Search Results (1,451)

Global Navigation Satellite System Reflectometry (GNSS-R), with its all-weather observation capability and low-cost advantage, provides an innovative solution for dynamic sea ice monitoring. In this paper, multi-dimensional features, including the GNSS-R Normalised Integrated Delay Waveform (N-IDW), the scattering coefficient and incidence angle derived from FY-3F satellite data, and the Delay Doppler Map (DDM) bistatic radar cross-section coefficient, are jointly used as model inputs. Experimental results show that this method successfully integrates FY-3F satellite data for sea ice thickness (SIT) retrieval, confirming the viability of employing FY-3F GNSS-R data for this purpose. An assessment of different algorithms in terms of their retrieval performance is conducted—covering RF, DT, KNN, SVM, ET, GBR, XGBR, and LR—and uses these eight models as base learners to construct different stacking models. After comparison, the ensemble stacking model using ET, LR, XGBR, and GBR as base models achieves the best retrieval performance. The MSE of this model for sea ice thickness retrieval reaches 0.0112 m, the RMSE reaches 0.1026 m and the correlation coefficient reaches 0.8876. Full article

Synthetic aperture radar (SAR) imagery provides all-weather, day-and-night observation capabilities that complement electro-optical (EO) imaging; however, the limited number of operational SAR satellites and the difficulty of acquiring expert-annotated SAR datasets constrain deep-learning-based SAR image analysis. In this paper, we propose EO2SAR-Diff, a conditional latent diffusion framework that translates EO aerial images into realistic synthetic SAR images. The framework comprises three core components: (1) domain-adaptive LoRA pre-training that anchors the Stable Diffusion backbone in the remote sensing domain, (2) a style extraction and injection network that captures SAR-specific visual characteristics via multi-scale feature encoding and parallel cross-attention, and (3) a multi-branch ControlNet with three parallel branches for complementary structural guidance. These components are coordinated by a dual-axis feature injection strategy that modulates conditioning strength along both spatial (per-block) and temporal (per-timestep) dimensions. Experiments on the DOTA 1.0 and SARDet-100K datasets demonstrate that EO2SAR-Diff ranks in the top tier among all compared methods in distributional alignment with real SAR imagery, in terms of FID and KID computed with two SAR-domain-adapted feature extractors. Augmenting the SAR training set with our synthetic images yields consistent improvements in downstream object detection performance, confirming the practical utility of the proposed framework. Full article

Ship detection in synthetic aperture radar (SAR) imagery, an indispensable all-weather technology for marine engineering and coastal safety, remains challenging in complex nearshore scenes due to coupled speckle noise, sea–land clutter, large scale variation, and extreme class imbalance. Existing decoupled pipelines fail to jointly mitigate these degradations, leading to high false alarm rates and poor generalization. We propose DN-AnchorNet, an end-to-end unified framework integrating a detection-oriented structure-preserving enhancement branch, a scale-adaptive anchor mechanism, and an adaptive weighted Smooth L1 loss. The detection-guided enhancement branch operates without paired clean data to preserve critical ship structures. The scale-adaptive anchor design enhances matching for small, elongated, and arbitrarily oriented ships, while the tailored loss improves regression robustness through dynamic threshold adjustment and valid positive-sample regression masking under class imbalance. Extensive experiments under the adopted fixed nearshore stress-test protocol of RSDD-SAR and SSDD+ show that DN-AnchorNet achieves the best overall performance among the compared representative oriented object detectors in this evaluation setting, with AP₅₀ values of 0.699 and 0.610, and F1-scores of 0.757 and 0.689, respectively. A strict zero-shot cross-dataset evaluation on HRSID provides supplementary evidence of DN-AnchorNet’s transferability to unseen marine SAR conditions. These results suggest that joint optimization can achieve a favorable accuracy–false-detection balance under challenging nearshore SAR detection conditions. Full article

