Conference Agenda

Smoothing aerial LiDAR point clouds is challenging, because they are often noisy, irregularly sampled, and sparse, as well as their inherent high degrees of freedom. Classic methods struggle on such datasets as they were designed for regularly sampled, dense datasets with moderate noise. To address the challenge, this paper proposes a constrained point cloud model with one degree of freedom. The point cloud model incorporates the sensing directions stored in the full waveform LiDAR datasets, and has theoretical advantages in terms of the statistical error bound for normal estimation. Based on the point cloud model, the directional total least square is formulated as a regularized convex optimization problem for points estimation on a tangent plane. Moreover, a non-convex regularizer along with the non-convex regularized directional total least square is proposed to improve the estimation

quality. To solve the proposed optimization problems, an accelerated Douglas-Rachford splitting algorithm is introduced. The proposed methods demonstrate better performances on simulated two-dimensional point clouds in terms of improved root-mean square-

error. For three-dimensional aerial LiDAR point clouds, implemented under the Savitzky-Golay filter framework with local

smoothness prior, the proposed methods demonstrate more smoothing power and robustness than the classic method.

Urban mapping for planning and monitoring requires high-resolution spatial data, especially in areas with high landcover diversity. Airborne LiDAR Scanning (ALS) provides accurate 3D point cloud data, but its classification remains challenging due to computational complexity, irregular point distribution, noise, mislabeling and outliers in the dataset. These challenges are amplified in dense urban environments with mixed vegetation and infrastructure. Existing local context-based classification methods consider all points equally, overlooking the impact of their spatial position of the point in the dataset. To address this, we propose a dynamic local context-based point confidence-based optimization that improves classification accuracy by leveraging the spatial context of each point. This approach selects points based on confidence levels derived from position indices in training data and predicted by binary classifiers in test data to enhance robustness of classifier. We evaluated the proposed approach using boosting-based machine learning classifiers on two datasets: Thiruvananthapuram Aerial LiDAR Dataset (TALD) from India and the ISPRS 3D semantic labeling dataset from Vaihingen, Germany. The results showed 90.3% accuracy on TALD and 90.0% on Vaihingen, achieving a 2-4% improvement over conventional local context-based classification.

Accurate interpretation and segmentation of 3D point clouds in real-world urban environments is a critical challenge in geospatial analysis, particularly due to the complexity of real-world scenes, inevitable data uncertainties, and potential annotation errors. This paper proposes a confidence-aware deep learning framework to refine the segmentation accuracy of real-world point cloud data. By incorporating multi-source information, such as aerial imagery, and embedding geospatial prior knowledge, this framework models data uncertainty through point-wise confidence scores. Besides, we design an iterative online learning strategy, allowing the network to improve both its predictions and the quality of training labels. Extensive experiments on large-scale airborne laser-scanned data demonstrate that our framework effectively enhances training data by reducing label noise and improving annotation quality, which leads to more robust, generalizable model performance. Our source code is publicly available at https://github.com/AutumnMoon00/RefineNet.

LiDAR sensors incorporated within modern smartphone and tablet devices enable relatively quick and inexpensive collection of ground-based LiDAR data applicable for ground truth mapping as needed for modelling understory vegetation. However, this LiDAR data often requires conversion and processing prior to research use. This study presents a workflow with algorithms utilizing structured query language (SQL) to efficiently process detailed rasterized features from LiDAR data collected by an iPhone Pro Max via the ForestScanner app. After transformation of the LiDAR data, SQL has been employed to voxelize the LiDAR data from which rasterized features have been derived. Various cell sizes for voxels and subsequent pixels have been investigated, leading to a recommended spatial resolution of 0.05 m for cell size dimension. SQL provides precise control for advanced querying to process ground-based LiDAR data for vegetational modelling applications.

In the era of information revolution, with data being the fuel of AI and analytics, efficient information extraction from LiDAR point clouds becomes indispensable for solving real-world problems and aiding decision-making in geospatial domain. Despite having geometric richness, the massive LiDAR point clouds are not only computationally demanding but also lack inherent semantics. The lack of semantics in LiDAR constrains effective data analysis. This paper presents a novel workflow by incorporating Deep Learning derived embeddings as attributes in the geospatial database for the spatio-semantic querying on Aerial LiDAR point clouds. This work leverages AI-based indexing, such as IVFFlat(Inverted File Index with Flat Quantization) on LiDAR point clouds for fast retrieval of queries. The pgPointCloud and pgVector extensions of PostgreSQL aid in importing point clouds into the database and performing similarity-based query retrieval on the embedding space of the point clouds. The methodology developed in this paper explores how semantic embeddings can handle inadequate semantics of point clouds by enabling direct and complex 3D intelligent queries within the database environment, thereby overcoming the limitations of traditional LiDAR representations. Few queries presented in this paper highlight the applications of this proposed framework in individual tree detection, tree species identification, utility management, urban planning and anomaly detection.

The development of global geographic information resource products involves massive information processing of PB-level multimodal spatiotemporal data, and faces technical challenges brought by the global scale. In response to the challenges, we have made technological innovations to break through the key technologies for the development of global geographic information data products. With the main themes of "intelligent interpretation of typical elements, multi-source geographic data mining, and intelligent hybrid compilation", we have conducted and completed the overall technical research on the construction of global geographic information resources, formed an autonomous construction capability. Firstly, through crowd-sourced data mining and fusion technology to achieve content information extraction and knowledge fusion; Secondly, using multiple source data features, fast automatic extraction and integration of elements based on deep learning models was processed, and produce digital line graph data based on intelligent hybrid compilation. Based on the automatic feature extraction technology of deep learning, the production of digital line graph data products has been updated, and the accuracy evaluation has reached over 85%.

This paper addresses the heavy reliance on manual operations in control point acquisition for UAV photogrammetry and proposes an encoded control point deployment and recognition method based on a Light-Marker UAV (LMUAV). Conventional approaches rely on manual placement of control points and manual identification and measurement in images for aerial triangulation, resulting in low efficiency. To address this limitation, an LMUAV equipped with an LED array actively broadcasts its positional information as quaternary optical signals. The observing UAV performs coarse localization of the target region by integrating communication priors with the imaging model, followed by light spot segmentation and graph construction within the region of interest (ROI). Node correspondences are then recovered by constructing a template graph and an observation graph and applying Reweighted Random Walks (RRWM) graph matching. The matching robustness is further enhanced by incorporating directional point constraints and RANSAC-based geometric filtering. Based on the recovered correspondences, the encoded information is decoded through color recognition and validation, enabling automatic control point recovery. Experimental results in a cross-flight-line scenario with a single target UAV demonstrate that the proposed method achieves stable node matching and encoding–decoding, with a sequence-level accuracy of 76.32%, and a final effective decoding rate of 71.05%, while maintaining centimeter-level positioning accuracy, thereby validating its effectiveness for automatic control point acquisition in UAV mapping.

Automated and selective harvesting of fruits is increasingly vital due to high costs and seasonal labor shortages in advanced economies. This paper explores 6D pose estimation of strawberries using synthetic data generated through a procedural pipeline for photorealistic rendering. We utilize the YOLOX-6D-Pose algorithm, a single-shot method leveraging the YOLOX backbone, known for its balance of speed and accuracy and its suitability for edge inference. To counter the lack of training data, we develop a robust and flexible pipeline for generating synthetic strawberry data from various 3D models in Blender, focusing on enhancing realism compared to prior efforts, thus providing a valuable resource for training pose estimation algorithms.

Quantitative evaluations show that our models achieve comparable accuracy on both the NVIDIA RTX 3090 and Jetson Orin Nano across several ADD-S metrics, with the RTX 3090 offering superior processing speed. However, the Jetson Orin Nano is particularly effective for resource-constrained environments, making it ideal for deployment in agricultural robotics. Qualitative assessments further validate the model's performance, demonstrating accurate pose inference for ripe and partially ripe strawberries, although challenges remain in detecting unripe specimens. This highlights opportunities for future enhancements, particularly in improving detection for unripe strawberries by exploring color variations. Moreover, the presented methodology can be easily adapted for other fruits, such as apples, peaches, and plums, broadening its applicability in agricultural automation.

Photogrammetric 3D reconstruction has long relied on traditional Structure-from-Motion (SfM) and Multi-View Stereo (MVS) methods, which provide high accuracy but face challenges in speed and scalability. Recently, learning-based MVS methods have emerged, aiming for faster and more efficient reconstruction. This work presents a comparative evaluation between a representative traditional MVS pipeline (COLMAP) and state-of-the-art learning-based approaches, including geometry-guided methods (MVSNet, PatchmatchNet, MVSAnywhere, MVSFormer++) and end-to-end frameworks (Stereo4D, FoundationStereo, DUSt3R, MASt3R, Fast3R, VGGT). Two experiments were conducted on different aerial scenarios. The first experiment used the MARS-LVIG dataset, where ground-truth 3D reconstruction was provided by LiDAR point clouds. The second experiment used a public scene from the Pix4D official website, with ground truth generated by Pix4Dmapper. We evaluated accuracy, coverage, and runtime across all methods. Experimental results show that although COLMAP can provide reliable and geometrically consistent reconstruction results, it requires more computation time. In cases where traditional methods fail in image registration, learning-based approaches exhibit stronger feature-matching capability and greater robustness. Geometry-guided methods usually require careful dataset preparation and often depend on camera pose or depth priors generated by COLMAP. End-to-end methods such as DUSt3R and VGGT achieve competitive accuracy and reasonable coverage while offering substantially faster reconstruction. However, they exhibit relatively large residuals in 3D reconstruction, particularly in challenging scenarios.

This study presents a comprehensive evaluation of six deep learning models for building exterior crack detection using UAV imagery. Our framework systematically compares Standard U-Net, ResNet34-UNet, UNet-Attention, UNet-Residual, HybridUNet, and TransUNet through rigorous ablation experiments. The models were trained on dedicated drone-captured crack imagery and evaluated using multiple loss functions and performance metrics.

