Bin Wang

Spaceborne Global Navigation Satellite System Reflectometry (GNSS-R) archives usually provide only one delay–Doppler map (DDM) for each recorded observation condition, limiting the representation of residual DDM variability. This study proposes a Position-Guided Conditional Normalizing Flow (PGCFlow) for observation-grounded probabilistic expansion of ocean bistatic radar cross section (BRCS) DDMs. PGCFlow uses four invertible affine coupling blocks to map a 17 × 11 DDM to an equal-dimensional Gaussian latent space. Wind–Auxiliary Condition Modulation incorporates a seven-dimensional condition vector into affine-parameter prediction, while Position-Guided Cross-Partition Aggregation (PGCA) uses deterministic grid descriptors to retain explicit cell locations and facilitate spatial-dependence modeling. Experiments used 5,819,042 quality-controlled CYGNSS observations from 2024. PGCFlow was compared with a conditional variational autoencoder and a generic conditional invertible neural network on 8000 held-out recorded conditions drawn from the same empirical observation domain, with 16 generated DDMs per condition. Although the cVAE achieved the highest balanced-aggregate structural similarity (SSIM) of 0.9396, PGCFlow obtained the lowest Fair Energy Score (FES) and Variogram Score (VS) of 0.2242 and 0.0641 and the closest relative local-neighborhood dispersion to unity at 1.0501. It also achieved the lowest frozen-estimator response RMSE and response MAE of 1.1900 and 0.9129 m/s, respectively. Ablation results indicated individual contributions from both proposed modules. Overall, PGCFlow achieved a favorable trade-off among the evaluated fidelity, dependence, dispersion, and response-consistency measures.

Soil salinization is a major ecological barrier to sustainable agricultural development. Global Navigation Satellite System Reflectometry (GNSS-R), as an emerging passive microwave remote sensing technology, has shown considerable potential for soil parameter monitoring due to its low cost, short revisit cycle, and strong penetrability. However, the significant heterogeneity between Delay–Doppler Maps (DDMs) and auxiliary features, together with the scarcity of in situ salinity observations, poses challenges for GNSS-R-based soil salinity retrieval. To address these issues, this study proposes a Cross-modal Adaptive Fusion Network (CAF-Net) that integrates GNSS-R DDM observations with multi-source auxiliary features through a cross-modal fusion framework to enhance feature representation and model generalization. In addition, a physical reference dataset is constructed by combining SMAP soil moisture products with dielectric models to alleviate the lack of ground salinity observations. Experiments conducted in the Yellow River Delta using Tianmu-1 GNSS-R data demonstrate that the proposed method achieves a correlation coefficient (R) of 0.9546 and a root mean square error (RMSE) of 0.5925 g/kg, outperforming several mainstream benchmark models. Moreover, the retrieved soil salinity shows good agreement with independent datasets in both spatial distribution and statistical characteristics. These results indicate that integrating GNSS-R observations with deep learning provides a promising approach for large-scale soil salinity monitoring.

Mountain forests on the western edge of the Sichuan Basin are challenging to monitor at high resolution because rugged topography, cloud cover, and Landsat-7 SLC-off artifacts create data gaps, while the 2008 Wenchuan earthquake and subsequent restoration further alter vegetation dynamics. We fused Landsat 5/7/8/9 surface reflectance with MODIS MOD13Q1 using an index-then-fusion STARFM framework to reconstruct a continuous 30 m NDVI record for 2000–2024 and quantified forest fraction dynamics using annual forest/non-forest maps, transition analysis, and K-means clustering of pixel-wise NDVI trajectories. To identify dominant controls, we applied a multi-output random forest with spatial block cross-validation and SHAP attribution. The fused NDVI agrees well with MODIS across 100,000 samples (R² = 0.953; RMSE = 0.032), and the regional mean NDVI increased from 0.711 (2000) to 0.774 (2024), showing a post-2008 decline–stagnation–recovery pattern. Forest fraction rose from 48.2% to 72.9%, with accelerated gains after 2010 (+21.4%), and improving trajectories dominated (70.95%), concentrating near the Longmenshan fault zone. The driver model generalized well (micro-mean R² = 0.875), and SHAP ranked elevation (32.6%) and initial forest fraction (32.3%) above temperature and precipitation. These results provide high-resolution evidence of mountain forest change and its primary controls to support terrain-informed ecological management.

