Remote Sensing

27 pages, 5279 KB

Open AccessArticle

Precise Cross-Sea Orthometric Height Determination Using GNSS Carrier-Phase Time-Frequency Transfer

by Kuangchao Wu, Wen-Bin Shen, Ziyu Shen, Hok Sum Fok, Yanming Guo, Kezhao Li, Weitao Yan, Zengzeng Lian, Jinjiang Wang and Huijia Guo

Remote Sens. 2025, 17(24), 3949; https://doi.org/10.3390/rs17243949 (registering DOI) - 6 Dec 2025

Abstract

State-of-the-art atomic clocks, in combination with high-precision time-frequency transfer techniques, have established a novel relativistic geodetic approach for determining the Earth’s geopotential. By exploiting ultra-stable atomic clocks and GNSS Precise Point Positioning (PPP) time-frequency transfer, this study investigates the cross-sea Orthometric Height (OH) [...] Read more.

State-of-the-art atomic clocks, in combination with high-precision time-frequency transfer techniques, have established a novel relativistic geodetic approach for determining the Earth’s geopotential. By exploiting ultra-stable atomic clocks and GNSS Precise Point Positioning (PPP) time-frequency transfer, this study investigates the cross-sea Orthometric Height (OH) determination between two remote stations separated by over 8000 km, corresponding to an OH difference of approximately 2260 m. Simulation results indicate that, when employing clocks with a frequency stability of 1 × 10

^{- 18}

, the remote OH determination could achieve a limiting accuracy of approximately 20 cm. This limitation is primarily attributed to the finite precision of the PPP time-frequency transfer, which constrains the ultimate performance of the OH determination. Furthermore, aggregating multiple observation periods could further enhance the accuracy to approximately 6 cm. These findings demonstrate that the PPP time-frequency transfer facilitates high-precision OH determination over intercontinental distances and thereby provides a feasible pathway toward the realization of a centimeter-level International Height Reference System (IHRS). Full article

(This article belongs to the Section Earth Observation Data)

24 pages, 2582 KB

Open AccessArticle

Temperature-Field Driven Adaptive Radiometric Calibration for Scan Mirror Thermal Radiation Interference in FY-4B GIIRS

by Xiao Liang, Yaopu Zou, Changpei Han, Pengyu Huang, Libing Li and Yuanshu Zhang

Remote Sens. 2025, 17(24), 3948; https://doi.org/10.3390/rs17243948 (registering DOI) - 6 Dec 2025

Abstract

To meet the growing demand for quantitative remote sensing applications in GIIRS radiometric calibration, this paper proposes a temperature field-driven adaptive scan mirror thermal radiation interference correction method. Based on the on-orbit deep space observation data from the Fengyun-4B satellite, this paper systematically [...] Read more.

To meet the growing demand for quantitative remote sensing applications in GIIRS radiometric calibration, this paper proposes a temperature field-driven adaptive scan mirror thermal radiation interference correction method. Based on the on-orbit deep space observation data from the Fengyun-4B satellite, this paper systematically analyzes the thermal radiation interference characteristics caused by scan mirror deflection and constructs the first scan mirror thermal radiation response model suitable for GIIRS. On the basis of this model, this paper further introduces the dynamic variation characteristics of the internal thermal environment of the instrument, enabling adaptive response and compensation for radiation disturbances. This method overcomes the limitations of relying on static calibration parameters and improves the generality and robustness of the model. Independent validation results show that this method effectively suppresses the interference of scan mirror deflection on instrument background radiation and enhances the consistency of the deep space and blackbody spectral diurnal variation time series. After correction, the average system bias of the interference-sensitive channel decreased by 94%, and the standard deviation of radiance bias from 2.5 mW/m²·sr·cm⁻¹ to below 0.5 mW/m²·sr·cm⁻¹. In the O-B test, the maximum improvement in relative standard deviation reached 0.15 K. Full article

(This article belongs to the Special Issue Remote Sensing Data Preprocessing and Calibration)

17 pages, 2489 KB

Open AccessArticle

Vegetation Changes and Its Driving Factors in the Three-River Headwaters Region from 1990 to 2022

by Chen Wang, Junbang Wang, Zhiwen Dong, Shaoqiang Wang and Xiaoyu Jiao

Remote Sens. 2025, 17(24), 3947; https://doi.org/10.3390/rs17243947 (registering DOI) - 6 Dec 2025

Abstract

Changes in vegetation coverage reflect the status and dynamic processes of ecosystems and serve as a crucial foundation for regional ecological protection. Using Landsat-5 and Sentinel-2 data, this study calculated the vegetation coverage in the Three-River Headwaters (TRH) region from 1990 to 2022 [...] Read more.