This study examines why ice covers on the Churchill River in Labrador crack during winter and how weather, river flow, freezing conditions, and riverbed features contribute to these events. Using data from 2010 to 2025 and satellite imagery, the study shows that cracks most often occur in December to February when heavy snow, rapid flow changes, or long cold periods place stress on the ice. Cracking also frequently starts near sandbars where the ice is weaker. The results highlight that no single factor causes cracking. Instead, a combination of snow load, temperature, flow variability, and local river conditions determines when and where cracks form. There is also a disconnect from flow regulation since cracks also formed in 2012 before the construction of the dam began in 2015. A field survey was also carried out employing a combination of borehole jack (BHJ) testing and ground-penetrating radar (GPR) surveys to quantify spatial variations in ice strength and thickness across a portion of the lower Churchill River across two sandbars. In situ BHJ measurements were conducted at multiple sites to determine confined compressive ice strength under both floating and grounded conditions, revealing substantial local variability linked to differences in ice support and the presence of white versus black ice. Complementary GPR transects using 500 MHz and 1000 MHz systems provided high-resolution profiles of ice thickness and internal structure, enabling identification of transitions between grounded and floating ice. The integrated BHJ–GPR approach allowed direct comparison between point-scale strength measurements and spatially continuous thickness and grounding patterns, demonstrating that grounded ice and ice containing higher proportions of white ice exhibited more complex stress states and greater variability in mechanical response. Together, these measurements highlight the importance of combining geophysical surveying with in situ mechanical testing to better understand how environmental conditions control ice integrity and potentially influence ice-jam lodgement propensity along regulated subarctic rivers. Full article

Autonomous driving necessitates a robust 3D perception system that includes accurate object detection, tracking, and segmentation. While recent low-cost camera-based methods have demonstrated promising results, these systems are prone to performance degradation under poor lighting conditions or adverse weather, resulting in considerable localization errors. In this paper, we present a novel approach called Frequency-aware Depth Association Radar-Camera (FDARC) Fusion. This method aims to generate semantically rich and spatially accurate Bird’s-Eye-View (BEV) feature maps by integrating data from both camera and radar sensors. Initially, the image features are enhanced using frequency-aware techniques. Subsequently, these features are transformed into BEV representation with the assistance of depth information estimated from both sensor modalities and radar measurements. This process, known as Depth Association (DA), facilitates more precise BEV representations. Following this, a Temporal and Deformable Cross-Fusion (TDCF) layer is utilized to encode multi-modal feature maps into a unified space-time dimension representation. Extensive experiments conducted on the nuScenes dataset show that FDARC achieves state-of-the-art performance in 3D detection tasks, markedly outperforming baseline models on the nuScenes val set using a ResNet-50 backbone, which attains 53.5% nuScenes Detection Score (NDS) and 44.7% mean Average Precision (mAP). Full article

Rapid and accurate detection of landslide-affected areas is critical for disaster response and risk mitigation. Sentinel-1 SAR imagery offers all-weather, day-and-night observation capability, but existing deep learning approaches treat landslide detection as a single-pass segmentation problem, which limits performance in complex terrain where backscatter changes are confounded by soil moisture, surface roughness, urban double bounce, shadow, and layover effects. MAD-SAR, a rule-based agentic framework that coordinates anomaly detection, super-resolution, object detection, and semantic segmentation under a planning orchestrator and a physics-aware validation engine is proposed. The orchestrator selects specialist modules, their execution order, and the number of refinement iterations according to a scene complexity score computed from SAR-derived statistics. The physics-aware validation engine cross-checks every candidate detection against backscatter change thresholds, DEM-derived slope constraints, and radar geometry masks before any detection is committed to the output. MAD-SAR is evaluated on three Japanese disaster datasets: Hiroshima 2018, Kumamoto 2016, and Ibaraki 2019. On the held-out Ibaraki test event, the framework achieves an F1-score of 0.863 and IoU of 0.759, outperforming all baselines and reducing false alarms by 45% relative to standalone SegFormer. Ablation results confirm that each module contributes to the final performance. These results suggest that multi-module orchestration with embedded physical validation can meaningfully improve SAR-based landslide mapping, though broader validation across regions, sensor configurations, and failure mechanisms remains necessary. Full article