Results show that TransUNet achieves optimal performance (87.66% F1 Score, 90.43% Precision, 89.99% Recall) by leveraging Transformer-based global context modeling. Notably, the performance gap among all six models remains minimal (<0.5% F1 Score difference), suggesting limited returns from increased architectural complexity alone. F1 Loss demonstrates the most balanced performance across architectures, while Focal-Dice-Loss offers superior optimization stability.

The study provides practical guidance for model selection: TransUNet with F1 Loss suits high-accuracy requirements, while simpler attention-enhanced U-Net variants offer cost-effective solutions for large-scale applications. These findings advance intelligent crack detection methodologies and emphasize balancing accuracy with computational efficiency for real-world structural health monitoring.

Unmanned Aerial Vehicles (UAVs) have contributed to a wide range of applications, becoming faster and more sustainable nowadays. However, given the significant increase in the number of UAVs, concerns regarding operational safety have grown.

Autonomous UAV path planning must ensure compliance with safety requirements. This study proposes a real-time path replanning method focused on ensuring compliance with regulations governing UAV operations. Considering no-fly zones (NFZs) defined by both static (buildings) and dynamic (people) obstacles, a low-cost and replicable solution was implemented in four main steps: 3D offline path planning using the A* algorithm and Digital Elevation Models; human detection in UAV imagery using the YOLO11m model; estimation of the person’s 3D coordinates using Monoplotting; and experiments of real-time path replanning. During flight

execution, imagery acquired by the UAV is transmitted to a server and, if a person is detected, path replanning is performed. The replanned route is then sent to the UAV controller to be executed via an SDK-based application. For flights at reduced speeds, the proposed method demonstrated feasibility in a computational environment (replanning time of 2.79 s). Simulated flight execution using the DJI Mobile SDK was successful. However, when relying on data transmission over Wi-Fi, the replanning duration on a local server (17.96 s) remained unsuitable for real-time operations. As future work, alternative solutions should be explored to ensure real-time processing. Despite the challenges, this study contributes by validating the open and free DJI MSDK application for path execution in a simulated environment, integrated with a listener application.

Point cloud registration involves aligning one point cloud with another or with a three-dimensional (3D) model, enabling the integration of multimodal data into a unified representation. This is essential in applications such as construction monitoring, autonomous driving, robotics, and virtual or augmented reality (VR/AR).With the increasing accessibility of point cloud acquisition technologies, such as Light Detection and Ranging (LiDAR) and structured light scanning, along with recent advances in deep learning, the research focus has increasingly shifted towards downstream tasks, particularly point cloud-to-model (PC2Model) registration. While data-driven methods aim to automate this process, they struggle with sparsity, noise, clutter, and occlusions in real-world scans, which limit their performance. To address these challenges, this paper introduces the PC2Model benchmark, a publicly available dataset designed to support the training and evaluation of both classical and data-driven methods. Developed under the leadership of ICWG II/Ib, the PC2Model benchmark adopts a hybrid design that combines simulated point clouds with, in some cases, real-world scans and their corresponding 3D models. Simulated data provide precise ground truth and controlled conditions, while real-world data introduce sensor and environmental artefacts. This design supports robust training and evaluation across domains and enables the systematic analysis of model transferability from simulated to real-world scenarios. The dataset is publicly accessible at: https://zenodo.org/records/17581812.

The cryosphere responds to variations in the climate. Monitoring glaciers requires research into their dynamics. The surface velocity of the Gangotri glacier was obtained in this study using the Sentinel-1 dataset. Modifying the laminar flow model improved estimates of ice thickness. Moreover, the glacier mass balance has been calculated using changes in ice thickness between 2017 and 2022. An average velocity of 0.09 m/day was observed with stretches from 0.12 to 0.23 m/day in the central trunk. A mean thickness of 189 ± 17.01 m was determined for the glacial ice. The thickest areas, with the least drag, were measured to be 587 ± 52.83 m in the middle part. Negative mass rates of -1.3 to -0.5 m.w.e./year were observed for the glacier system (with thickness changes of -3 to -0.6 m/year) due to the glacier's decreased thickness throughout time.

The Himalayan region hosts the largest reservoir of snow and ice outside the polar regions. However, ongoing climate change has resulted in widespread glacier retreat, heightening the frequency and magnitude of extreme events, including flashfloods, landslides, and Glacier Lake Outburst Floods. The Beas Basin in the northwestern Himalaya exemplifies this vulnerability, where cryospheric transformations directly threaten downstream communities, hydropower systems, and infrastructure. Despite its critical importance, long-term basin-scale records remain limited. Therefore, this study investigates the long-term cryospheric evolution of the Beas Basin and identifies emerging glacial lakes using an integrated remote-sensing and modelling-approach. Glacier mass balance from 1951 to 2024 was estimated using an Improved Accumulation-Area-Ratio method, incorporating equilibrium-line-altitudes derived from ASTER-DEM and meteorological data, alongside glacier extents from Landsat and Sentinel imagery. Current glacier ice reserves were quantified using laminar-flow and volume–area scaling methods, with surface velocities derived from sub-pixel Landsat image correlation, and slope from DEMs. Future glacial lake formation was assessed using the HIGTHIM tool, which integrates ice thickness, bed topography, and moraines. Results indicate a mean area-weighted mass balance of –0.46±0.26m.w.e.a⁻¹, corresponding to 17.75Gt cumulative ice loss (~48% of glacier-stored mass) since 1951 and a current ice reserve of 19.60±3.5 Gt. Sixty-three potential glacial lake sites were identified, with four existing lakes projected to expand, totalling 122±22 million-m³of water. These findings reveal extensive cryospheric reorganisation, with significant implications for hydrology, water security, and hazard management. The study demonstrates the value of combining satellite observations with process-based modelling for monitoring Himalayan glacier dynamics in data-sparse regions.

The intrusion of relatively warm ocean waters beneath Antarctic ice shelves is a key driver of basal melting and strongly influences ice-shelf stability. However, previous studies investigating warm-water pathways have largely relied on single-source datasets, such as ship-based Conductivity–Temperature–Depth (CTD) measurements, which are spatially sparse and limited to a few well-surveyed regions. Recent advances in multi-source remote sensing datasets provide new opportunities to address these limitations. In this study, a multi-source remote sensing–based framework is developed to identify potential pathways of relatively warm water intrusion beneath Antarctic ice shelves and to quantify the associated basal melting. The Moscow University Ice Shelf (MUIS) is used as a case study. Across the continental shelf, CTD observations, sub-ice-shelf bathymetry, and modeled ocean circulation are integrated to infer potential intrusion routes. At the ice-shelf front and base, EN4 reanalysis data are used to characterize seawater properties, while satellite-derived basal melt products are applied to analyze spatial and vertical patterns of basal melting. Results indicate that relatively warm water is mainly concentrated at depths of 300–500 m, coinciding with bathymetric depressions that facilitate its intrusion beneath MUIS. Enhanced basal melting occurs near the ice front and grounding line, primarily within the upper 0–500 m of the ice-shelf draft, with an average melt rate of ~6 m yr⁻¹. The proposed framework provides a transferable approach for investigating ocean-driven melting beneath Antarctic ice shelves.

The Antarctic Ice Sheet (AIS), the largest terrestrial ice mass on Earth, contains approximately 90% of the planet's total ice volume. This study quantifies ice discharge and associated uncertainties in AIS estimates through Input-Output method, evaluating the impact of flux gate locations on discharge magnitude and measurement uncertainty. Through analysis of key factors contributing to discharge uncertainty, we propose a gate positioning strategy that optimizes the balance between proximity to the grounding line and uncertainty minimization.

NASA’s ICESat-2, a single-photon lidar satellite launched in 2018, has for six years delivered pole-wide elevation data with <0.4 cm/yr precision. To verify and exploit these data over Antarctica, we built a “space-air-ground” calibration chain. (1) A cross-track array of corner-cube retro-reflectors (CCRs) was installed at Kunlun, Taishan and Zhongshan stations; one deployment captures both ascending and descending passes, doubling efficiency. GNSS-PPP/RTK solutions overcome the absence of fixed reference points and position CCRs to within 1 cm; comparison with ICESat-2 tracks shows sub-4 cm vertical accuracy, confirming stable on-orbit performance. (2) UAV photogrammetry during the 36th CHINARE expedition produced 5 cm-resolution DEMs of crevassed ice margins at Zhongshan/Prydz Bay. Fused with RTK ground control, these reveal ICESat-2 planimetric offsets of 2–5 m and serve as “truth” for a new Photon-Cloud algorithm that corrects slope-induced positioning errors and extends the mission’s utility in rugged terrain. (3) Whole-continent cross-over analysis of repeat tracks shows millimetre-level consistency between ascending and descending orbits; an improved cross-track model extracts robust elevation-change time series for stable ice interiors. The integrated framework provides ICESat-2 Antarctic accuracy metrics, refined processing tools and a transferable protocol for future polar photon-counting altimetry missions.

Accurate mapping of exposed rock is fundamental for cryospheric and geospatial analyses in Antarctica, yet existing products are of limited resolution and tend to underestimate true rock exposure. We present a weakly supervised deep-learning framework that refines existing rock masks by combining Sentinel-2 multispectral imagery with elevation and slope data from the Reference Elevation Model of Antarctica (REMA). A U-Net with eight input channels (six spectral bands, elevation, slope) is trained using imperfect Landsat- and GeoMap based labels. Trained on data from the Antarctic Peninsula, the model produces a 10~m rock mask that delineates small and shaded outcrops more effectively than existing datasets. While quantitative evaluation is constrained by imperfect reference data, qualitative inspection indicates improved rock–snow separation. The workflow is fully automated, requires no manual annotation, and scales efficiently to all rock-hosting regions of the continent reachable by Sentinel-2 multispectral coverage. Beyond rock mapping, the framework is transferable to other scenarios with incomplete or uncertain reference data, such as vegetation, snow, or water mapping. The resulting rock mask for complete Antarctica, together with the trained model and preprocessing scripts, will be released to support reproducible large-scale mapping and future cryospheric research.