Changes in vegetation coverage reflect the status and dynamic processes of ecosystems and serve as a crucial foundation for regional ecological protection. Using Landsat-5 and Sentinel-2 data, this study calculated the vegetation coverage in the Three-River Headwaters (TRH) region from 1990 to 2022 with the pixel dichotomy model, identified land cover changes over the past three decades via a deep neural network, and analyzed the primary influencing factors behind vegetation coverage dynamics. The results indicate that vegetation coverage in TRH has generally increased, as very high vegetation coverage expanded by 10.3%, while very low and low vegetation coverage decreased by 4.2%. Extensive bare land in the western region decreased and transformed into grassland, while the areas of shrubland and forest in the central and eastern TRH areas increased. The areas of grassland, shrubland, and forest increased by 3.7 × 10⁴ km², 2.1 × 10⁴ km², and 4.7 × 10³ km², respectively. Precipitation, elevation, and temperature are the main factors influencing the spatial variation in vegetation coverage. We found that the contributions of the permafrost active layer thickness and precipitation to changes in vegetation coverage are high. Finally, we provide a detailed and timely analysis of recent vegetation distribution and type changes on the Tibetan Plateau, offering a strengthened scientific foundation for monitoring, assessment, and ecological conservation efforts aimed at supporting ecosystem restoration in the region. Full article

► Show Figures

Figure 1

19 pages, 5281 KB

Open AccessArticle

Morphometric Analysis and Emplacement Dynamics of Folded Terrains at Avernus Colles, Mars

by Caitlin Ahrens and Rachel A. Slank

Remote Sens. 2025, 17(24), 3946; https://doi.org/10.3390/rs17243946 (registering DOI) - 6 Dec 2025

Abstract

Folded, arcuate terrains on the surface of Mars provide insight into the volcanic properties of surface materials and emplacement dynamics. This research focused on the analysis of folded terrains in the chaotic-terrain Avernus Colles region, located near Elysium Planitia, using images from the [...] Read more.

Folded, arcuate terrains on the surface of Mars provide insight into the volcanic properties of surface materials and emplacement dynamics. This research focused on the analysis of folded terrains in the chaotic-terrain Avernus Colles region, located near Elysium Planitia, using images from the Mars Odyssey Orbiter and altimetry data from the Mars Orbiter Laser Altimeter (MOLA). The combined data revealed areas of deformation, which is inferred to be the result of compressions and possibly collapse from the late Amazonian period. We identified and measured 19 distinct folds, with morphometric wavelengths ranging from 0.7 to 1.75 km. These measurements were applied to a simple two-layer regolith model to better understand the folding patterns observed. The model suggests that these folds could have formed with an upper viscous boundary layer less than 0.55 km thick and strain rates approximately 10⁻⁷ s⁻¹. These strain rates indicate that the deformation of the terrains likely occurred over a relatively short period of time, ranging from 16 to 38 days. By studying these deformation patterns, we can enhance our understanding of the volcanic history and surface processes on Mars, offering insight into the planet’s geologic evolution and material properties. Full article

(This article belongs to the Special Issue Planetary Geologic Mapping and Remote Sensing (Third Edition))

► Show Figures

Figure 1

40 pages, 4012 KB

Open AccessReview

Soil Moisture Monitoring Method and Data Products: Current Research Status and Future Development Trends

by Ruihao Liu, Cun Chang, Ruisen Zhong and Shiyang Lu

Remote Sens. 2025, 17(24), 3945; https://doi.org/10.3390/rs17243945 - 5 Dec 2025

Abstract

Soil moisture (SM) is a key variable regulating land–atmosphere energy exchange, hydrological processes, and ecosystem functioning. Though important, there are still unresolved problems in accurate SM monitoring and the practical application and validation of existing methods. In this review, we integrate mechanistic classification [...] Read more.

Soil moisture (SM) is a key variable regulating land–atmosphere energy exchange, hydrological processes, and ecosystem functioning. Though important, there are still unresolved problems in accurate SM monitoring and the practical application and validation of existing methods. In this review, we integrate mechanistic classification and applicability and constraint discussions to develop a coherent understanding of current SM monitoring approaches. Within this framework, in situ measurements, optical and thermal infrared methods, active and passive microwave remote sensing (RS) techniques, and model-based simulations are compared, and publicly accessible SM dataset products are comparatively analyzed in terms of product characteristics and application limitations. Different from other published reviews, this study covers a large scope of SM monitoring methods varying from in situ observation to RS inversion, and classifies them based on their mechanisms, thereby constructing a complete comparative framework for SM research. Moreover, three types of open-access SM dataset products are investigated, optical and microwave RS products, model simulation and data fusion products, and reanalysis dataset products, and evaluated according to their resolution, depth, applicability, advantages, and limitations. By doing so, it is concluded that in situ observations remain essential for calibration and validation but are spatially limited. Optical and thermal infrared methods are restricted by atmospheric conditions and a shallow penetration depth, while microwave techniques exhibit varying performances under different vegetation and soil conditions. Existing datasets differ significantly in resolution, consistency, and coverage, making no single product universally applicable. Future research should focus on multi-source and spatiotemporal data fusions, the integration of machine learning with physical mechanisms, enhancement for cross-sensor consistency, the establishment of standardized uncertainty evaluation frameworks, and the refinement of high-order RTMs and parameterization. Full article

► Show Figures

Figure 1

18 pages, 6983 KB

Open AccessArticle

Predictability of Landfalling Typhoon Tracks in East China Based on Ensemble Sensitivity Analysis

by Jing Zhang, Shoupeng Zhu, Yan Tan and Chen Chen

Remote Sens. 2025, 17(24), 3944; https://doi.org/10.3390/rs17243944 - 5 Dec 2025

Abstract

Accurate typhoon track forecasting is vital for disaster mitigation in East China, a region frequently impacted by landfalling typhoons. Despite advances in numerical weather prediction, uncertainties remain high, especially within 48 h of landfall, due to complex interactions among tropical cyclones, the subtropical [...] Read more.