Synthetic Aperture Radar (SAR) is widely used for marine oil spill surveillance due to its all-weather capabilities and sensitivity to sea surface roughness. However, oil slicks often appear as dark formations that can be confounded with visually similar “look-alikes”, making automated detection and boundary delineation challenging. This study proposes a two-stage deep learning framework for oil spill mapping in Sentinel-1 SAR imagery. First, a ConvNeXt-T classifier screens image patches for likely slick presence, reducing the search space for dense prediction. Second, spill boundaries are extracted with a domain-adapted Segment Anything Model (SAM) configured for prompt-free, single-shot segmentation. The input representation is enhanced by combining preprocessed Sentinel-1 VV backscatter with Gray-Level Co-occurrence Matrix (GLCM) texture measures (homogeneity and variance) to better separate oil from heterogeneous background sea at the segmentation level. Quantitative evaluation against established segmentation baselines demonstrates that our adapted SAM achieves the highest overall accuracy, reaching an F1-score of 0.86. This outperforms traditional models such as UNet and CBDNet (0.83), as well as DeepLabV3, SegNeXt, and OFCNet (all at 0.82). Furthermore, an analysis of the wind speed on the test set shows that wind speed affects detectability but does not by itself determine segmentation quality. The results indicate that combining transformer-based screening with efficient foundation-model adaptation can provide accurate and scalable oil spill mapping for operational SAR monitoring. Full article

To address the limitations of the traditional cross-scanning method in absolute calibration of weather radars using metal spheres, including insufficient spatial coverage, limited target acquisition efficiency, and echo underestimation in inter-range bins, this study proposes a sector scanning field calibration method. In this approach, standard metal spheres are suspended from UAVs, and a three-dimensional scanning volume around their theoretical positions is constructed to enable high-density echo sampling. By applying drive backlash correction, quadratic Gaussian surface fitting, and three-dimensional ellipsoid model inversion, key radar parameters can be retrieved. Experimental results show that the improved sector scanning method enhances automation, accuracy, and robustness in field environments and minor target drifts. The experiments were conducted under low-wind and low-clutter conditions. The average calibration error of antenna pointing is 0.08°, the average error of echo intensity calibration is 0.3 dB, the average beamwidth error is 0.07°, the range resolution is 6.6 m, and the average radial ranging error is 14 m. These results indicate that the proposed method can meet the main calibration requirements of weather radars in the present experiments. Full article

Adverse weather conditions, particularly thunderstorms, are the primary cause of flight delays and safety threats, accounting for approximately 58.7% of irregular flights in 2025. Traditional static rerouting methods often fail to adapt to the non-linear evolution of convective weather. This paper proposes a high-fidelity dynamic rerouting framework to enhance flight safety and efficiency. In the perception layer, a RainNet deep learning model is employed for short-term recursive nowcasting of radar reflectivity, which is subsequently transformed into Dynamic Avoidance Zones (DAZ) via clustering and convex hull algorithms. In the decision layer, a two-stage improved Genetic Algorithm (GA) is developed to solve the rerouting path. The first stage generates initial collaborative solutions under a receding-horizon framework, while the second stage applies a “path-straightening” module to reduce cumulative turning angles and curvature fluctuations. The comparative results in actual scenarios demonstrate a distinct dual-advantage over baseline methodologies. Compared to sampling-based strategies, the proposed model reduces the path length by 14.79%. Furthermore, when compared to heuristic algorithms, it actively trades a negligible 1% distance margin to achieve a massive 92.7% reduction in the cumulative turning angle. With a maximum single turn of only 32.51°, the trajectory completely eliminates sawtooth jitter and redundant detours. Ultimately, this research provides essential technical support for improving air traffic management efficiency and reducing controller workload during severe weather events. Full article

Extreme weather events exacerbated by global warming pose severe threats to urban safety, underscoring the urgent need for highly accurate precipitation nowcasting. Short-term local heavy precipitation remains a particular challenge for traditional forecasting due to its suddenness and high disaster potential. To address this, we propose a multi-modal fusion framework that integrates ground-based GNSS-derived Precipitable Water Vapor (GNSS-PWV) and ground-based Radar Composite Reflectivity (CR). While GNSS-PWV keenly captures pre-convective atmospheric water vapor accumulation, radar CR details the morphological distribution of hydrometeors. Specifically, we developed the Spatio-Temporal Enhanced Attention Swin U-Net (STEA-Swin) model to synergize these heterogeneous datasets over the Beijing–Tianjin–Hebei region. High-precision PWV was retrieved from 250 Continuously Operating Reference Stations (CORS) using the dual-frequency ionosphere-free Precise Point Positioning (PPP) method, achieving a strong correlation (>0.97) with ERA5 reanalysis data. Validated against measured data from the 2025 flood season, the STEA-Swin model achieved a Probability of Detection (POD) of 0.68 for torrential rain events at a +1 h forecast lead time. Notably, compared to single-source models, the Critical Success Index (CSI) and POD for torrential rain improved by 18.5% and 21.5%, respectively. These findings demonstrate that coupling deep learning with ground-based GNSS-derived atmospheric thermodynamic information can significantly enhance early warning capabilities, providing a promising technical approach for regional disaster prevention and climate resilience. Full article