Under the background of city physical examination and assessment in territorial spatial planning, urban livability has become a focus of interest. Urban livability reflects residents' overall satisfaction with their living environment. Previous studies have been constrained by issues such as low data precision, coarse spatial scales, and limited practical applicability. To address these limitations, this study developed a refined livability evaluation framework by multi-source remote sensing data, with a primary emphasis on high-resolution domestic satellite imagery, including Gaofen (GF-1) and Ziyuan (ZY-3). Integrated with Suomi NPP night-time light data and socio-economic datasets, the research assessed four key dimensions, which were safety and resilience, residential comfort, recreation convenience, and quality and vitality in the city of Wuhan and Yibin at a detailed kilometer-grid scale. Results revealed distinct spatial patterns of urban livability of the two cities: Wuhan's central urban areas exhibited higher, more clustered livability, driven largely by quality and vitality, whereas Yibin showed a more fragmented pattern with strengths in recreation convenience but relative weaknesses in residential comfort and urban vitality. This study underscores the significant value of high-resolution, multi-source remote sensing data in enabling precise, spatially explicit livability analysis, thereby providing a scientific basis for targeted spatial planning and urban quality enhancement.

Dengue transmission in rapidly urbanising tropical cities is shaped by the combined influence of climate variability, urban morphology, and short-term surface water dynamics. This study develops a remote sensing and GIS-based framework to investigate the interaction between built-up density, waterlogging, and dengue incidence in Chennai from 2021 to 2023. Multi-source datasets, including Sentinel-2 imagery, NICFI high-resolution LULC, NDVI, and NDWI indices, Google Open Buildings footprints, IMD daily climate variables, and geocoded dengue case records, were integrated into a harmonised spatial grid for systematic analysis. Waterlogging-prone zones were delineated using a Sentinel-2 water-frequency method to capture the post-rainfall surface water accumulation rather than only persistent water bodies. Spatial clustering of dengue cases was examined using kernel density estimation, Global and Local Moran’s I, and Getis-Ord Gi*, revealing strong spatial autocorrelation and persistent hotspots in older, densely built neighbourhoods such as Kodambakkam, Adyar, Guindy, Saidapet, and Velachery, where compact built-up patterns and drainage limitations facilitate vector breeding. Peripheral areas showed weaker clustering and lower disease intensity. To assess the climatic influences, a Distributed Lag Non-linear Model (DLNM) was employed to quantify the delayed and non-linear effects of rainfall, maximum temperature, and minimum temperature on dengue incidence. Results showed notable lagged responses, with rainfall and minimum temperature exhibiting strong delayed associations aligned with mosquito development and viral incubation cycles. By integrating climatic, hydrological, and urban structural metrics, this study provides a replicable geospatial workflow for identifying micro-scale dengue-risk environments, supporting evidence-based vector-control strategies and climate-resilient urban planning in tropical cities.

Ticks are key vectors of human and animal disease, with Borrelia burgdorferi sensu stricto, the causative agent of Lyme disease, posing the greatest risk in North America. In Canada, Lyme disease cases are rising as the blacklegged tick (Ixodes scapularis) expands northward, driven by climate change, land cover shifts, and host movement. The Kingston, Frontenac, Lennox and Addington (KFL&A) region is a well-established hotspot, highlighting the importance of mechanistic models that realistically represent heterogeneous environmental drivers of transmission.

This study integrates multi-sensor Earth observation (MODIS, GEDI, Landsat) with climate, habitat, and ecological data to improve mechanistic tick phenology models. A hierarchical framework incorporates microclimate, landscape, and regional variables, enabling assessment of how sensor type, spatial resolution, and environmental gradients influence seasonal tick activity predictions. Model calibration and validation use field-collected tick and pathogen data, supplemented by citizen science observations.

By systematically linking EO to disease modeling, this approach improves the representation of environmental drivers, enhances predictive performance, and supports public health planning. The framework is transferable to other vector-borne diseases, advancing the integration of remote sensing into epidemiological forecasting at regional to national scales.

Illegal waste disposal sites pose significant ecological and public-health risks yet remain difficult to track with traditional field inspections. We propose a multi-agent detection framework that fuses textural, spectral, and contextual cues from medium-resolution satellite imagery for this work. Three specialised agents - Waste-Pile, Road, and Industry detectors - are implemented as YOLO (You Only Look Once) convolutional models that generate partial hypotheses, which are then hierarchically aggregated through rule weights learned from expert-labelled samples. The system provides an interpretable set of object relations, allowing regulators to trace how individual cues contribute to the final decision. The method was validated on an independent test area near Taromske (Dnipropetrovsk region, Ukraine) and corroborated by ground surveys. Joint aggregation raised the posterior probability of the primary target cluster from 0.27 (single-detector confidence) to 0.91, while maintaining robustness to label noise and heterogeneous sensor characteristics. Compared with conventional CNN baselines, the proposed approach delivers three key advantages: explicit explainability of outputs, transferability to 10 m spatial resolution without extensive retraining, and seamless integration of heterogeneous evidence sources. The proposed framework can serve as a cost-effective backbone for regional and national waste-monitoring systems. Future work will focus on near-real-time processing of Sentinel-2 time series, incorporation of hyperspectral and thermal methane indicators to assess remediation stages, and extension of the array of features to other anthropogenic disturbances such as open-pit mining and construction debris.

Based on traditional PS-InSAR technology, this study proposes a building elevation estimation method based on long and short baseline iteration. It utilizes long-temporal SAR images for multiple iterations to calculate building heights, which are used as prior information. Combined with the Interferometric Point Target Analysis (IPTA) method, it inverts building deformation information. The K-means clustering method is employed for PS point clustering analysis, classifying PS points with similar deformation trends and mapping them to buildings. A building safety risk assessment system is established, which comprehensively evaluates the cumulative deformation amount and deformation rate of both the building structure and its foundation. In this paper, the feasibility of the above method is verified by an example. The deformation of 9442 buildings is extracted in the study area, of which 245 buildings are in a high security risk state, and 2 buildings are in a high security risk state. Through this study, it can provide comprehensive auxiliary decision-making reference data covering macro wide-area and micro single buildings for urban construction management departments.

Beef cattle farming is a significant activity in Brazil, and forage quality has a direct impact on animal performance. However, traditional methods for estimating dry mass, which involve cutting, drying and weighing plant material, are slow and labor-intensive. UAVs equipped with multispectral sensors, such as the DJI Mavic 3M, offer a faster and more scalable alternative for monitoring mixed-forage pastures. This study estimates the dry mass of forage mixtures using multispectral UAV data in two scenarios: (i) using only spectral information and (ii) combining spectral data with canopy height measured in the field. Model performance was evaluated using R², RMSE, and percentage error. The multispectral-only model explained 55% of dry mass variability (720.56 kg/ha; 23.67%), while adding canopy height improved performance to 80% and reduced the error to 589.41 kg/ha (19.36%). Results show that canopy height enhances the accuracy and operational potential of UAV-based methods for estimating dry mass in mixed-forage areas.

Mining companies are continuously looking for cost efficient methods to monitor the success of their rehabilitation efforts. Although open access satellite imagery is available at regular temporal intervals, its usefulness for grassland biodiversity monitoring has been questioned due to its coarse spatial resolution with respect to the species size. To compensate for the low spatial resolution, previous studies have successfully explored the benefits of using a multitemporal set of Sentinel-2 (S2) images. However, unless the temporal patterns are studied as a whole, some of the phenological information such as growth rates are lost, and delayed snow cover may spread events like growth onset over multiple dates between plots. This study aims to explore the added value of temporal fitting of Sentinel-2 time series (ts) over existing baseline models applied using the full time series as such. Our set of temporal features included functional components, harmonic decomposition, frequency decomposition, and phenological metrics. Out of the compared models, the Random Forest regression model using a set of fitted temporal features achieved the highest holdout prediction accuracy (R2 = 0.36, RMSE = 3.87, relative RMSE = 0.20) and cross-validation accuracy similar to the baseline models. However, all the compared regression models underestimated extreme plant diversity to some extent. Future studies should account for varying vegetation cover and terrain features by incorporating auxiliary data.

Woody plant encroachment (WPE) threatens grassland ecosystems across the Canadian Prairies, causing grassland biodiversity loss with substantial economic impacts due to reduced forage production. While remote sensing offers scalable monitoring capabilities, existing approaches lack frameworks for distinguishing shrub and tree encroachment and often require extensive ground truth data. This study developed an ensemble machine learning framework integrating Sentinel-1 SAR and Sentinel-2 optical imagery with UAV-derived training data to map fractional shrub and tree cover across Saskatchewan's Aspen Parkland and Moist Mixed Grassland ecoregions, SK. A stacking ensemble combining Random Forest, Support Vector Machine, XGBoost, and Artificial Neural Network models with Ridge regression meta-learning outperformed individual algorithms, achieving mean R² values of 0.65 for shrub and 0.68 for tree cover prediction. Multi-scale training incorporating features at 10, 30, 50, and 70 m resolution improved performance by 15% for shrub and 24% for tree compared to single-scale approaches. Feature importance analysis revealed that shrub detection relied primarily on red-edge bands and moisture indices, while tree detection depended heavily on SAR backscatters. Quantile histogram matching enabled successful model transfer from Foam Lake Community pasture to Aberdeen Community Pasture, with resulting maps indicating that total WPE exceeded 50% in both study areas, with shrubs occupying 23.7% (Foam Lake) and 18.5% (Aberdeen) of both regions at rates higher than 5% shrub cover. The present framework provides a scalable, cost-effective approach for operational woody encroachment monitoring, enabling early detection and targeted functional management interventions to preserve grassland ecosystems.