Accurate typhoon track forecasting is vital for disaster mitigation in East China, a region frequently impacted by landfalling typhoons. Despite advances in numerical weather prediction, uncertainties remain high, especially within 48 h of landfall, due to complex interactions among tropical cyclones, the subtropical high, and mesoscale systems. This study applies Ensemble-based Sensitivity Analysis (ESA) within a high-resolution regional ensemble prediction system (Shanghai Weather And Risk Model System-Ensemble Prediction System, SWARMS-EN) to investigate forecast uncertainties of three representative typhoons—Gaemi, Bebinca, and Kong-rey—that made landfall in East China in 2024. Our results reveal consistent sensitivity patterns across diverse large-scale environments, particularly around the western flank of the subtropical high and in proximity to nearby low-pressure systems. Track uncertainty was closely tied to fluctuations in the steering flow, notably its zonal component. Moreover, binary typhoon interactions emerged as key drivers of forecast divergence. ESA effectively identified sensitive regions where small initial perturbations exert significant downstream influence on typhoon tracks. This study demonstrates the operational value of ESA for diagnosing forecast error sources and guiding targeted observations. By linking forecast uncertainty to physical mechanisms, this research enhances our understanding of typhoon predictability and supports the development of more adaptive and accurate regional forecasting systems. Full article

(This article belongs to the Special Issue Multi-Source Atmospheric Remote Sensing: Enabling High-Precision Meteorological Monitoring and Forecasting)

27 pages, 12824 KB

Open AccessArticle

Multiscale Attention-Enhanced Complex-Valued Graph U-Net for PolSAR Image Classification

by Wanying Song, Qian Liu, Kuncheng Pu, Yinyin Jiang and Yan Wu

Remote Sens. 2025, 17(24), 3943; https://doi.org/10.3390/rs17243943 - 5 Dec 2025

Abstract

The powerful graph convolutional network (GCN) for polarimetric synthetic aperture radar (PolSAR) image classification generally relies on real-valued features, ignoring the phase information and thus limiting the modeling of complex-valued (CV) polarization characteristics. To address this issue, this paper proposes a novel multiscale [...] Read more.

The powerful graph convolutional network (GCN) for polarimetric synthetic aperture radar (PolSAR) image classification generally relies on real-valued features, ignoring the phase information and thus limiting the modeling of complex-valued (CV) polarization characteristics. To address this issue, this paper proposes a novel multiscale attention-enhanced CV graph U-Net model, abbreviated as MAE-CV-GUNet, by embedding CV-GCN into a graph U-Net framework augmented with multiscale attention mechanisms. First, a CV-GCN is constructed based on the real-valued GCN, to effectively capture the intrinsic amplitude and phase information of the PolSAR data, along with the underlying correlations between them. This way can well lead to an improved feature representation for PolSAR images. Based on CV-GCN, a CV graph U-Net (CV-GUNet) architecture is constructed by integrating multiple CV-GCN components, aiming to extract multi-scale features and further enhance the ability to extract discriminative features in the complex domain. Then, a multiscale attention (MSA) mechanism is designed, enabling the proposed MAE-CV-GUNet to adaptively learn the importances of features at various scales, thereby dynamically fusing the multiscale information among them. The comparisons and ablation experiments on three PolSAR datasets show that MAE-CV-GUNet has excellent performance in PolSAR image classification. Full article

(This article belongs to the Special Issue Advances in AI-Driven Synthetic Aperture Radar (SAR): Data Processing to Automatic Interpretation)

22 pages, 13202 KB

Open AccessArticle

Deep Learning-Based Remote Sensing Monitoring of Rock Glaciers—Preliminary Application in the Hunza River Basin

by Yidan Liu, Tingyan Xing and Xiaojun Yao

Remote Sens. 2025, 17(24), 3942; https://doi.org/10.3390/rs17243942 - 5 Dec 2025

Abstract

Rock glaciers have been recognized as key indicators of geomorphic and climatic processes in high mountain environments. In this study, Sentinel-2 MSI imagery and topographic data were integrated to construct enhanced feature sets for rock glacier identification. Three state-of-the-art deep learning models (U-Net, [...] Read more.

Rock glaciers have been recognized as key indicators of geomorphic and climatic processes in high mountain environments. In this study, Sentinel-2 MSI imagery and topographic data were integrated to construct enhanced feature sets for rock glacier identification. Three state-of-the-art deep learning models (U-Net, DeepLabV3+, and HRnet) were employed to perform semantic segmentation for extracting rock glacier boundaries in the Hunza River Basin, located in the eastern Karakoram Mountains. The combination of spectral and terrain features significantly improved the differentiation of rock glaciers from surrounding landforms, establishing a robust basis for model training. A series of comparative experiments were conducted to evaluate the performance of each model. The HRnet model achieved the highest overall accuracy, exhibiting superior capabilities in high-resolution feature representations and generalization. Using the HRnet framework, a total of 597 rock glaciers were identified, covering an area of 183.59 km². Spatial analysis revealed that these rock glaciers are concentrated between elevations of 4000 m and 6000 m, with maximum density near 5000 m, and a predominant south and southwest orientation. These spatial patterns reflect the combined influences of topography, thermal conditions, and snow accumulation on the formation and preservation of rock glaciers. The results confirm the effectiveness of deep learning-based semantic segmentation for large-scale rock glacier mapping. The proposed framework establishes a technical foundation for automated monitoring of alpine landforms and supports future assessments of rock glacier dynamics under climate variability. Full article