Soil moisture (SM) is a critical variable in land–atmosphere water and energy exchange, and synthetic aperture radar (SAR) observations offer an effective means for large-scale and fine-resolution SM monitoring. Sentinel-1A, with its all-time and all-weather capability, has become an indispensable data source for SM retrieval, while comprehensive comparisons of machine learning and deep learning methods for regional and global scale SM retrieval remain insufficient. In this study, four widely used machine learning (ML) algorithms, including random forest (RF), eXtreme gradient boosting (XGBoost), convolutional neural network (CNN), and long short-term memory (LSTM), are evaluated for SM retrieval from Sentinel-1A observations across the International Soil Moisture Network (ISMN) at global and regional scales. Multiple-source dynamic parameters, including Sentinel-1A observations, MODIS vegetation parameters, ERA5-Land meteorological and soil variables, are used as inputs, as well as static geospatial parameters. Validation results demonstrate that tree-based ensemble methods (RF and XGBoost) consistently outperform deep learning methods across all scales. Specifically, XGBoost achieves the best performance with satisfactory SM retrieval results. Moreover, XGBoost is insensitive to Sentinel-1A viewing geometry, allowing fusion of multi-orbit observations to improve temporal resolution without accuracy loss. These findings demonstrate the effectiveness of tree-based ML for global/regional SM retrieval from Sentinel-1A. In addition, this study performs a comprehensive evaluation of spatial generalization ability and orbit robustness of different retrieval models under global heterogeneous environments, and proposes a reliable scheme for generating high-spatiotemporal-resolution SM products. Full article

Flash-flood and floodplain inundation simulations are highly sensitive to the spatiotemporal variability of convective rainfall, particularly during the initial runoff generation stage. However, coarse-resolution numerical weather prediction (NWP) forcing tends to smooth localized rainfall extremes, limiting its ability to accurately represent hydrological responses in low-relief floodplains. In this study, we couple a diffusion-enhanced radar nowcasting model, Diff_ConvLSTM, with a spatial resolution of 1 km and a temporal resolution of 6 min, to assess the hydrological value of high-resolution rainfall forcing over the middle Yangtze River floodplain. We introduce a monotone piecewise cubic Hermite interpolation scheme to ensure a stable transition from discrete high-frequency rainfall inputs to continuous hydrodynamic integration. Evaluation using a radar dataset from 2023 to 2024 shows that Diff_ConvLSTM better preserves intense convective echoes and rainband structures compared to the baseline ConvLSTM, increasing the Probability of Detection at the 40 dBZ threshold by 65.8%. A forcing-replacement experiment for the flood event on 30 June 2023 demonstrates that AI-based nowcasting rainfall forcing reduces peak-discharge underestimation, improves volumetric consistency, and produces inundation patterns that are closer to the observation-driven reference than those generated by low-resolution forecast forcing, although positive biases in inundation area and water depth persist. An additional event in 2024 confirms that the improvements are primarily reflected in discharge magnitude and flood volume representation, while enhancements in peak timing remain limited. Overall, the results illustrate both the added value and the remaining limitations of AI-enhanced nowcasting for hydrologically informed flood forecasting. Full article