Fractional vegetation cover (FVC) is a key land surface parameter describing vegetation abundance and structure, defined as the fraction of the ground area occupied by vegetation when viewed from nadir. FVC provides essential insights into ecosystem condition, productivity, and disturbance, making it a critical variable for biodiversity monitoring and habitat assessment. However, generating accurate and repeatable FVC estimates remains challenging due to scale effects, spatial resolution constraints, and inconsistencies in available validation data across time and space. This research develops a machine learning (ML) framework for large-scale FVC estimation that addresses these challenges by combining multi-sensor Earth observation data and Active Learning (AL) model refinement techniques. The ML framework is applied within key wood bison habitat in northern Alberta, focusing on mapping six vegetation components: spruce, pine, deciduous, shrub, herbaceous, and moss. The approach integrates Sentinel-1, Sentinel-2, Landsat-9, and GLO-30 data, optimized through feature selection and ensemble-based Random Forest modeling. The resulting FVC maps achieved strong predictive performance (R² = 0.50–0.88) and capture fine-scale spatial variability in vegetation composition. The ML pipeline provides a scalable and adaptive framework for FVC estimation that supports provincial landcover updates, improves understanding of wood bison habitat features, and contributes to ongoing ecosystem monitoring and conservation planning across boreal Alberta.

Wildlife monitoring from aerial imagery often requires precise animal localization under practical constraints where only object counts are needed. Traditional detection methods rely on bounding-box annotations, introducing unnecessary cognitive load for small objects spanning only a few dozen pixels. This work introduces DINOKey, a modified DINO transformer-based detector adapted to operate natively on point annotations rather than bounding boxes. Key contributions include: (1) architectural modifications to the DINO decoder, detection head, and denoising queries to directly predict 2D keypoints; (2) a combined loss function integrating L1 regression, focal loss, and average Hausdorff distance, with ablations validating each component; (3) open-source implementation within an existing detection framework; and (4) demonstration of improved small-object localization and reduced false positives on an aerial elephant dataset compared to box-supervised baselines. Ablation studies show that the Hausdorff distance term provides the largest accuracy gain by effectively reducing false positives, while focal loss improves stability in densely clustered regions. The proposed method achieves 0.786 mAP and accurately localizes animal centers across diverse environmental conditions, offering a practical solution for conservation practitioners working under tight logistical constraints.

Fog-dependent ecosystems in the Atacama Desert host highly specialized vegetation, yet monitoring their functional traits remains challenging due to remoteness and limited spectral detectability. The bromeliad Tillandsia landbeckii exhibits extremely low reflectance in the VIS/NIR range, rendering conventional multispectral approaches ineffective.

This study evaluates the potential of a novel UAV-based VNIR/SWIR multi-camera system (camSWIR) for estimating canopy water content (CWC) in Tillandsia landbeckii. A UAV survey conducted in northern Chile acquired high-resolution (≈3 cm GSD) SWIR imagery across four operational bands (1100–1650 nm). Field-based destructive sampling (n = 20) provided reference CWC measurements, and a statistically rigorous workflow was applied to mitigate overfitting in a high-dimensional predictor space.

Results show that the spectral slope between 1200 and 1510 nm is the most informative predictor of CWC, with cross-validated performance indicating moderate predictive skill (LOOCV R² ≈ 0.52), but reduced stability under nested validation. The repeated selection of predictors within this wavelength region confirms a physically meaningful relationship with liquid water absorption.

Despite limitations due to a small sample size and species-specific optical properties, particularly the dense trichome layer that affects light interactions, the study demonstrates the feasibility of SWIR-based, non-destructive CWC estimation in hyper-arid ecosystems. These findings provide a proof of concept for future upscaling, highlighting the need for larger calibration datasets and improved modelling to enable reliable spatial mapping of plant water status.

Aerial imagery provides a powerful avenue for monitoring wildlife populations, yet automated detection remains challenging. Animals typically occupy only a tiny fraction of large-scale aerial imagery, may be partially obscured, and appear against highly diverse Arctic and sub-Arctic backgrounds. Suppervised deep-learning detectors also depend on large, fully annotated datasets, making broad ecological surveys labor-intensive and slow to scale.

This study explores an alternative perspective: viewing wildlife as rare events within mostly background imagery. Instead of training on annotated animal samples, an anomaly-detection framework learns the visual patterns of normal landscapes and identifies deviations from these patterns as potential animal locations. To guide the model without costly labels, simple animal-like shapes are inserted into background patches during training, encouraging the network to recognise features associated with real targets while avoiding the need for detailed masks or bounding boxes.

The approach generates two outputs: patch-level predictions distinguishing empty from potentially occupied areas, and pixel-level anomaly maps highlighting likely target locations. When evaluated on a highly varied Arctic dataset, the method remains reliable despite major shifts in terrain, surface texture, animal distributions and postures, and pronounced class imbalance that often degrade supervised models. Unlike distribution-based anomaly approaches that rely on stable normal-feature statistics and frequently misinterpret natural texture variability as anomalies, this method handles heterogeneous environments more effectively.

Overall, the study shows that anomaly-oriented frameworks, typically used in industrial and medical settings, have strong potential to ease annotation demands and support scalable, automated wildlife detection in complex remote-sensing environments.

This study examines PIX4D’s implementation of Gaussian Splatting for reconstructing complex geometries, with a focus on underwater photogrammetry for coral reef mapping. Unlike standard Gaussian Splatting pipelines that emphasize photorealistic rendering, our approach prioritizes high-precision geometric reconstruction, especially for thin structures and heavily occluded regions. We compare the method against conventional multi-view stereo techniques using both real underwater imagery collected in Moorea (French Polynesia) and synthetic datasets generated with the POSER underwater simulation framework.

Stand-level merchantable tree stem volume estimation in temperate forests is critical for data-driven forest management decision-making. Mobile laser scanning (MLS) has greatly improved data-collection efficiency for forest biometrics; however, automated analysis of massive, structurally complex MLS point clouds remains limited. This study presents an automated framework to estimate stand-level merchantable stem volume from backpack mobile Light Detection and Ranging (LiDAR) data. The framework comprises three stages: (1) point cloud reconstruction using the Integrated-Scan Simultaneous Trajectory Enhancement and Mapping (IS²-TEAM) method; (2) individual tree segmentation via a multistage geometric pipeline; and (3) merchantable stem volume estimation based on skeletonization-derived stem modeling. The proposed approach is evaluated on a forest-scale dataset collected in temperate natural forests in the United States. Results demonstrate operational feasibility at scale, with practical processing times and robust geometric consistency. Validation against destructively measured reference volumes shows that the proposed approach outperforms baseline quantitative structure modeling (QSM) methods, achieving a coefficient of determination (R²) of 0.97, a bias of −0.06 m³, and a root mean square error (RMSE) of 0.21 m³. The proposed framework enables reliable, automated estimation of merchantable stem volume from MLS data and supports deployment from individual-tree to forest scales with minimal manual intervention.

TRACE is a training-free framework for instance-level open-vocabulary inventory generation in 3D forensic evidence reconstruction.

Starting from multiview RGB imagery, prompt-based 2D object masks are extracted using SAM3 and associated across views via geometry-aware and appearance-aware multiview instancing. Based on COLMAP geometry and DINOv2/v3 descriptors, the proposed framework establishes globally consistent same-class object identities across the scene. The resulting global instances are then encoded with SigLIP2 to obtain language-aligned instance descriptors and subsequently lifted into a 3D Gaussian Splat representation by assigning instance-level semantics to geometrically supported Gaussian subsets. This yields an enriched 3D scene representation that jointly preserves spatial structure, object-level identity, and language-accessible semantics, thereby enabling instance-aware open-vocabulary querying in 3D.

The purpose of this study is to automatically classify water and land areas with LiDAR point clouds. After determining the average water level, the water and land surfaces were classified. Previous studies have focused on supervised classification based on land sampling or deep learning techniques using photographs. However, these classification techniques are expensive and require long calculation times. In this study, a method is proposed for the automatic classification of water and land areas without land surveys using the coordinate and reflection values of LiDAR point clouds. The bounding box method was used to detect water surface levels. The correlations between the min-box level, mean box height, and mean box reflection values of the LiDAR point data were used to determine the water surface level. The results show that the method is suitable for the fast classification of water surfaces from LiDAR point clouds. Thus, shoreline changes in large areas can be detected automatically without the need for land surveying. The proposed bounding box classification method can be applied independently of LiDAR point cloud density. The extended version of this method can also be used to detect vehicles and objects on a water surface.

Underground environments such as sewer networks are critical infrastructure whose condition directly affects public health, environmental protection, and maintenance costs. Conventional inspection workflows largely rely on monocular CCTV systems and manual video review, providing limited 3D understanding and often missing subtle or spatially complex defects. At the same time, sewer environments are characterised by challenging imaging conditions, including low illumination, specular surfaces, water films and occlusions, which further complicate reliable assessment.

In this extended abstract, we present a real-time inspection concept that combines (i) stereo camera-based SLAM for geometric mapping and pose estimation, (ii) Vision Transformer (ViT) based anomaly detection trained on the public SewerML dataset, and (iii) lightweight Gaussian Splatting modules that create local high-resolution 3D reconstructions only in the vicinity of detected defects. The system is targeted at embedded hardware, specifically an NVIDIA Jetson Nano, and is designed for deployment and evaluation in real sewer environments. The overall goal is to provide inspectors and asset managers with spatially anchored 3D visualisations of anomalies that can be integrated into digital-twin workflows for decision support and long-term monitoring.