26 pages, 14433 KB

Open AccessArticle

Decrypting Spatiotemporal Dynamics and Optimization Pathway of Ecological Resilience Under a Panarchy-Inspired Framework: Insights from the Wuhan Metropolitan Area

by An Tong, Yan Zhou, Jiazi Zheng and Ziqi Liu

Remote Sens. 2025, 17(24), 3941; https://doi.org/10.3390/rs17243941 - 5 Dec 2025

Abstract

Environmental degradation from rapid urbanization significantly threatens ecological resilience (ER). Nevertheless, accurately evaluating ER remains a persistent challenge. Prior studies’ limited attention to resilience’s cross-scale complexity has hindered evidence-based management. This study, based on long-term time series remote sensing and multi-source data, developed [...] Read more.

Environmental degradation from rapid urbanization significantly threatens ecological resilience (ER). Nevertheless, accurately evaluating ER remains a persistent challenge. Prior studies’ limited attention to resilience’s cross-scale complexity has hindered evidence-based management. This study, based on long-term time series remote sensing and multi-source data, developed a cross-scale spatiotemporal ER analysis framework integrating landscape ecology and panarchy perspectives. A local “resistance–adaptation–recovery” substrate resilience evaluation was combined with telecoupling-based global network resilience to quantify multi-scale ER from 2000 to 2020. Key drivers across time scales were identified using a hybrid XGBoost–SHAP and genetic algorithm (GA)–optimized dynamic Bayesian network (DBN), and spatial optimization scenarios were simulated with patch-generating land use simulation (PLUS) model. ER decreased slightly from 0.4856 in 2000 to 0.4503 in 2020, with dynamic fluctuations across periods. A clear spatial pattern emerged, with higher ER in the east and lower in the west. Forest land contributed strongly to ER, while construction and cropland reduced it. Spatial composition factors—especially the proportions of forest and construction land—were dominant drivers, outweighing structural factors such as landscape pattern. DBN backward inference revealed nonlinear threshold effects among socio–natural–spatial drivers. Scenario-based simulations confirmed that regulating spatial composition via our optimization pathway can enhance ER. This is particularly effective when expanding forestland in mountainous regions while restraining the growth of built-up areas. This study proposes an integrated framework of “resilience assessment—driver analysis—spatial optimization,” which not only advances the theoretical basis for nested ER assessment but also offers a transferable approach for optimizing spatial patterns and sustainable land management, thereby enhancing ecological resilience in rapidly urbanizing regions. Full article

(This article belongs to the Section Ecological Remote Sensing)

► Show Figures

Figure 1

18 pages, 4569 KB

Open AccessArticle

Accuracy Assessment of Shoreline Extraction Using MLS Data from a USV and UAV Orthophoto on a Complex Inland Lake

by Mariusz Specht and Oktawia Specht

Remote Sens. 2025, 17(24), 3940; https://doi.org/10.3390/rs17243940 - 5 Dec 2025

Abstract

Accurate shoreline determination is essential for the study of coastal and inland water processes, hydrography, and the monitoring of aquatic and terrestrial environments. This study compares two modern remote sensing technologies: MLS conducted with a USV and photogrammetry using a UAV. The research [...] Read more.

Accurate shoreline determination is essential for the study of coastal and inland water processes, hydrography, and the monitoring of aquatic and terrestrial environments. This study compares two modern remote sensing technologies: MLS conducted with a USV and photogrammetry using a UAV. The research was carried out on Lake Kłodno, characterised by a complex shoreline with vegetation and hydrotechnical structures. Both approaches satisfied the accuracy requirements of the IHO Special Order for shoreline extraction (≤5 m at the 95% confidence level). For the UAV-derived orthophoto, the error within which 95% of shoreline points were located (corresponding to 2.45·σ) was 0.05 m for the natural shoreline and 0.06 m for the variant including piers, both well below the IHO threshold. MLS achieved a 95% error of 1.16 m, which also complies with the Special Order criteria. UAV data enable clear interpretation of the land–water boundary, whereas MLS provides complete three-dimensional spatial information, independent of lighting conditions, and allows surveys of vegetated or inaccessible areas. The results demonstrate the complementarity of the two approaches: UAV is well suited to highly accurate shoreline mapping and the identification of hydrotechnical structures, while MLS is valuable for analysing the nearshore zone and for surveying vegetated or inaccessible areas. The findings confirm the value of integrating these approaches and highlight the need to extend research to other types of waterbodies, to consider seasonal variability, and to develop methods for the automatic extraction of shorelines. Full article

► Show Figures

Figure 1

25 pages, 4675 KB

Open AccessArticle

DLiteNet: A Dual-Branch Lightweight Framework for Efficient and Precise Building Extraction from Visible and SAR Imagery

by Zhe Zhao, Boya Zhao, Ruitong Du, Yuanfeng Wu, Jiaen Chen and Yuchen Zheng

Remote Sens. 2025, 17(24), 3939; https://doi.org/10.3390/rs17243939 - 5 Dec 2025

Abstract

High-precision and efficient building extraction by fusing visible and synthetic aperture radar (SAR) imagery is critical for applications such as smart cities, disaster response, and UAV navigation. However, existing approaches often rely on complex multimodal feature extraction and deep fusion mechanisms, resulting in [...] Read more.

