CCS, it is necessary to monitor whether injected CO₂ is properly stored and not leaking. The behavior of CO₂ can be investigated using a 4D seismic survey that compares seismic data before and after injection. We proposed a two-step monitoring with diffraction-angle filtering (DAF) to effectively locate the CO₂ plume. Because DAF allows us to control wavenumber components, the gradient of full-waveform inversion (FWI), which is the first step, is composed of low-wavenumber components, and reverse time migration (RTM) for seismic imaging is carried out with high-wavelength components. To verify our method, we implemented FWI and RTM with and without DAF using the velocity model in the Volve oil field in the North Sea. Numerical examples show that the CO₂ plume is properly detected from the difference between baseline and post-injected survey and the extension of the reflective boundary is improved compared to the results of the conventional method. With our proposed method, local minima problem is mitigated in FWI, and the boundaries between layers can be clearly distinguished in RTM.

This study utilizes domain adaptation to enhance the integration of diverse geoscience datasets, aiming to improve the identification of ore bodies. Traditional mineral exploration methods often face challenges in merging different geoscience data types, which leads to models that do not perform well across varying domains. Domain adaptation is a deep learning strategy aimed at adapting a model developed in one domain (source) to perform well in a different domain (target). To adapt models trained on detailed, labeled drilling data (source) to interpret broader, unlabeled geophysical data (target), Domain-Adversarial Neural Networks (DANNs) were applied, chosen for their robust performance in scenarios where the target domain does not provide labels. This approach was indirectly validated through the minimal overlap between regions identified as candidate ore and borehole locations marked as host rocks, with qualitative validation provided by t-Distributed Stochastic Neighbor Embedding (t-SNE) visualizations showing improved data integration across domains.

Elastic wavefield decomposition into quasi-compressional (qP) and quasi-shear-vertical (qSV) modes is essential for elastic imaging and inversion in VTI media, but becomes computationally expensive when polarization vectors vary strongly in space. I propose a tensor-train (TT) representation of mixed-domain decomposition projectors, constructed via TT-cross with a single user-specified tolerance and applied efficiently using FFT-based operations. A residual-orthogonal strategy extracts qSV from the residual wavefield after qP removal to suppress mode leakage. The method is implemented in Python/PyTorch with GPU acceleration. Numerical experiments on three 2D VTI models (a two-layer benchmark, a BP 2007 benchmark subset, and an Overthrust-based structurally complex model) demonstrate reconstruction errors of 0.094–0.89% for TT, compared to 1.67–6.44% for a conventional CUR low-rank approach (4–46× improvement), with consistently lower cross-talk and near-unity energy ratios. Time-domain receiver traces further confirm that TT yields smaller reconstruction residual spikes and reduced cross-mode leakage than CUR. Runtime tests show that CUR can be faster on smaller grids, whereas TT with GPU acceleration becomes competitive and can outperform CUR for larger models. The TT representation scales linearly with tensor $\mathcal{O}\left(d N_s r^2\right)$—enabling practical extension to higher-dimensional projector tensors where conven-tional methods become impractical.