Low-level jets (LLJs) play an important role in the transport of heat, water vapor, and atmospheric pollutants. Based on one year (1 September 2023 to 31 August 2024) of tropospheric wind profiler radar (RWP) observations at the Wenjiang Meteorological Observation Base in Chengdu, this study systematically investigates the wind field structure in the middle and lower troposphere over the Chengdu region and the vertical distribution and evolution characteristics of LLJs. The effective detection height of the RWP reaches at least 7.4 km throughout the year, demonstrating good consistency with concurrent radiosonde data. Horizontal wind speed accelerates markedly above 3 km, with the strongest vertical gradient observed in winter. In the lower layer, the prevailing wind direction is primarily controlled by mountain-valley breezes; with increasing altitude, the westerly belt gradually becomes the dominant wind system. Within the atmospheric boundary layer (below 1 km), the wind field exhibits a distinct diurnal cycle: easterly winds dominate in the afternoon, shifting to northerly winds at night. Surface wind speed peaks in the afternoon, whereas upper-level wind speed peaks at night. The occurrence frequency of LLJs is highest in July (26.3% for LLJ-1), followed by April (17.8%). The prevailing wind directions of LLJs are north-northeasterly and northeasterly, and jet core heights are mainly distributed between 0.7 and 1.9 km. For weaker LLJs (LLJ-1 and LLJ-2), both frequency and intensity are higher at night than during the day, peaking at 22:00. These findings deepen our understanding of boundary layer dynamics over complex basin terrain and provide a high-resolution observational benchmark for model improvements and weather warnings. Full article

Multimodal object detection based on synthetic aperture radar (SAR) and optical imagery is of great significance in remote sensing, particularly under adverse weather conditions, nighttime environments, and complex background scenarios. Although SAR imagery has unique advantages under all-weather conditions, its object detection performance still faces challenges in low-texture regions and cluttered scenes. Optical imagery provides rich spatial and texture information, but its applicability is limited in harsh environments. To overcome the limitations of unimodal SAR object detection, this paper proposes a novel multimodal object detection framework, termed StarRoute-DBNet, to improve detection accuracy and robustness through multimodal data fusion and efficient feature interaction. Specifically, a FocusGraph (Graph Convolution-Based Feature Relationship Modeling) module is first designed to adaptively model the spatial relationships between optical and SAR features via graph convolutional networks (GCNs), thereby capturing complex cross-modal spatial dependencies. This module enhances feature interaction across modalities, improves the localization accuracy of oriented targets, and shows clear advantages for small-object detection in complex backgrounds. Second, to alleviate the loss of critical information during downsampling, a PhaseRoute (Sparse Routing Polyphase Downsampling Module) is introduced, which combines multi-phase decomposition with a Top-2 sparse routing strategy to preserve informative spatial cues. By incorporating Gumbel noise into the routing process, the proposed module further improves routing flexibility, detection accuracy, and model robustness. In addition, a Multi-Scale Shuffle-Gated Fusion (MSSGF) module is proposed to address the multi-scale issue in multimodal feature fusion. This module integrates multi-scale convolutional branches, channel shuffles, and dual-attention mechanisms to enhance feature interaction across scales, while an adaptive weighted fusion strategy is employed to dynamically adjust the fusion weights of multimodal features. As a result, the proposed method significantly improves detection accuracy and robustness, especially in complex scenes. Extensive experiments conducted on the MVSDA dataset and the M4-SAR dataset demonstrate that the proposed StarRoute-DBNet consistently outperforms existing state-of-the-art methods under complex backgrounds and adverse conditions. In particular, it achieves clear advantages in oriented object detection and small-object detection, verifying its effectiveness and robustness for cross-modal remote sensing object detection. Full article

A microscale vortex embedded in a cold front over the Pearl River Estuary was observed by weather radars in Hong Kong on the evening of 2 March 2026. This paper presents an observational and simulation study of this vortex. In addition to the reflectivity and Doppler velocity data, the three-dimensional wind field associated with this vortex was analyzed using two radar-based analysis methods. Updrafts were present within the vortex, and the formation of the vortex appears to be related to the horizontal wind shear within the frontal zone and vertical motion triggered by a mid-tropospheric wave. Three commercial aircraft flew across the vortex at low altitude southwest of Lantau Island. Flight data showed marked fluctuations in vertical velocity, including both upward and downward air motions, together with severe turbulence within the vortex. The vortex is therefore of both meteorological interest and operational significance for aviation safety. The event was also simulated using the Weather Research and Forecasting (WRF) model with 200 m resolution. The model reproduced the observed vertical motions and turbulence intensity reasonably well in comparison with aircraft observations. Sensitivity tests with varying sea surface temperature and local terrain over Hong Kong showed no significant impact on the formation of the vortex, confirming that the event was primarily driven by horizontal wind shear in the frontal zone and vertical motion triggered by mid-tropospheric waves. Full article