High-precision and efficient building extraction by fusing visible and synthetic aperture radar (SAR) imagery is critical for applications such as smart cities, disaster response, and UAV navigation. However, existing approaches often rely on complex multimodal feature extraction and deep fusion mechanisms, resulting in over-parameterized models and excessive computation, which makes it challenging to balance accuracy and efficiency. To address this issue, we propose a dual-branch lightweight architecture, DLiteNet, which functionally decouples the multimodal building extraction task into two sub-tasks: global context modeling and spatial detail capturing. Accordingly, we design a lightweight context branch and spatial branch to achieve an optimal trade-off between semantic accuracy and computational efficiency. The context branch jointly processes visible and SAR images, leveraging our proposed Multi-scale Context Attention Module (MCAM) to adaptively fuse multimodal contextual information, followed by a lightweight Short-Term Dense Atrous Concatenate (STDAC) module for extracting high-level semantics. The spatial branch focuses on capturing textures and edge structures from visible imagery and employs a Context-Detail Aggregation Module (CDAM) to fuse contextual priors and refine building contours. Experiments on the MSAW and DFC23 Track2 datasets demonstrate that DLiteNet achieves strong performance with only 5.6 M parameters and extremely low computational costs (51.7/5.8 GFLOPs), significantly outperforming state-of-the-art models such as CMGFNet (85.2 M, 490.9/150.3 GFLOPs) and MCANet (71.2 M, 874.5/375.9 GFLOPs). On the MSAW dataset, DLiteNet achieves the highest accuracy (83.6% IoU, 91.1% F1-score), exceeding the best MCANet baseline by 1.0% IoU and 0.6% F1-score. Furthermore, deployment tests on the Jetson Orin NX edge device show that DLiteNet achieves a low inference latency of 14.97 ms per frame under FP32 precision, highlighting its real-time capability and deployment potential in edge computing scenarios. Full article

(This article belongs to the Special Issue Advances in Multiple Sensor Fusion and Classification for Object Detection and Tracking)

► Show Figures

Figure 1

27 pages, 19042 KB

Open AccessArticle

A Global Distribution-Aware Network for Open-Set Hyperspectral Image Classification

by Fengcheng Ji, Wenzhi Zhao, Qiao Wang and Rui Peng

Remote Sens. 2025, 17(24), 3938; https://doi.org/10.3390/rs17243938 - 5 Dec 2025

Abstract

Recently, developments in hyperspectral image (HSI) classification have brought increasing attention to the challenges of the open-set problem. However, current open-set methods generally overlook the intra-class multimodal structure, making it difficult to comprehensively capture the global data distribution, which in turn reduces their [...] Read more.

Recently, developments in hyperspectral image (HSI) classification have brought increasing attention to the challenges of the open-set problem. However, current open-set methods generally overlook the intra-class multimodal structure, making it difficult to comprehensively capture the global data distribution, which in turn reduces their ability to distinguish known from unknown classes. To address this, we propose a novel global distribution-aware network (GDAN) that jointly performs pixel-wise HSI classification and trustworthy uncertainty-aware identification of unknown class. First, a generative adversarial network (GAN) is employed as the backbone, enhanced with a self-attention (SA) module to capture long-range dependencies across the extensive spectral bands of hyperspectral data. Second, an interpretable open-set HSI classification framework is designed, combining GAN with Markov Chain Monte Carlo (MCMC) to model global distribution by exploring intra-class multimodal structures and estimate predictive uncertainty. In this framework, the traditionally fixed discriminator weights are reformulated as probability distributions, and posterior inference is conducted using MCMC within a Bayesian framework. Finally, accurate categories and predictive uncertainty of ground objects can be obtained through posterior sampling, while samples with high uncertainty are assigned to the unknown class, thus enabling accurate open-set HSI classification. Extensive experiments on three benchmark HSI datasets demonstrate the superiority of the proposed GDAN for open-set HSI classification, yielding overall classification accuracies of 94.6%, 92.6%, and 94.8% in the 200-sample scenario. Full article

(This article belongs to the Special Issue Deep Learning for Spectral-Spatial Hyperspectral Image Classification (2nd Edition))

► Show Figures

Figure 1

20 pages, 3790 KB

Open AccessArticle

Characteristics of Planetary Boundary Layer Height (PBLH) over Shenzhen, China: Retrieval Methods and Air Pollution Conditions

by Yaqi Zhou, Yong Han, Zhiyuan Hu, Qicheng Zhou, Yan Liu, Li Dong and Peng Xiao

Remote Sens. 2025, 17(24), 3937; https://doi.org/10.3390/rs17243937 - 5 Dec 2025

Abstract

The PBLH affects the intensity of the surface turbulence and the state of pollutant dispersion. Current research on PBLH characteristics and their relationship with pollution in coastal megacities remains insufficient. Moreover, existing studies rarely evaluate the consistency of various boundary layer solution methods, [...] Read more.