This paper proposes a robust cross-modal matching framework for tunnel inspection, specifically designed to address the unique challenges posed by low-texture environments often encountered in tunnel linings. Traditional image-based matching techniques struggle in these environments due to the lack of distinctive surface features and limited texture variation. To overcome these challenges, the proposed method leverages the global prior knowledge of tunnel geometry. By jointly projecting LiDAR point clouds and multi-camera images onto a shared parameterized cylindrical surface, the method constructs a unified geometric space that facilitates accurate 3D–2D correspondences. This dual-projection strategy significantly improves the alignment of structural features such as segment joints, line grooves, and equipment brackets, which are critical for defect detection in tunnel inspection. The enhanced matching ability allows for more reliable multi-sensor data fusion, thereby supporting the automated analysis of tunnel defects. This framework lays a solid foundation for intelligent tunnel inspection systems, offering a powerful solution for real-time monitoring and analysis of tunnel infrastructure.

Flood susceptibility mapping (FSM) is a crucial component of flood risk assessment; however, traditional statistical and machine learning methods for FSM are limited in their predictive capabilities. FSM approaches typically use static inputs, relying solely on geospatial factors, and fail to consider the spatiotemporal aspects (antecedent conditions) that trigger flood events. This study addresses this gap by developing a hybrid model that combines static geospatial features with dynamic temporal meteorological data, which is often excluded in FSM. The proposed hybrid model consists of two branches: (1) a 2D Convolutional Neural Network (CNN) to extract the features from geospatial inputs (i.e., slope and surficial geology) and (2) a Convolutional Long Short-Term Memory (ConvLSTM2D) network to learn the temporal antecedent conditions from Daymet precipitation, temperature and snow-water equivalent. This model was trained and tested in the Saint John River basin, New Brunswick, Canada — a region that has experienced significant historical flooding. Three hyperparameters were investigated: temporal sequence length (1–4-month timesteps), resampling ratio (0.1-0.7), and positive class weight (1.5 or 2.0). The optimal model was achieved with a 3-month timestep, a 0.2 resampling ratio, and a 1.5 positive class weight, resulting in an F1 score of 0.89. The model performance was highest when using a 3-month timestep, which captured the full snowmelt-to-rain spring cycle, outperforming models that used timesteps of 1, 2, or 4 months. The proposed 2D CNN-ConvLSTM2D architecture is effective in simultaneously learning the static geospatial features and temporal meteorological sequences, highlighting the importance of seasonal antecedent conditions in FSM.

Optical satellite sensors are widely used for rapid flood mapping due to their global coverage and free availability. Thresholding spectral indices, such as the Normalized Difference Water Index (NDWI), can detect water pixels rapidly and with good precision. However, small shifts in threshold values can lead to large differences in flood area and data-driven approaches for threshold selection remain a challenge. At the same time, new foundation segmentation models, such as the Segment Anything Model (SAM), can extract object boundaries from images without task-specific training, though it lacks flood-specific contextual awareness. To address these limitations, we propose a risk-guided segmentation framework that combines risk-weighted optimization of NDWI thresholding, and further refinement of the NDWI mask using SAM. The goal is to improve flood delineation by incorporating information on where a flood is more likely to occur (flood hazard maps) and how flood boundaries appear visually (SAM). We evaluate the method on the 2018 spring flood along the Saint John (Wolastoq) River in New Brunswick, Canada, across five study regions for both Sentinel-2 and Landsat-8 scenes using imagery captured on May 2, 2018 (peak flood for the study regions). We show that a higher risk score corresponds to a higher segmentation accuracy, demonstrating that flood hazard maps can help guide NDWI threshold selection. Moreover, refinement with SAM improves segmentation quality compared to the baseline NDWI masks, demonstrating that the use of risk-guided spectral thresholding with foundation models can improve flood delineation in optical satellite imagery.

Floods remain among the most damaging natural hazards worldwide. In Canada’s capital region—Ottawa and its surrounding areas—flood prediction is crucial, most especially in flood-prone zones, to mitigate recurring events such as the 2017 and 2019 Ottawa floods, which caused extensive damage to homes and infrastructure. This study integrates 18 flood conditioning factors with remote sensing indices and ensemble machine learning to improve flood susceptibility mapping in the Ottawa River watershed. A complementary HEC-RAS 2D hydraulic model simulated flow depth and velocity under a 100-year flood scenario. The ensemble model achieved strong predictive performance (Kappa, F1-score, and AUC > 0.979) and demonstrated high transferability across sub-regions (Kappa > 0.85; F1-score > 0.92; AUC > 0.99). HEC-RAS results indicated spatial variability in flood depth (up to 15 m) and velocity (up to 15 m/s). SHAP analysis identified Elevation, HAND, MNDWI, NDWI, and Aspect as the dominant flood-driving factors. The integrated framework enhances flood susceptibility assessment and supports Natural Resources Canada’s efforts to strengthen flood risk management and resilience in the Ottawa River watershed and similar regions.

Machine learning for flood susceptibility mapping (FSM) has traditionally relied on narrowly scoped events and temporally constrained datasets, limiting the generalizability and long-term utility of predictive models. We present a multi-event, multi-temporal modelling framework that leverages discrete flood occurrences from 2005 to 2023 to train a unified model capable of inference across an extended temporal horizon. Each flood event was treated as a spatio-temporal marker, enabling the model to learn evolving driver–event relationships and underlying temporal trends. Dynamic inputs (e.g., climate data, land use/land cover) are integrated with static geophysical features (e.g., digital terrain model and derivatives) to capture both transient and persistent influences on flood susceptibility. An XGBoost model was trained, tested, and validated using a 70/15/15 split, achieving an overall accuracy of 0.945, with true positive and true negative rates of 0.95 and 0.94, respectively. Precision scores for wet (flood-prone) and dry (non-flood-prone) classes are 0.94 and 0.95. Generated yearly national FSM maps from 2000 to 2023 were evaluated against published flood event datasets. Validation using national flood records, climate variability bulletins, and spatio-temporal analyses of year-to-year raster correlations confirms that years with elevated predicted susceptibility correspond to observed flood events. In addition, a weighted wetness score identified the years with both widespread and extreme flood-prone conditions, highlighting the model’s ability to capture multi-scale temporal dynamics. These results demonstrate that multi-event, multi-temporal modelling enhances the temporal reach and robustness of geospatial flood prediction, providing a foundation for long-term monitoring, trend analysis, and policy-relevant scenario planning.

This study presents a high-resolution geomorphological analysis of river terraces along the urban corridor of the La Silla River (Monterrey Metropolitan Area, Mexico) using UAV-based LiDAR and photogrammetry, with a DJI Matrice 350 RTK equipped with a Zenmuse L2 sensor, generating dense point clouds, DEMs, and orthomosaics. These products allowed for the precise identification of three terrace levels (T1-T3), their geomorphometric attributes, and their lithological composition. The results reveal contrasting degrees of anthropogenic modification: while terrace 1 retains its natural morphology, terraces 2 and 3 show substantial alterations due to residential expansion, public infrastructure, and road construction, which alter the original geomorphological surfaces. Temporal satellite images also show the sensitivity of terrace geomorphology to extreme hydrometeorological phenomena, with cyclones such as Hanna (2020) and Alberto (2024) causing vegetation loss, surface restructuring, and local modification of terraces. Overall, UAV-LiDAR proved to be very effective for mapping terraces in restricted urban environments, providing essential details for monitoring, risk assessment, and sustainable management of urban rivers.

Terrestrial laser scanning technology can efficiently acquire high-precision three-dimensional spatial information in complex forest environments, making it an important technical means for detailed analysis of forest structure and resource monitoring. However, traditional terrestrial laser scanners planning methods are prone to coverage gaps and data redundancy due to factors such as tree obstructions, terrain undulations, and canopy overlap, making it difficult to simultaneously balance observation completeness and scanner station deployment cost. To address this, this paper proposes an intelligent survey station planning for terrestrial laser scanners in complex forest environments. The method first uses airborne LiDAR data to build a prior forest model, which is then used to quantitatively evaluate forest visibility features by calculating the cumulative visible central angle through visibility analysis. Finally, an integer linear programming model is further introduced to achieve global optimization of the station set based on an initial feasible coverage solution obtained using a greedy algorithm. To test the performance of the proposed method, this paper applies the proposed method to the forest plot located in Lushan city, Jiangxi province, China. Experimental results indicate that the proposed method achieves an overall coverage rate of 94.55% with only seven stations, reducing the number of stations by approximately 30% and 22% compared with the greedy algorithm and genetic algorithm, respectively. The results demonstrate the effectiveness and superiority of this method for station planning in complex forest areas and provide efficient and precise technical support for forest structure monitoring and spatial information acquisition.

Reflectance is a widely used feature for all types laser scanning data. Thus, the accuracy and improvement of the reflectance parameter is a persistent topic of research. For short laser pulses with medium-sized footprints, previous work has investigated the effects of inclined targets on the recorded waveform of full-waveform LiDAR systems. In this work, a new methods to extract incidence angle from only a single waveform can be leveraged to improve reflectance values through recalculation based on the laser-radar equation and correcting for angle of incidence artifacts. The results of the proposed method are evaluated with two datasets based on two different topo-bathymtric laser scanners. For both systems, we calculated the relative biconical reflectance and relative averaged bidirectional reflectance distribution function (rBRDF) and evaluated them on homogeneous roof faces. The two reflectance measures are then compared to the initial reflectance values of the laser scanners used in the study. Both measures showed improvements compared to the standard values. The biconical reflectance shows the best overall mean score for all surveyed roofs with an MAD improvement of 0.80 dB to 62 dB for Sensor I and 0.61 dB to 0.56 dB for Sensor II, in addition the rBRDF also displays an improvement with varying results depending on the deployed system. These results highlight the advantages of the proposed reflectance measures and the potential improvement of the widely used LiDAR attribute.