Modeling the microwave backscatter of snow-covered sea ice is a significant challenge in cryospheric remote sensing. However, due to the incoherent scattering assumptions, traditional electromagnetic scattering models are inadequate for capturing the nonmonotonic fluctuations in the angular response of the C-band backscatter observed in measured data. To address this issue, this article proposes a coherent multilayer small perturbation method (named COH-ML-SPM), which explicitly incorporates coherent scattering contributions from the air–snow and snow–ice interfaces within a multilayer framework. The proposed model is validated through two levels of experiments. First, COH-ML-SPM is evaluated against C-band backscatter measurements over snow-covered sea ice in Resort Bay, Canada, and compared with three widely used models—ML-SPM, the advanced integral equation model (AIEM), and the snow microwave radiative transfer (SMRT, in independent backscattering approximation (IBA) and dense media radiative transfer (DMRT) configurations)—across snow depths of 14, 8, and 4 cm. Results show that COH-ML-SPM significantly improves simulation accuracy, achieving an average root-mean-square error (RMSE) of 1.3 dB and an $R^{2}$ of 0.9, and successfully reproduces the observed nonmonotonic angular behavior. Notably, within the incidence-angle range of 42°–54°, the proposed model reduces systematic deviations present in traditional models, yielding accuracy improvements of up to 30%. Further validation using independent field measurements from different regions, periods, and environmental conditions demonstrates the robustness and adaptability of the model under diverse snow–ice configurations. Sensitivity analysis of model parameters reveals that dielectric-related properties such as density, temperature, and salinity significantly affect both the amplitude and phase of coherent scattering terms, whereas interface roughness parameters primarily influence scattering intensity with minimal impact on phase. Moreover, analysis of the total backscatter response indicates that snow density and surface roughness are the dominant controlling factors, thereby providing a physically grounded basis for potential applications.

Recognizing the critical role of polar Sea Ice Concentration (SIC) in climate feedback mechanisms, this study presents the first comprehensive investigation of China’s Fengyun-3E(FY-3E) GNOS-II Global Navigation Satellite System Reflectometry (GNSS-R) for bipolar SIC retrieval. Specifically, reflected signals from multiple Global Navigation Satellite Systems (GNSS) are utilized to extract characteristic parameters from Delay Doppler Maps (DDMs). By integrating regional partitioning and dynamic thresholding for sea ice detection, a Random Forest Regression (RFR) model incorporating a rolling-window training strategy is developed to estimate SIC. The retrieved SIC products are generated at the native GNSS-R observation resolution of approximately 1 × 6 km, with each SIC estimate corresponding to an individual GNSS-R observation time. Owing to the limited daily spatial coverage of GNSS-R measurements, the retrieved SIC results are further aggregated into monthly composites for spatial distribution analysis. The model is trained and validated across both polar regions, including targeted ice–water boundary zones. Retrieved SIC estimates are compared with reference data from the OSI SAF Special Sensor Microwave Imager Sounder (SSMIS), demonstrating strong agreement. Based on an extensive dataset, the average correlation coefficient (R) reaches 0.9450 in the Arctic and 0.9602 in the Antarctic for the testing set, with corresponding Root Mean Squared Error (RMSE) of 0.1262 and 0.0818, respectively. Even in the more challenging ice–water transition zones, RMSE values remain within acceptable ranges, reaching 0.1486 in the Arctic and 0.1404 in the Antarctic. This study demonstrates the feasibility and accuracy of GNSS-R-based SIC retrieval, offering a robust and effective approach for cryospheric monitoring at high latitudes in both polar regions.

