Qilin Zhang

Continuous monitoring and nowcasting of tornadic near-storm environments remain challenging, particularly in regions with limited ground-based weather radar coverage. High-spatiotemporal-resolution geostationary satellite remote sensing offers a valuable approach to track the evolution of severe convective storms. Combining 10 min cloud-top brightness temperature (TBB) data from the Himawari-8 satellite and ERA5 reanalysis, this study investigates the atmospheric environments of 177 documented tornadoes in China from 2016 to 2023. Tracking storm convective centers using TBB minima reveals clear regional differences in tornadogenesis paradigms. Southern China tornadoes exhibit a “dynamically driven” pattern within quasi-steady, warm, and moist environments. These environments feature low Lifted Condensation Levels (LCL; ~790 m) and weak Convective Inhibition (CIN). Intense low-level wind shear and storm-relative helicity (SRH) dominate the convective triggering. Northern China tornadoes follow a “coupled thermodynamic-kinematic” paradigm under relatively drier and cooler backgrounds. Their initiation relies on the rapid, synchronized accumulation of Mixed-Layer convective available potential energy (MLCAPE) and deep-layer SRH. Furthermore, intensity-based comparative analysis indicates that significant tornadoes (Enhanced Fujita [EF] scale, EF ≥ 2) are favored by higher MLCAPE, deep-layer shear, and lower LCLs compared to weak ones (EF ≤ 1). Himawari-8 TBB data capture a more rapid pre-storm convective cloud-top cooling for strong tornadoes, with medians reaching −73 °C. This study demonstrates that combining high-frequency satellite observations with reanalysis data provides quantitative precursor signals for regional severe tornado nowcasting.

To improve the feature representation and cross-condition generalization of bearing fault diagnosis, this paper proposes an adaptive multi-sensor fusion network with a skip fusion module and a parameter-efficient dual-convolution diagnosis block. The vibration and current signals are first augmented by overlapping segmentation and transformed into the frequency domain using FFT. Multi-scale depthwise convolutions are then employed in parallel branches to capture fault patterns at different receptive fields, and an attention-based skip fusion mechanism selectively aggregates cross-sensor features for complementary enhancement. After fusion, self-calibrated convolution and dilated convolution are alternately applied to strengthen discriminative representation without increasing model complexity. Experiments on multiple bearing datasets under both constant and variable operating conditions demonstrate that the proposed method achieves consistently higher accuracy and robustness than representative CNN-based baselines, verifying its effectiveness for practical bearing fault diagnosis.

The electric fields generated by lightning discharges propagate upward and couple with the lower ionosphere, triggering various mesospheric optical emissions. The potential role of local terrain in modulating the lightning-generated electric fields in the lower ionosphere remains poorly understood. To investigate the effect of mountain terrain on the lightning-generated electric fields at high altitudes (70–85 km), a two-dimensional (2D) finite-difference time-domain (FDTD) simulation model was developed. The simplified mountain is parameterized by its height, width, and horizontal distance from the lightning channel. Simulation results show that mountain terrain significantly influences the lightning-driven electric field waveforms in the initiation region of sprite halos. Increased mountain height leads to greater attenuation of the high-altitude electric field amplitudes, thereby suppressing sprite halos initiation. The shielding effect of mountain width on the electric fields is less pronounced than that of mountain height, and it stabilizes when the width exceeds 40 km. When the horizontal distance between the mountain and lightning channel is less than 40 km, the electric field attenuation increases significantly with decreasing distance. The attenuation effect gradually weakens beyond a distance of 40 km, yet the electric field waveforms exhibit considerable fluctuations due to the reflection process.

Utilizing multi-source meteorological data for lightning identification is crucial for monitoring severe convective weather. However, several key challenges persist in this field: dimensional imbalance and modal competition among multi-source heterogeneous data, model training bias caused by the extreme sparsity of lightning samples, and an imbalance between false alarms and missed detections resulting from complex background noise. To address these challenges, this paper proposes a lightning identification network guided by physical priors and constrained by supervision. First, to tackle the issue of modal competition in fusing satellite (high-dimensional) and radar (low-dimensional) data, a physical prior-guided asymmetric radar information enhancement mechanism is introduced. This mechanism uses radar physical features as contextual guidance to selectively enhance the latent weak radar signatures. Second, at the architectural level, a multi-source multi-scale feature fusion module and a weighted sliding window–multilayer perceptron (MLP) enhanced decoding unit are constructed. The former achieves the coupling of multi-scale physical features at a 2 km grid scale through cross-level semantic alignment, building a highly consistent feature field that effectively improves the model’s ability to detect lightning signals. The latter leverages adaptive receptive fields and the nonlinear modeling capability of MLPs to effectively smooth spatially discrete noise, ensuring spatial continuity in the reconstructed results. Finally, to address the model bias caused by severe class imbalance between positive and negative samples—resulting from the extreme sparsity of lightning events—an asymmetrically weighted BCE-DICE loss function is designed. Its “asymmetric” characteristic is implemented by assigning different penalty weights to false-positive and false-negative predictions. This loss function balances pixel-level accuracy and inter-class equilibrium while imposing high-weight penalties on false-positive predictions, achieving synergistic optimization of feature enhancement and directional suppression. Experimental results show that the proposed method effectively increases the hit rate while substantially reducing the false alarm rate, enabling efficient utilization of multi-source data and high-precision identification of lightning strike areas.