Accurate classification of topo-bathymetric LiDAR data remains challenging due to the heterogeneous nature of land-water transitional environments, where terrestrial, water surface, and submerged features must be distinguished simultaneously. This study presents a multi-branch deep learning architecture for classifying bathymetric LiDAR data into different classes: soil ground, trees and vegetation, water surface, seabed, aquatic plants and other underwater objects (dead wood, coral reef). The proposed framework employs three parallel feature extraction branches, while the first branch captures spatial structure by focusing on three-dimensional geometric coordinates (XYZ), the other two branches use two independent 1D U-Net architectures to extract signal-based features from RGB spectral reflectance and waveform-derived attributes (intensity, return number, number of returns). The discrete LiDAR attributes, though represented as point-wise numerical values, preserve signal characteristics derived from full-waveform analysis. The encoder-decoder of 1D U-Net architecture with skip connections effectively captures sequential patterns and multi-return patterns in different classes especially in vegetation canopies. The three feature streams are fused through fully-connected layers before final classification. Evaluation using different metrics demonstrates the capability of the framework to simultaneously classify diverse coastal zone and inland waters contexts spanning terrestrial and submerged domains within a unified processing pipeline, eliminating the need for separate terrestrial and bathymetric classification workflows.

Permanent laser scanning (PLS) is an effective tool for near-continuous monitoring of topographical changes in beach-dune systems. While PLS systems were traditionally costly, the emergence of affordable LiDAR sensors enables larger-scale setups with multiple scanners or sites. However, the different characteristics compared to high-end devices, create challenges for one-on-one replacement. To assess how low-cost sensors can replace high-end sensors, we compare the performance of a setup with several low-cost Livox AVIA sensors to a single high-end RIEGL VZ-2000i sensor in its ability to capture an embryonic dune field with large variation in topography. This is evaluated using HELIOS++ virtual laser scanning (VLS). To also assess the representativeness of the simulations, we further compare the VLS to real-world measurements with the Livox AVIA. Based on a VLS setup with six AVIAs mounted on tripods at 2 m above ground, a coverage of 52% can be obtained, which is similar to the coverage of a single RIEGL VZ-2000i on a tower 8 m high. The real-world experiments confirm the VLS results with a slightly lower point cloud coverage of 42%. Furthermore, the effective range of the Livox AVIA in a beach-dune system lies around 100-150 m. At larger ranges, only pulses at high incidence angles (angle between surface and incoming beam, >20°) are registered at the scanner. The variations in coverage between the VLS and real-world scans highlight the need for careful consideration of the occlusion potential of different representations of the topography, beam divergence shapes, and the moisture conditions.

The aim of this abstract is to assess the accuracy of the trajectory of the handheld laser scanner CHCNAV RS10. The trajectory data of this PLS device is compared with a simultaneously measured total station measurement.

Green-wavelength LiDAR systems enable high-resolution 3D sensing in underwater environments, but the geometric evaluation of measurements across the waterline remains difficult. A main challenge is that traceable reference instruments usually operate only in air, while refraction at the air-water interface systematically affects both the reconstructed 3D point cloud and the geometry of partially submerged objects. To address this problem, this study presents a controlled experimental framework for evaluating waterline-induced effects in an Underwater LiDAR (ULi) system, using the Z+F IMAGER 5016A as an in-air reference. A rigid reference frame (RRF) spanning the waterline was deployed in a swimming pool. The RRF was first scanned by the IMAGER in air to establish the reference geometry and was then measured by the ULi system under waterline conditions. The analysis considered the above-water, cross-waterline, and underwater parts of the RRF. The evaluation was based not only on overall geometric deviations but also on rigid-body-invariant internal quantities, especially pairwise distances that are independent of the pose of the RRF. In addition, the sensitivity of the reconstructed geometry to the refractive-index setting used in processing was assessed by perturbing the refractive index and quantifying the resulting changes. The proposed workflow provides a practical and traceable basis for isolating and evaluating waterline-related refraction effects in controlled ULi experiments.

The rapid evolution of three-dimensional (3-D) geospatial science has redefined the standards of national mapping and cadastral agencies (NMCAs). Traditionally bodies of authoritative 2-D topographic products, these organisations now face the challenge of producing, maintaining, and disseminating national-scale 3-D geospatial datasets that support applications ranging from climate adaptation and urban planning to disaster response and digital twins. This paper presents a comparative study of five NMCAs, comprising IGN (France), BKG (Germany), Kadaster (The Netherlands), GSI (Japan) and USGS (United States of America). By examining agency structure, economic models, and 3-D data collection programmes, this paper identifies converging trends in AI integration, national surveys, along with divergences in funding and implementation. The analysis highlights insights and potential lessons for organisations at early stages of national 3-D dataset implementation.

Frame-wise semantic segmentation of indoor lidar scans is a fundamental step toward higher-level 3D scene understanding and mapping applications.

However, acquiring frame-wise ground truth for training deep learning models is costly and time-consuming. This challenge is largely addressed, for imagery, by Visual Foundation Models (VFMs) which segment image frames. The same VFMs may be used to train a lidar scan frame segmentation model via a 2D-to-3D distillation pipeline. The success of such distillation has been shown for autonomous driving scenes, but not yet for indoor scenes. Here, we study the feasibility of repeating this success for indoor scenes, in a frame-wise distillation manner by coupling each lidar scan with a VFM-processed camera image. The evaluation is done using indoor SLAM datasets, where pseudo-labels are used for downstream evaluation. Also, a small manually annotated lidar dataset is provided for validation, as there are no other lidar frame-wise indoor datasets with semantics. Results show that the distilled model

achieves up to 56% mIoU under pseudo-label evaluation and around 36% mIoU with real-label, demonstrating the feasibility of

cross-modal distillation for indoor lidar semantic segmentation without manual annotations.

Recent advances in image-based satellite 3D reconstruction have progressed along two complementary directions. On one hand, multi-date approaches using NeRF or Gaussian-splatting jointly model appearance and geometry across many acquisitions, achieving accurate reconstructions on opportunistic imagery with numerous observations. On the other hand, classical stereoscopic reconstruc- tion pipelines deliver robust and scalable results for simultaneous or quasi-simultaneous image pairs. However, when the two images are captured months apart, strong seasonal, illumination, and shadow changes violate standard stereoscopic assumptions, causing existing pipelines to fail. This work presents the first Diachronic Stereo Matching method for satellite imagery, enabling reliable 3D reconstruction from temporally distant pairs. Two advances make this possible: (1) fine-tuning a state-of-the-art deep stereo network that leverages monocular depth priors, and (2) exposing it to a dataset specifically curated to include a diverse set of diachronic image pairs. In particular, we start from a pretrained MonSter model, originally trained on a mix of synthetic and real datasets such as SceneFlow and KITTI, and fine-tune it on a set of stereo pairs derived from the DFC2019 remote sensing challenge. This dataset contains both synchronic and diachronic pairs under diverse seasonal and illumination conditions. Experiments on multi-date WorldView-3 imagery demonstrate that our approach consistently surpasses classical pipelines and unadapted deep stereo models on both synchronic and diachronic settings. Fine-tuning on temporally diverse images, together with monocular priors, proves essential for enabling 3D reconstruction from previously incompatible acquisition dates.

Unmanned Aerial Vehicles (UAV)-based photogrammetry provides an efficient solution for shallow water bathymetry, yet its accuracy is fundamentally constrained by light refraction at the air-water interface, which violates the central geometric assumptions of traditional photogrammetry.

Existing approaches, ranging from empirical corrections and iterative post-processing to black-box deep learning, often compromise geometric fidelity, physical interpretability, or generalization.

We address this challenge through Refraction-Aware Gaussian Splatting (RA-GS), which embeds a physically rigorous two-media refraction model directly into the Gaussian Splatting (GS) reconstruction pipeline.

Rather than relying on computationally expensive per-pixel ray tracing, we formulate an analytical parameter transformation that maps the true underwater position, scale, and opacity of each Gaussian to their apparent states observed through a planar refractive interface.

Through this fully differentiable transformation, true underwater 3D geometry and photorealistic appearance are jointly optimized by directly minimizing the photometric error within the standard GS framework.

This approach relies solely on RGB imagery, eliminating the need for external depth priors or deep learning networks.

Using a physically based, ray-traced synthetic riverbed dataset, we isolate and explicitly correct refractive distortions.

Our method achieves a geometric F1-score of 94\% (10 cm threshold at 10 m depth) and produces high-quality novel view synthesis with a PSNR of 25.9 dB and SSIM of 0.93.

Field experiments on real UAV data corroborate the practical utility for high-precision bathymetric mapping under calm-surface conditions.

By resolving the fundamental refractive difficulty, the proposed framework provides a physically grounded, computationally efficient, and practically useful solution for next-generation photogrammetric bathymetry.

Understanding vegetation dynamics in three-dimensional, high-temporal resolution is essential for advancing ecological research and sustainable forest management. This study introduces a novel methodology for tracking branch kinematics in trees using hyper-temporal terrestrial laser scanning (TLS) data. Focusing on a solitary pedunculate oak (Quercus robur) over a one-year period, we employed a geometric feature detection algorithm combined with quantitative structure modeling (QSM) to identify and track distinctive point cloud sections on first- and second-order branches. By leveraging an iterative closest point (ICP) alignment process, branch kinematics were analyzed across multiple epochs, yielding detailed three-dimensional movement trajectories. The results demonstrate that branch movements exhibit screw-shaped patterns. Temporal resolution analysis revealed that a one-week recording interval is sufficient for our study subject to reliably capture kinematic dynamics, whereas longer intervals (e.g., three weeks) result in significant deviations from actual trajectories. The proposed method proved robust against partial occlusions from leaf growth but struggled under extensive occlusions. This research highlights the potential of hyper-temporal TLS for non-contact, high-resolution monitoring of tree canopy dynamics and provides a foundational approach for future studies aimed at modeling vegetation movement and structural changes over time.