The PBLH affects the intensity of the surface turbulence and the state of pollutant dispersion. Current research on PBLH characteristics and their relationship with pollution in coastal megacities remains insufficient. Moreover, existing studies rarely evaluate the consistency of various boundary layer solution methods, making it difficult to identify deviations in single methods. So, we conducted enhanced observation experiments in Shenzhen, a megacity in China, between March and July 2023. The characteristics of the PBLH was analyzed by five months of observations from Micro-Pulse Lidar (MPL) and Microwave Radiometer (MWR). Five retrieval methods (Parcel, GRA, STD, WCT, and Theta) were applied for comparative assessment. The results shows that all methods captured similar diurnal patterns. During daytime, the PBLH ranged from 512 to 1345 m, with Theta yielding the highest and STD the lowest average values. At night, PBLH decreased overall, and method-dependent differences persisted. Under different pollution levels, this study also discussion the properties of PBLH using MPL and microwave radiometer. And aerosol optical depth (AOD) and PBLH showed a strong negative correlation, indicating aerosol-induced suppression of boundary layer growth. The study of boundary layer characteristics in coastal megacities can provide reference for atmospheric physics research in economically developed coastal areas. Full article

(This article belongs to the Special Issue Advanced Remote Sensing Approaches for Multi-Scale Atmospheric Components Monitoring and Impact Assessment)

► Show Figures

Figure 1

24 pages, 15414 KB

Open AccessArticle

TAF-YOLO: A Small-Object Detection Network for UAV Aerial Imagery via Visible and Infrared Adaptive Fusion

by Zhanhong Zhuo, Ruitao Lu, Yongxiang Yao, Siyu Wang, Zhi Zheng, Jing Zhang and Xiaogang Yang

Remote Sens. 2025, 17(24), 3936; https://doi.org/10.3390/rs17243936 - 5 Dec 2025

Abstract

Detecting small objects from UAV-captured aerial imagery is a critical yet challenging task, hindered by factors such as small object size, complex backgrounds, and subtle inter-class differences. Single-modal methods lack the robustness for all-weather operation, while existing multimodal solutions are often too computationally [...] Read more.

Detecting small objects from UAV-captured aerial imagery is a critical yet challenging task, hindered by factors such as small object size, complex backgrounds, and subtle inter-class differences. Single-modal methods lack the robustness for all-weather operation, while existing multimodal solutions are often too computationally expensive for deployment on resource-constrained UAVs. To this end, we propose TAF-YOLO, a lightweight and efficient multimodal detection framework designed to balance accuracy and efficiency. First, we propose an early fusion module, the Two-branch Adaptive Fusion Network (TAFNet), which adaptively integrates visible and infrared information at both pixel and channel levels before the feature extractor, maximizing complementary data while minimizing redundancy. Second, we propose a Large Adaptive Selective Kernel (LASK) module that dynamically expands the receptive field using multi-scale convolutions and spatial attention, preserving crucial details of small objects during downsampling. Finally, we present an optimized feature neck architecture that replaces PANet’s bidirectional path with a more efficient top-down pathway. This is enhanced by a Dual-Stream Attention Bridge (DSAB) that injects high-level semantics into low-level features, improving localization without significant computational overhead. On the VEDAI benchmark, TAF-YOLO achieves 67.2% mAP₅₀, outperforming the CFT model by 2.7% and demonstrating superior performance against seven other YOLO variants. Our work presents a practical and powerful solution that enables real-time, all-weather object detection on resource-constrained UAVs. Full article

(This article belongs to the Special Issue Target Detection, Recognition, Tracking, and Positioning Using Remote Sensing and AI Techniques)

► Show Figures

Figure 1

30 pages, 8582 KB

Open AccessArticle

Machine Learning Approaches for Assessing Avocado Alternate Bearing Using Sentinel-2 and Climate Variables—A Case Study in Limpopo, South Africa

by Muhammad Moshiur Rahman, Andrew Robson and Theo Bekker

Remote Sens. 2025, 17(24), 3935; https://doi.org/10.3390/rs17243935 - 5 Dec 2025

Abstract

Alternate (irregular) bearing, characterized by large fluctuations in fruit yield between consecutive years, remains a major constraint to sustainable avocado (Persea americana) production. This study aimed to assess the potential of satellite remote sensing and climatic variables to characterize and predict [...] Read more.

Alternate (irregular) bearing, characterized by large fluctuations in fruit yield between consecutive years, remains a major constraint to sustainable avocado (Persea americana) production. This study aimed to assess the potential of satellite remote sensing and climatic variables to characterize and predict alternate bearing patterns in commercial orchards in Tzaneen, Limpopo Province, South Africa. Historical yield data (2018–2024) from 46 “Hass” avocado blocks were analyzed alongside Sentinel-2 derived vegetation indices (NDVI, GNDVI, NDRE, CIG, CIRE, EVI2, LSWI) and flowering indices (WYI, NDYI, MTYI). To align temporal scales, all VIs and FIs were aggregated into eight quarterly averages from the two years preceding each yield year and spatially averaged across each orchard block. Climatic predictors including maximum temperature (Tmax), minimum temperature (Tmin), vapor pressure deficit (VPD), and precipitation were screened against historical yields to identify critical periods, with June–October emerging as the most influential months, and these variables were aggregated accordingly to match annual alternate bearing patterns. Five machine learning (ML) algorithms—Random Forest, XGBoost, CATBoost, LightGBM, and TabPFN—were trained and tested using a Leave-One-Year-Out (LOYO) approach. Results showed that VPD, Tmin, and Tmax during the flowering period (July–September) were the most influential variables affecting subsequent yields. TabPFN achieved the highest predictive accuracy (Accuracy = 0.88; AUC = 0.95) and strongest temporal generalization. Spectral gradients between flowering and early fruit drop were lower during “on” years, reflecting stable canopy vigor. This combined use of remote sensing and climatic variables in a ML framework represents a novel approach, and the findings demonstrate that integrating remote sensing and climatic indicators enables early discrimination of “on” and “off” years, supporting proactive orchard management and improved yield stability. Full article

