Junhui Xing

Ground-penetrating radar (GPR) is a nondestructive electromagnetic (EM) method used for subsurface target detection and stratigraphic imaging. The use of a single-frequency GPR profile is inherently limited by the trade-off between penetration depth and vertical resolution. For example, low-frequency antennas provide deeper penetration with lower resolution, whereas high-frequency antennas perform better in shallow areas but attenuate rapidly. Hence, multifrequency GPR data fusion can have complementary advantages. Conventional fusion methods usually depend on handcrafted feature extraction and fusion rules, whereas supervised deep learning-based methods are difficult to apply in polar field surveys due to the unavailability of absolute paired ground-truth labels. To address these limitations, this study proposes an unsupervised dual-branch attention network (UDBA-Net) for the fusion of 400/900 MHz multifrequency GPR data and applies it to Arctic multiyear sea ice exploration. The proposed network combines frequency-specific dual-branch encoders, a convolutional block attention module (CBAM)-like channel- or spatial-attention recalibration module, and an unsupervised hybrid loss comprising local intensity, structural similarity, and Laplacian-gradient constraints. The low- and high-frequency profiles are encoded separately to reduce early cross-band feature aliasing, whereas the attention module adaptively recalibrates the concatenated features based on channel contribution and spatial reflection-event salience. The main objective of the method is the structural and morphological enhancement, rather than preservation, of absolute electromagnetic reflection coefficients. Experimental analyses using controlled synthetic data and Arctic field GPR profiles confirm that UDBA-Net enhances relative structural contrast, improves reflection-event continuity, and integrates shallow, high-resolution textures with deeper, low-frequency interfaces under label-scarce conditions.

Traditional geomagnetic field modeling requires balancing computational efficiency with the resolution of complex dynamics. A dual-architecture deep learning framework is introduced as a high-efficiency surrogate for global modeling. A sinusoidal representation network (SIREN) captures the continuous, quasi-static internal field, and a transformer with an ‘adaptive recursion’ mechanism models the dynamic external field. The recursive mechanism adjusts computational depth according to geomagnetic activity, optimizing efficiency and accuracy. Trained on CHAOS-8.4 data, the framework reproduces the global field with high fidelity. Under identical CPU hardware conditions, the SIREN-based surrogate model achieved an inference speed-up of approximately 1100 times compared to traditional implementations. Native graphics processing unit (GPU) acceleration further enables large scale parallel computations without additional engineering effort. This study validated the feasibility of deep learning as a robust real-time surrogate for complex geophysical simulations.

Neutrino Earth tomography provides an observational approach complementary to seismology for probing the Earth's deep three-dimensional structure. However, existing studies relied on one-dimensional density models that fail to capture lateral heterogeneity, limiting the capability to detect deep structures. To address this, we integrated multi-source data including PREM, CRUST1.0, and HMSL-S06 on a tesseroid grid to construct a non-spherically symmetric three-dimensional density model incorporating large low-shear-velocity provinces (LLSVP). We proposed a matching neutrino trajectory algorithm and derived closed-form expressions for Earth's total mass and moment of inertia to impose global physical constraints. Within the exact three-flavor oscillation propagation framework, we performed forward calculations using public Super-Kamiokande (SK) data and quantitatively compared differences in predicted event counts between the three-dimensional (3D) model and a traditional one-dimensional (1D) spherically symmetric model. The results showed that, compared to the 1D spherically symmetric model, the 3D density model produced only limited changes in total counts of angular event-count distributions as functions of energy and zenith angle, yet it significantly affected flavor-resolved event differences, demonstrating the potential of this framework to discriminate deep 3D structures. With accumulating data from next-generation detectors such as HUNT and Baikal-GVD, the framework established in this study could be applied in practice to neutrino Earth tomography and is expected to provide new observational constraints on Earth’s interior.

Sea ice, as a complex and highly variable medium covering the ocean surface, exhibits intricate internal structural variations that significantly affect the ocean’s thermodynamic, dynamic, and acoustic properties. However, the extreme conditions of the polar environment pose major challenges for conducting in-situ surface experiments, resulting in limited research on the acoustic structure of Arctic sea ice. This paper presents the results obtained from cross-hole acoustic measurements conducted during the 13th Chinese National Arctic Research Expedition. A traveltime tomography method is employed to invert the acoustic velocity distribution within sea ice. The Dijkstra algorithm is used for ray tracing, and damped least-squares inversion is applied to enhance solution stability and control ray path curvature to model acoustic wave propagation paths accurately. Additionally, the attenuation characteristics of sea ice are analyzed by monitoring changes in signal amplitude. This study also investigates the influence of temperature on sound speed and amplitude. The findings of this work contribute valuable insights to future studies on sea ice-climate interactions and acoustic wave propagation in polar ocean environments.

Beachrocks are common coastal sedimentary rocks in tropical and subtropical seas. They are widely spread especially in islands and coastal areas. These rocks are important for island geological evolution research. Research on beachrocks aids in protecting island ecosystems and enhances islands’ ability to prevent and mitigate damage from natural disasters. This study uses unmanned aerial vehicle (UAV) images and the U-Net model based on deep learning to identify beachrocks. To enhance identification accuracy, the efficient channel attention (ECA) mechanism was integrated, leading to improvements of 0.49% in overall accuracy, 1.41% in precision, 0.97% in recall, 1.10% in F1-score, and 2.09% in intersection over union (IoU) compared to the baseline U-Net model. The final results demonstrate that the model effectively identified beachrocks, achieving 97.47% accuracy, 93.27% precision, 94.73% recall, 93.95% F1-score, and 88.65% IoU. This study offers a valuable tool for island geological evolution research and supports the development of large-scale island conservation efforts.

Distributed acoustic sensing (DAS) is an emerging vibration signal acquisition technology that transforms existing fiber-optic communication infrastructure into an array of thousands of seismic sensors. Due to its advantages of low cost, easy deployment, continuous measurement, and long-distance measurement, DAS has rapidly developed applications in the field of marine geophysics. This paper systematically summarizes the status of DAS technology applications in marine seismic monitoring, tsunami and ocean-current monitoring, ocean thermometry, marine target monitoring, and ocean-bottom imaging; analyzes the problems faced during its development; and discusses prospects for further applications in marine geoscience and future research directions.