3D instance segmentation for laser scanning (LiDAR) point clouds remains a challenge in many remote sensing-related domains. Successful solutions typically rely on supervised deep learning and manual annotations, and consequently focus on objects that can be well delineated through visual inspection and manual labeling of point clouds. However, for tasks with more complex and cluttered scenes, like in-field plant phenotyping in agriculture, such approaches are often infeasible. In this study, we tackle such a task - in-field wheat head instance segmentation using terrestrial laser scanning (TLS) point clouds. To address the problem and circumvent the need for manual annotations, we propose a novel two-stage pipeline. To obtain the initial 3D instance proposals, the first stage uses 3D-to-2D multi-view projections, the Grounded SAM pipeline for zero-shot 2D object-centric segmentation, and multi-view label fusion. The second stage uses these initial proposals as noisy pseudo-labels to train a supervised 3D panoptic-style segmentation neural network. Our results demonstrate the feasibility of the proposed approach and show significant performance improvements (up to +50\% in F1-score) relative to Wheat3DGS, a recent alternative solution for in-field wheat head instance segmentation without manual 3D annotations based on multi-view RGB images and Gaussian Splatting, showcasing TLS as a competitive sensing alternative. Moreover, the results show that both stages of the proposed pipeline can deliver usable 3D instance segmentation without manual annotations, indicating promising, low-effort transferability to other comparable TLS-based point cloud segmentation tasks.

Prompted by the rapid advancements in software and hardware, 3D building data for numerous different applications is nowadays often captured via mobile or kinematic laser scanning. However, in contrast to other laser scanning methods, there exist only a few approaches tailored for the planning of a kinematic laser scan survey, and none of them provides an optimality guarantee. Therefore, we propose a novel approach based on Mixed Integer Linear Programming (MILP) to find the optimal trajectory for such a survey. To obtain a high-quality point cloud, we account for scanner-related constraints that influence the quality of the resulting point cloud. Moreover, we enable the introduction of tie points to mitigate the effects of uncertainties in the position estimation that are propagated in the acquired data. In our problem formulation, we aim to find the best tour in a properly weighted graph. For this, we propose two different weight settings to either enable a purely length-based optimization or to increase the redundancy in the measurements by incorporating a Visibility Ratio Factor (VRF) into the objective function.

To prove the applicability of our approach for offline panning, we apply our formulation to three different scenarios. In this context, the VRF-based weighting enables a significant speed-up of the solving process while resulting in only slightly prolonged routes. This approach paves the way for applying exact algorithms with an optimality guarantee in the planning process for efficient kinematic laser scanning surveys.

This contribution introduces a novel method for non-contact, marker-free modal analysis of wind turbine blades using terrestrial laser scanning (TLS). As part of a research initiative, TLS's potential for assessing modal properties like natural frequencies and mode shapes—key for extending blade service life—is explored. Traditionally, this analysis relies on numerous accelerometers, incurring high costs and effort. TLS is evaluated as a viable alternative.

In laboratory tests, TLS and photogrammetry were used on a 4-meter test object in vibration. Photogrammetric data, serving as a reference, used 3D coordinates from retroreflective markers for frequency analysis via Fast Fourier Transform (FFT). TLS data were similarly segmented, with frequencies derived using FFT, and both methods showed consistent results, validating TLS's feasibility.

Building on lab results, the method was applied to an 88-meter rotor blade in a field experiment. The laser scanner collected profile data along the blade's longitudinal axis, converted to the object coordinate system. By segmenting the blade, eigenfrequencies were determined. The calculation process was validated with simulations, achieving precise results even with manual blade excitation and amplitudes up to 20 cm.

TLS measurements reveal valuable insights into eigenfrequencies and modal shapes along the blade. This approach offers a cost-effective, efficient alternative to traditional sensor-based analysis, proving its practicality for the wind energy industry.

Hybrid airborne imaging systems combining oblique cameras and LiDAR sensors offer significant advantages for applications requiring both geometric precision and rich texture information, including infrastructure monitoring, facility surveying, and detailed urban modeling. Despite capturing temporally consistent multi-modal data, achieving pixel-level registration between imagery and LiDAR remains fundamentally challenging due to insufficient calibration infrastructure and the technical complexity of deeply integrating heterogeneous sensors.

A critical bottleneck is that standard photogrammetric workflows exhibit non-linear cumulative drift, particularly across extended flight strips. This spatially varying deformation causes systematic misalignments when photogrammetric reconstructions are overlaid with LiDAR geometry. Conventional approaches applying global rigid transformations fail to address this issue because photogrammetric drift is inherently non-uniform—a single global registration cannot correct localized geometric deviations throughout the scene.

This work introduces a novel view-dependent registration framework that synergizes LiDAR's global geometric fidelity with photogrammetry's local density. Rather than attempting to warp entire models through global transformations, we decompose the registration problem by treating the geometry within each camera frustum as an independent rigid body. Building upon initial georeferencing, we perform fine-grained local SE(3) rigid registration to anchor each Multi-View Stereo (MVS) depth map directly to sparse LiDAR geometry within its corresponding viewing frustum.

This localized approach enables pixel-accurate alignment within individual frames while effectively compensating for accumulated photogrammetric drift and interpolation errors. By addressing registration at the frustum level rather than globally, our method achieves practical pixel-level fusion of hybrid airborne datasets, unlocking the full potential of integrated camera-LiDAR systems for high-precision geospatial applications.

High-resolution Digital Elevation Models (DEMs) are essential for urban flood modelling, where small elevation differences govern surface drainage and inundation extent. DEMs frequently contain hydrological inconsistencies: elevated infrastructure such as bridges, tunnels and culverts may appear as artificial barriers disrupting flow continuity, while linear structures such as retaining walls may be underrepresented depending on spatial resolution or point density. These inconsistencies propagate errors through downstream hydraulic simulations.

This paper presents an automated, open-source Python pipeline for generating hydrologically conditioned DEMs by integrating classified airborne LiDAR data with OpenStreetMap (OSM) infrastructure features. The workflow is tested on a 16 km2 area of central Copenhagen using a 2023 national LiDAR acquisition at 13.5 pts/m2. A 0.5 m resolution DSM is generated from LiDAR ground and building classes via Inverse Distance Weighting interpolation, with Nearest Neighbour gap-filling for hydraulic model continuity. Hydrological conditioning is performed through four sequential operations: bridge burning, tunnel enforcing, culvert enforcing, and barrier rasterization. Barrier top-of-wall elevations are estimated directly from the LiDAR point cloud.

Vertical accuracy is assessed by pixel-wise comparison against the Danish national terrain model DHM/Terraen (NMAD = 0.066 m, LE90 = 0.265 m) and by independent checkpoint validation against the HojdefikspunktDanmark geodetic network. The inclusion of shallow tunnel underpasses proved a significant addition: tunnel features alone contributed approximately half of the total depression volume reduction. The conditioned DSM is designed as input for an urban flood simulation chain; full hydraulic validation will be performed by the Danish Meteorological Institute within the CLEAR-EO project.

Kelp forests are vital marine habitats with significant ecological, cultural, and economic importance. These ecosystems, found along coastlines, are susceptible to regional and global stressors (such as coastal development and climate change). This paper presents Satellite-based Kelp Mapping (SKeMa), a novel framework for automatically mapping canopy-forming kelp forests using Sentinel-2 satellite imagery along the British Columbia coast, specifically to support First Nations marine planning for these species. SKeMa employs a deep learning semantic segmentation model, offering an efficient alternative to traditional, labor-intensive, and time-consuming kelp mapping methods. A cross-validation study with independent test sets yields a mean Intersection over Union (IoU) of 0.5326, demonstrating the model’s capability to detect kelp canopies across diverse coastal regions, particularly for larger kelp beds.

The research extended DualSeq, an advanced

machine-learning model for predicting sea surface temperature

(SST), crucial for understanding oceanic ecosystems and climate

patterns. Traditional SST prediction methods typically employ

time-series regressions focusing on nonlinear temporal patterns,

but often overlook vital spatial correlations in SST dynamics,

limiting their accuracy. DualSeq addresses this by integrating

spatial and temporal uncertainty quantification, with a particular

focus on the Arabian Sea. It utilises LSTM and GRU networks to

effectively harness the SEVIRI-IO-SST dataset, which contains

five years of remote-sensing data. A distinctive aspect of DualSeq

is its incorporation of a weighted normalized linear equation,

which significantly improves the accuracy of SST predictions and

enhances the dependability of spatial and temporal uncertainty

assessments. The model stands out in its ability to forecast up to

one month in advance, significantly outperforming others. For 1-

month forecasts, DualSeq shows a remarkable R² value of 0.983,

surpassing the LSTM-attention model by 7.4% and reducing

RMSE and MAE by about 65.4% and 82.4%, respectively.

This performance illustrates DualSeq’s superior capability in

capturing both short-term and long-term uncertainties in SST

forecasting.

This study presents a Station-Centric Geographically weighted Regression (SCGWR) framework for Chlorophyll-a prediction in the optically complex waters of the Gulf of İzmir using Sentinel-2 imagery. Unlike traditional global multiple regression model, the proposed approach calibrates an individual model for each sampling station while using 16 outer Moore-neighbor pixels (range 2) from surrounding stations as independent validation data in the model optimization, thereby preventing adjacency bias and information leakage in performance assessment. Compared to multiple linear regression (MLR) against 20 independent in-situ measurements, SCGWR method offers a robust, reproducible alternative for local-scale water-quality mapping in coastal environments where bio-optical variability is high.

Coastal areas play an important role economically, socially and environmentally due to their many functions.

However, these regions are at risk of erosion, which is further exacerbated by human-driven climate change.

Tracking and monitoring coastal boundaries enable efficient allocation of conservation and protection efforts.

Due to the vast size and complexity of coastal areas, on-site monitoring to track erosion is inefficient.

Artificial intelligence has shown impressive results in segmenting and extracting these boundaries from remote sensing imagery.