(This article belongs to the Special Issue Artificial Intelligence-Based Remote Sensing for Crop Information Extraction and Status Monitoring)

► Show Figures

Figure 1

28 pages, 4643 KB

Open AccessArticle

JM-Guided Sentinel 1/2 Fusion and Lightweight APM-UNet for High-Resolution Soybean Mapping

by Ruyi Wang, Jixian Zhang, Xiaoping Lu, Zhihe Fu, Guosheng Cai, Bing Liu and Junfeng Li

Remote Sens. 2025, 17(24), 3934; https://doi.org/10.3390/rs17243934 - 5 Dec 2025

Abstract

Accurate soybean mapping is critical for food–oil security and cropping assessment, yet spatiotemporal heterogeneity arising from fragmented parcels and phenological variability reduces class separability and robustness. This study aims to deliver a high-resolution, reusable pipeline and quantify the marginal benefits of feature selection [...] Read more.

Accurate soybean mapping is critical for food–oil security and cropping assessment, yet spatiotemporal heterogeneity arising from fragmented parcels and phenological variability reduces class separability and robustness. This study aims to deliver a high-resolution, reusable pipeline and quantify the marginal benefits of feature selection and architecture design. We built a full-season multi-temporal Sentinel-1/2 stack and derived candidate optical/SAR features (raw bands, vegetation indices, textures, and polarimetric terms). Jeffries–Matusita (JM) distance was used for feature–phase joint selection, producing four comparable feature sets. We propose a lightweight APM-UNet: an Attention Sandglass Layer (ASL) in the shallow path to enhance texture/boundary details, and a Parallel Vision Mamba layer (PVML with Mamba-SSM) in the middle/bottleneck to model long-range/global context with near-linear complexity. Under a unified preprocessing and training/evaluation protocol, the four feature sets were paired with U-Net, SegFormer, Vision-Mamba, and APM-UNet, yielding 16 controlled configurations. Results showed consistent gains from JM-guided selection across architectures; given the same features, APM-UNet systematically outperformed all baselines. The best setup (JM-selected composite features + APM-UNet) achieved PA 92.81%, OA 97.95, Kappa 0.9649, Recall 91.42%, IoU 0.7986, and F1 0.9324, improving PA and OA by ~7.5 and 6.2 percentage points over the corresponding full-feature counterpart. These findings demonstrate that JM-guided, phenology-aware features coupled with a lightweight local–global hybrid network effectively mitigate heterogeneity-induced uncertainty, improving boundary fidelity and overall consistency while maintaining efficiency, offering a potentially transferable framework for soybean mapping in complex agricultural landscapes. Full article

(This article belongs to the Special Issue Machine Learning of Remote Sensing Imagery for Land Cover Mapping)

► Show Figures

Figure 1

23 pages, 12696 KB

Open AccessArticle

KADL: Knowledge-Aided Deep Learning Method for Radar Backscatter Prediction in Large-Scale Scenarios

by Dong Zhu, Peng Zhao, Qiang Zhao, Qingliang Li, Jinpeng Zhang and Lixia Yang

Remote Sens. 2025, 17(24), 3933; https://doi.org/10.3390/rs17243933 - 5 Dec 2025

Abstract

Radar backscatter from large-scale scenarios plays a crucial role in remote sensing applications. However, due to the diversity and heterogeneity of the natural environment, traditional empirical methods which rely on simplified physics and a limited set of parameters, fail to adequately model land [...] Read more.

Radar backscatter from large-scale scenarios plays a crucial role in remote sensing applications. However, due to the diversity and heterogeneity of the natural environment, traditional empirical methods which rely on simplified physics and a limited set of parameters, fail to adequately model land backscatter, thus exhibiting significant limitations. While purely data-driven deep learning (DL) methods offer flexibility, they often struggle to ensure physical consistency and effectively generalize to unseen scenarios. To address these issues, we propose a novel knowledge-aided (KA) DL-based method (called KADL) in this paper for predicting the radar backscatter from large-scale scenarios. The proposed KADL is implemented in three parts. First, based on multi-source remote sensing data, the dielectric properties of land surface, i.e., soil moisture and leaf area index (LAI) are incorporated as priori physical knowledge into the Multi-Feature Clutter Dataset (MFCD) to obtain initialized input. Second, a knowledge perception module (KPM) is introduced into the cascaded deep neural network (DNN) solver to exploit the representative features within the inputs. Third, an efficient knowledge-weighted fusion (KWF) strategy is developed to further enhance the discriminative features and simultaneously suppress the non-informative features. For better comparison, we refitted the specific empirical models based on the measured data and introduced an advanced nonhomogeneous terrain clutter model (termed ANTCM) derived from our previous work. Extensive experiments conducted on the measured data demonstrate that KADL achieves a root mean square error (RMSE) of 4.74 dB and a mean absolute percentage error (MAPE) of 8.7% on independent test data. Furthermore, KADL exhibits superior robustness, with a standard deviation of RMSE as low as 0.18 dB across multiple trials. All these results validate the superior accuracy, robustness, and generalization ability of KADL for large-scale backscatter prediction. Full article