Global navigation satellite system reflectometry (GNSS-R) is widely used for soil moisture retrieval. To mitigate heterogeneity between delay–doppler map (DDM) and auxiliary features and to impose physical constraints in fusion modeling, we propose a physics-guided cross-feature fusion network (PCF-Net) for the Yellow River Delta, comprising a multi-feature input, cross-feature fusion, and a physics-constrained retrieval module. Specifically, the multi-feature input module employs convolutional neural networks (CNNs) to extract spatial–local features from DDM and deep semantic information from auxiliary data, enabling structural alignment and unified embedding of heterogeneous modalities. The cross-feature fusion module adopts mamba-based cross-feature fusion (MCF) block and transformer-based cross-feature fusion (TCF) block to achieve bidirectional interaction and deep coupling between one-dimensional auxiliary features and DDM image features. The physics-constrained retrieval module constructs a physics-consistency loss between measured and simulated GNSS-R surface reflectivity to regularize the training process. We further construct a multi-source GNSS-R dataset from cyclone global navigation satellite system (CYGNSS), Tianmu-1, and fengyun-3E (FY-3E) and validate against soil moisture active passive (SMAP). The joint use of multi-source GNSS-R data markedly improves retrieval accuracy, achieving R of 0.9805 and RMSE of 0.0176 cm³/cm³, demonstrating effectiveness at the regional scale.

Due to the limited availability of subsurface and deep sea observation data, accurately reconstructing high spatiotemporal resolution ocean interior temperatures is both a crucial and challenging task. This study proposes Patch Self attention and Season Attention Generative Adversarial Network (PSSA-GAN) for reconstructing high spatiotemporal resolution ocean subsurface temperature fields. The model utilizes sea level anomaly (SLA), sea surface temperature (SST), and sea surface wind (SSW) data as inputs, extracting multi-scale features and seasonal signals from the remote sensing data to reconstruct the ocean subsurface temperature fields. To improve the model's ability to capture global features and to allow the model to incorporate seasonal variations in ocean temperature during training. The patch self-attention and season attention mechanism are integrated into the generator. To ensure the reconstructed temperature fields more align with the real ocean environment, a re-weighting loss function is designed. The loss function jointly leverages ocean reanalysis data and Argo observations as ground truth for model training. The daily temperature fields in the Pacific Ocean (104°E-127°W, 20°S-44°N) are reconstructed by PSSA-GAN. The construction results demonstrate that PSSA GAN consistently captures the spatial patterns across various depth layers, aligning well with reanalysis data and reflecting fine scale ocean variations in the tropical eastern Pacific. The average RMSE of the reconstruction results across 17 layers above 1062m is 0.5213°C, outperforming both the Attention U-net and Pix2Pix models. The evaluation experiments based on Argo in-situ observations show that the reconstructed daily mean temperature field achieves a numerical accuracy comparable to that of ocean reanalysis data.

Synthetic aperture radar (SAR) imagery is indispensable for rapid, weather-independent marine oil-spill monitoring. Yet, the acute shortage of annotated SAR spill imagery severely limits deep learning detectors. While generative adversarial networks (GANs) have been used to generate synthetic data, their inherent limitations—training instability and mode collapse—often result in blurred and semantically inconsistent outputs. To overcome these challenges, we introduce a classification-guided diffusion model (CG-DM), which integrates the expressive power of diffusion processes with task-specific guidance. CG-DM incorporates three key innovations: 1) morphology-aware classification guidance: SAR oil-spill images are categorized into different morphological categories (blocky, elongated, and patchy); this category information conditions every step of the reverse-diffusion process, enabling fine-grained control over the global geometry of generated spills while preserving intracategory diversity; 2) label-synchronized generation: the model simultaneously generates the SAR image and its corresponding pixel-level annotation mask, which eliminates the need for the time-consuming and error-prone manual labeling process; and 3) spatial-aware attention mechanism: a lightweight attention mechanism performs localized window self-attention with relative positional offsets, which significantly enhances the sharpness of spill edges and the fidelity of speckle-textured details. Evaluated on the M4D benchmark semantic segmentation dataset of the Information Technologies Institute (ITI)/European Maritime Safety Agency (EMSA) with Sentinel-1 SAR imagery from 2015 to 2017, CG-DM achieves a Fréchet inception distance (FID) of 203.14 and a kernel inception distance (KID) of 0.124, surpassing state-of-the-art GAN baselines by substantial margins. Besides, ablation studies confirm the critical contributions of each component: spatial-aware attention mechanism significantly enhances generation quality, while classification guidance effectively preserves morphological feature diversity. Crucially, under extreme data scarcity, training augmentation with CG-DM synthetic samples improves oil-spill detection IoU by up to 81%, demonstrating strong practical utility. This study establishes the first annotation-free paradigm for SAR oil-spill data generation, paving the way for high-accuracy maritime disaster monitoring systems.