To address the challenges of scarce labeled samples, strong instantaneity, and variable morphology in lightning optical classification—issues that traditional methods struggle to handle efficiently and often require extensive manual intervention—we propose a frame difference triplet network (FD-TripletNet) tailored for few-shot lightning recognition. The lightning optical dataset used in this study was collected from two observation stations over six months, comprising 459 video samples that include lightning events with diverse morphologies (e.g., branched, spherical) and non-lightning events prone to misclassification (e.g., strong light interference, moving objects). Considering the critical feature of lightning—abrupt single-frame changes—we introduce adjacent frame difference matrices as model input to explicitly capture transient brightness variations, reducing noise from static backgrounds. To enhance discriminative ability in few-shot scenarios, the model leverages Triplet Loss to compact intra-class features and separate inter-class features, combined with a dynamic sample matching strategy to focus on challenging cases. The experimental results show that FD-TripletNet achieves a classification accuracy of 94.8% on the dataset, outperforming traditional methods and baseline deep learning models. It effectively reduces the False Negative Rate (FNR) to 3.2% and False Positive Rate (FPR) to 7.4%, successfully distinguishing between lightning and non-lightning events, thus providing an efficient solution for real-time lightning monitoring in meteorological applications.

Radar echo extrapolation based on real-world data is a fundamental problem in meteorological forecasting. Existing extrapolation models typically assume complete radar echo sequences, but in practice, data loss frequently occurs due to equipment failures and communication disruptions. Although traditional solutions can handle missing values through a interpolation-then-prediction pipeline, they suffer from a major limitation: interpolating the missing data and then extrapolating will introduce a cumulative error. To address these issues, we propose MissPred, a radar echo extrapolation model specifically designed for missing data patterns. MissPred employs a dual encoder–decoder architecture. Specifically, the training process involves the sequential execution of interpolation and extrapolation as two distinct serial tasks. In order to circumvent the occurrence of cumulative errors, interpolation and extrapolation are required to share encoder parameters. Furthermore, a missing spatiotemporal feature fusion module (MSTF) that is absent has been designed for the purpose of extracting fine-grained complete spatiotemporal features. Finally, the incorporation of adversarial training is introduced to enhance the authenticity of the prediction results. In order to evaluate the proposed model, case studies are conducted on real radar datasets. Our dataset covers missing rates ranging from 10% to 50%. The experimental results show that the model outperforms the baseline model with the prior interpolation of missing data in the missing mode with stable robustness.

Lightning-induced voltages (LIVs) computation is crucial for lightning protection of power systems and equipment, yet the effect of complex terrain on LIVs remains not fully evaluated. This study establishes a three-dimensional finite-difference time-domain model to investigate the LIVs over Gaussian-shaped mountainous terrain, considering different lightning strike locations. Simulation results show that the influence of Gaussian-shaped mountains on LIVs is directly related to the lightning strike location. Compared with the flat ground scenario, the LIVs’ amplitude can increase by approximately 56% when lightning strikes the mountain top. However, for lightning strikes to the ground adjacent to the mountain, the LIVs’ amplitude is attenuated to varying degrees due to the shielding effect of the mountain. Additionally, the influences of line configuration, as well as mountain height and width on the LIVs, are evaluated.

Satellite optical observations of lightning are influenced by a variety of factors. Studying these factors can provide valuable reference information for applications such as lightning parameter inversion. However, due to the variability of natural factors and the high cost of field observations, research requiring controlled variables often relies heavily on effective simulation models. To this end, we applied our developed optical transmission model for lightning, which can simultaneously account for the spatiotemporal characteristics of lightning sources and observation angles, as well as inhomogeneous and irregular cloud environments, to analyze an unexplained hypothesis from previous studies—that non-planar cloud tops may also be an influencing factor. Our analysis confirms that non-planar cloud tops are indeed an important factor that must be considered, especially under smaller or larger observation angles. In the simulation results, undulations caused an energy increase of up to 43.19% at a 0° observation angle, while at a 60° observation angle, the undulations resulted in an additional attenuation of approximately 17.5%.

The single-site lightning detection system can provide timely and effective information on lightning activity in areas where a multi-site lightning network cannot be built. Using deep learning, the single-site lightning detection achieves better performance than traditional methods, but it is highly dependent on the quality of the training dataset. To address this, this paper proposes a method called Tri-Pre to improve dataset quality and thereby enhance the performance of single-site cloud-to-ground lightning detection based on deep learning. After using the Tri-Pre method, the location model’s distance estimation error decreases by 36.08%. For lightning with propagation distances greater than 1000 km, the average relative error of the results from the built model based on the Tri-Pre method is 3.78%. When verified using additional measured data, the model also shows satisfactory accuracy, particularly for lightning with propagation distances beyond 1000 km. Specifically, for lightning with propagation distances between 1500 and 1600 km, the average relative location error is approximately 5.46%.