(This article belongs to the Section AI Remote Sensing)

► Show Figures

Figure 1

18 pages, 2281 KB

Open AccessArticle

Evaluating Remotely Sensed Spectral Indices to Quantify Seagrass in Support of Ecosystem-Based Fisheries Management in a Marine Protected Area of Western Australia

by Nick Konzewitsch, Lara Mist and Scott N. Evans

Remote Sens. 2025, 17(24), 3932; https://doi.org/10.3390/rs17243932 - 5 Dec 2025

Abstract

Understanding and monitoring benthic habitat distribution is essential for implementing ecosystem-based fisheries management (EBFM). Satellite remote sensing offers a rapid and cost-effective approach to marine habitat assessments; however, its application requires context-specific adjustment to account for environmental variability and differing study aims. As [...] Read more.

Understanding and monitoring benthic habitat distribution is essential for implementing ecosystem-based fisheries management (EBFM). Satellite remote sensing offers a rapid and cost-effective approach to marine habitat assessments; however, its application requires context-specific adjustment to account for environmental variability and differing study aims. As such, predictor variables must be tailored to the specific site and target habitat. This study uses Sentinel-2 Level 2A surface reflectance satellite imagery and stability selection via Random Forest Recursive Feature Elimination to assess the importance of remote sensing indices for mapping moderately deep (<20 m) seagrass habitats in relation to the Marine Stewardship Council-certified Western Australia Enhanced Greenlip Abalone Fishery (WAEGAF). Of the seven indices tested, the Normalised Difference Aquatic Vegetation Index (NDAVI) and Depth Invariant Index for the blue and green bands were selected in the optimal model on every run. The kernelised NDAVI and Water-Adjusted Vegetation Index also scored highly (both 0.92) and were included in the final classification and regression models. Both models performed well and predicted a similar cover and distribution of seagrass within the fishery compared to the surrounding area, providing a baseline and supporting EBFM of the WAEGAF within the surrounding marine protected area. Importantly, the use of indices from freely accessible ready-to-use satellite products via Google Earth Engine workflows and expedited ground truth image annotation using highly accurate (0.96) automatic image annotation provides a rapidly repeatable method for delivering ecosystem information for this fishery. Full article

(This article belongs to the Special Issue Advanced Applications of Remote Sensing in Monitoring Marine Environment (Second Edition))

► Show Figures

Figure 1

20 pages, 2107 KB

Open AccessArticle

Evaluating the Performance of the STEMMUS-SCOPE Model to Simulate SIF and GPP Under Drought Stress Using Tower-Based Observations of Maize

by Mengchen Li, Xinjie Liu and Liangyun Liu

Remote Sens. 2025, 17(24), 3931; https://doi.org/10.3390/rs17243931 - 5 Dec 2025

Abstract

With advancements in solar-induced fluorescence (SIF) observation technology and the evolution of vegetation radiative transfer models, SIF signals can now be more effectively interpreted and leveraged from a mechanistic perspective. This, in turn, facilitates a deeper understanding of the mechanistic link between SIF [...] Read more.

With advancements in solar-induced fluorescence (SIF) observation technology and the evolution of vegetation radiative transfer models, SIF signals can now be more effectively interpreted and leveraged from a mechanistic perspective. This, in turn, facilitates a deeper understanding of the mechanistic link between SIF and photosynthesis. Considering the impact of water stress on terrestrial ecosystems, this paper simulated SIF and gross primary productivity (GPP) values using the STEMMUS-SCOPE model at half-hour scales from 2017 to 2023 at the Daman site. The simulation results were compared and validated against flux tower observations and SCOPE model outputs. Taking advantage of irrigation events in the semi-arid irrigated farmland, we assessed the accuracy of STEMMUS-SCOPE in simulating SIF and GPP under drought stress, as well as its capability to quantitatively analyze the impacts of water stress on SIF and GPP. The results show that the accuracy of the SIF and GPP values simulated by the STEMMUS-SCOPE model is higher than that of the SCOPE model. The averaged R² and RMSE between the SIF simulated by STEMMUS-SCOPE model and the observed SIF values are 0.66 and 0.29 mW m⁻² nm⁻¹, and the averaged R² and RMSE between the GPP simulated by the STEMMUS-SCOPE model and the observed GPP values from 2017 to 2023 are 0.88 and 4.93 µmol CO₂ m⁻² s⁻¹, respectively. Especially under relatively drought conditions, the R² between the SIF simulated values and observed values is 0.84, and the R² between the GPP simulated values and observed values is 0.96. By further combining soil moisture content (SMC) and canopy conductance (Gs) analyses, we found that the response of the STEMMUS-SCOPE simulations under water stress was consistent with previous findings on the impacts of water deficits, thereby confirming the model’s reliability for drought conditions. Under drought stress, the decline in fluorescence emission efficiency (ΦF) with decreasing Gs and SMC was smaller than that of the light use efficiency (LUE). Therefore, the STEMMUS-SCOPE model is promising for investigating the SIF–GPP relationship under drought stress. Full article

► Show Figures

Figure 1

Journal Description

Remote Sensing

Latest Articles

Journal Menu

Journal Browser

Highly Accessed Articles

Latest Books

E-Mail Alert

News

Topics

Conferences

Special Issues

Topical Collections

Further Information

Guidelines

MDPI Initiatives

Follow MDPI