As an emerging remote sensing technology, Global Navigation Satellite System Reflectometry (GNSS-R) offers moderate penetration capabilities and supports all-weather observations. It has been widely applied in soil moisture retrieval. However, GNSS-R-based soil moisture inversion faces significant challenges due to the marked heterogeneity, complex feature associations, and difficulty in unified modeling among Delay-Doppler Map (DDM) imagery, time delay waveforms, and auxiliary features. To address these issues, this study introduces a multi-modal fusion strategy combined with a deep learning approach that possesses strong nonlinear modeling capabilities, and proposes a novel GNSS-R soil moisture retrieval model termed the Multi-modal Cross-Branch Fusion Network (MCBF-Net). MCBF-Net consists of three parallel branches designed to process different modalities of input data: a one-dimensional convolutional neural network (1D CNN) module for efficiently extracting features from time delay waveforms; a multi-scale fusion Transformer module for capturing the global features of DDM images; and a backpropagation (BP) neural network module that leverages its powerful nonlinear mapping capabilities to deeply learn complex patterns in the auxiliary data. Additionally, a multi-modal feature fusion module and a fusion loss function are embedded within the tri-branch architecture to ensure effective integration of features across modalities and to enhance the synergy among the network branches. Through this design, the model enables deep and comprehensive feature extraction and fusion from time delay waveforms, DDM imagery, and auxiliary information, significantly improving the accuracy of soil moisture retrieval. This study uses the Yellow River Delta region as the experimental area, focusing on the urgent needs of ecological environmental protection and sustainable agricultural development in the region. The soil moisture observation data provided by the Soil Moisture Active Passive (SMAP) satellite were used to validate the effectiveness of the proposed method. The retrieved results achieved a root mean square error (RMSE) of 0.0284 cm³/cm³ and a correlation coefficient (R) of 0.9232, indicating high accuracy and reliability. Furthermore, the study analyzed the spatiotemporal distribution characteristics of soil moisture in the Yellow River Delta, offering a feasible and efficient solution for regional-scale soil moisture inversion and refined monitoring.

Deep learning models have obvious advantages in detecting oil spills, but the training of deep learning models heavily depends on a large number of samples of high quality. However, due to the accidental nature, unpredictability, and urgency of oil spill incidents, it is difficult to obtain a large number of labeled samples in real oil spill monitoring scenarios. Surprisingly, few-shot learning can achieve excellent classification performance with only a small number of labeled samples. In this context, a new cross-domain few-shot SAR oil spill detection network is proposed in this paper. Significantly, the network is embedded with a hybrid attention feature extraction block, which consists of a coordinate attention module to perceive the channel information and spatial location information, as well as a global self-attention transformer module capturing the global dependencies and a multi-scale self-attention module depicting the local detailed features, thereby achieving deep mining and accurate characterization of image features. In addition, to address the problem that it is difficult to distinguish between the suspected oil film in seawater and real oil film using few-shot due to the small difference in features, this paper proposes a double loss function category determination block, which consists of two parts: a well-designed category-perception loss function and a traditional cross-entropy loss function. The category-perception loss function optimizes the spatial distribution of sample features by shortening the distance between similar samples while expanding the distance between different samples. By combining the category-perception loss function with the cross-entropy loss function, the network’s performance in discriminating between real and suspected oil films is thus maximized. The experimental results effectively demonstrate that this study provides an effective solution for high-precision oil spill detection under few-shot conditions, which is conducive to the rapid identification of oil spill accidents.