Historical remote sensing data make it possible to track long-term erosion but remain challenging due to the coarse resolution of older data.

Our work proposes studying the impact of super-resolution on coastal boundary segmentation using high-resolution imagery.

ESRGAN and SRCNN have proven highly beneficial in improving the quality of coarse-resolution samples, achieving superior performance compared to bicubic interpolation across scaling factors ranging from ×2 to ×12.

ESRGAN super-resolved samples achieved F1-scores ranging from 97.75% to 89.92% for scaling factors ×2 to ×12, while bicubic interpolation achieved between 97.34% and 65.27%.

These improvements demonstrate that SR enhances boundary delineation and robustness across scales.

Our work also explores the applicability of tracking erosion through historical data.

Results demonstrate a coastal boundary change of 0.23 m per year over seven years, which is on par with expected values.

This study introduces a novel framework that integrates a region-aware Full-Waveform Figure Descriptor (FWFD) with a Convolutional Vision Transformer (CvT) for underwater terrain classification using bathymetric LiDAR data. The FWFD converts sequential waveform returns into a multi-directional image-like representation, enabling the preservation of spatial correlations among neighboring laser footprints. By combining convolutional token embedding and self-attention mechanisms, the CvT effectively learns both local and global waveform features. Experiments on a YellowScan full-waveform LiDAR dataset over coastal Australia demonstrate that the proposed FWFD-CvT model achieves 95.55 % overall accuracy under moderate waveform smoothing and exceeds 98 % accuracy for underwater objects. The framework shows robust performance across complex seafloor morphologies and maintains consistency in mixed land-water environments. This research contributes a transferable paradigm for region-aware waveform interpretation and establishes a foundation for extending full-waveform analysis to terrestrial, multispectral, and topographic LiDAR applications requiring fine-scale surface characterization.

Shallow-water cultural heritage occupies a dynamic land-sea interface where coastal erosion, sediment transport, limited visibility and burial processes hinder conventional archaeological investigation. This paper presents an integrated geoinformatics framework for reconstructing the cultural dynamics of coastal and shallow submerged archaeological landscapes in southeastern Crete, Greece. The methodology combines multispectral remote sensing, satellite-derived and in situ bathymetry, UAV and shallow-water photogrammetry, marine geophysics, GIS-based coastal vulnerability, fuzzy logic multi-criteria risk assessment and digital dissemination through augmented reality. The workflow was applied at five representative case studies, including Stomio, Ierapetra harbour, Koufonisi, Chryse and associated coastal sectors. Optical data from Pleiades-1A, PlanetScope, and Sentinel-2A were used for shoreline mapping, feature enhancement, and satellite-derived bathymetry. Geophysical and bathymetric surveys covered more the 4.5 and 10 hectares respectively. UAV photogrammetry produced high resolution orthomosaics, while the proposed experimental Remote Control (RC) boat extends documentation potential to very shallow submerged environments. Integrated interpretation clarified palaeo-shorelines, submerged harbour structures, fish tanks, architectural continuities and archaeological risk hotspots. The results demonstrate a scalable and transferable framework for documenting, interpreting, monitoring, and communicating endangered shallow-water cultural landscapes.

This paper presents an efficient approach to the accessibility assessment of tram transport based on a simulation within a digital twin environment. We propose a novel framework that integrates several advanced data acquisition and processing steps: mobile mapping of the tram routes, detection of rail tracks and tram stops, and the final assessment of tram accessibility by simulating the MAL deployment in the digital twin. Our experimental evaluation demonstrates that the digital twin provides a practical and reliable tool for assessing tram accessibility.

As-built Building Information Models (BIMs) are crucial for building digitalisation, structural analysis, and life cycle management. Despite recent advances, automated reconstruction of structural elements from point clouds remains a challenging task, particularly in ensuring geometric accuracy and topological consistency within a storey and across consecutive storeys. This paper proposes an automated method for refining topological inconsistency between columns, beams, and slabs, ensuring consistent as-built BIMs. The method places Constructive Solid Geometry (CSG) at the core of the refinement process, driven by fundamental structural principles. The method starts by creating solid rectangular prisms from labelled point clouds. Beams are then aligned both vertically and horizontally within each storey. Columns are vertically aligned across consecutive storeys. Topology relationships between the elements are retrieved and encoded in graphs. These graphs, together with a set of Boolean operations, are used to resolve gaps and trim overlaps between the connected elements. The refined elements are represented in accordance with the IFC standards. The proposed method was validated on two multi-storey case studies representing frame and flat-slab building structures. Both qualitative and quantitative evaluations confirmed the effectiveness of the approach, achieving significant geometric accuracy and topological consistency. In addition, the method exhibits efficient runtime performance, indicating its promise for scalable Scan-to-BIM automation.

Bridge infrastructure plays a critical role in transportation networks, requiring reliable and efficient methods to detect missing bolts to ensure structural integrity and prevent failures. This study proposed a novel methodology integrating point cloud-based Digital Twins (DTs) with Topological Data Analysis (TDA), specifically using Persistent Homology (PH), for robust and accurate missing bolt detection. The framework combines 3D photogrammetric reconstruction to generate point cloud-based DTs, Convolutional Neural Networks (CNNs) for precise bolt localization, and PH to identify and quantify missing bolts. Through parameter evaluations and a real-world bridge case study, the proposed approach demonstrated high detection accuracy, effectively identifying missing bolts with a false positive rate below 10%. These findings confirm the reliability and effectiveness of integrating DTs with TDA as an advanced data-driven approach for automated structural inspection and bridge health monitoring.

Semantic segmentation of indoor point clouds is a fundamental task in 3D scene understanding, supporting applications such as virtual reality, indoor navigation, and building management. Point-based transformer models achieve high accuracy but require substantial computational resources, while superpoint-based methods are more efficient yet often less precise. To address this trade-off, we propose LGFormer, a lightweight framework that integrates Graph Convolutional Networks (GCN) and transformers to jointly capture local and global contextual features. The method constructs a superpoint-based topology graph, where local features are extracted using GCN and global dependencies are modeled through transformer layers. Experiments on the S3DIS and ScanNet++ datasets demonstrate that LGFormer achieves 90.7% and 88.5% segmentation accuracy, respectively, while reducing inference time by more than 99% compared with point-based transformers. By effectively leveraging superpoints and local-global feature fusion, LGFormer dlivers competitive accuracy with significantly lower computational cost, making it suitable for large-scale indoor scene analysis.

Corroding reinforcement leads to cross section loss and reduced structural capacity of concrete bridges. Detecting exposed rebars

(ER) is crucial during bridge inspection to plan countermeasures early and prevent further corrosion. With advancements in deep

learning, several public datasets derived from inspection imagery have been released to identify ER and other concrete damage

automatically. At the same time, Uncrewed Aerial Systems (UAS) have become more capable of navigating even underneath the

bridge deck. This combination holds promise to improve efficiency of bridge inspection methods, but obtained imagery differs

from available datasets, featuring very small damages and complex backgrounds. To address this mismatch, this work reviews

publicly available ER datasets, presents a UAS-based bridge inspection dataset for evaluating ER damage (UBID-ER-val), and

quantifies similarities and differences between them. We train several YOLOv8 models on conventional inspection documentation

images and benchmark the reviewed datasets, scoring F2 = 0.229 at S2DS, F2 = 0.430 at CODEBRIM, F2 = 0.584 at Dacl10k,

compared to F2 = 0.505 at UBID-ER-val. We analyse factors influencing performance and find that tiled inference raises Recall

(+0.166) but drastically reduces Precision (−0.309), while matching training and validation image resolution underperforms across

all datasets (−0.061 to −0.129). The differences in best-performing combinations underscore the underlying domain shift that

complicates practical deployment. As a practical outcome of this work, UBID-ER-val is made publicly available to enable objective

benchmarking of ER detection models and to assess their reliability under field conditions.

Detecting visible electrical utilities is a prerequisite for developing advanced reasoning strategies to reconstruct hidden in-wall networks. This paper investigates the detection of visible power-related utilities using a domain-adaptive deep learning-based vision pipeline based on the YOLOv11-L, object detection model. Four publicly available datasets containing power sockets, power strips, and light switches were curated, relabeled, and merged into a unified training dataset of 3,459 images. The resulting model achieved a mean average precision (mAP) of 0.74 for power sockets and strips and 0.98 for light switches, demonstrating strong detection performance. Real-time evaluation on a low-cost smartphone via the Ultralytics HUB App indicates reliable detection in small-scale real-world environments and detected utilities could be integrated automatically into semantic indoor models using a marker-less referencing approach. The work further highlights broader applications, including Augmented Reality-based visualization to reduce cognitive load for project managers and inspectors or construction workers and electricians, and its potential use as input for existing and future reasoning methods for hidden-utility reconstruction. The prepared dataset, trained model and source code is available at: https://github.com/hcu-cml/indoor-electrical-facility-detection

Accurately representing forest canopies in atmospheric models remains a major challenge due to the complex ways in which trees interact with airflow and modulate surface--atmosphere exchanges. Aerodynamic roughness is a key control variable in modelling frameworks related to air quality, meteorology, and atmospheric transport processes. In this study, we develop a physically based and spatially resolved framework to estimate aerodynamic roughness length from remote sensing observations. Specifically, using AHN (Actueel Hoogtebestand Nederland) airborne laser scanning data over a coniferous forest in Loobos, located within the Veluwe Natura 2000 region in the central Netherlands, we derive geometric roughness parameters and compare them qualitatively against eddy-covariance (EC) tower measurements at the site.

Results show that LiDAR-based roughness captures strong directional and structural variability driven by forest stand height and canopy heterogeneity, patterns that closely align with the anisotropy observed in the EC-derived displacement height and roughness length. Seasonal differences between leaf-on and leaf-off conditions further demonstrate the importance of canopy phenology in shaping aerodynamic behaviour. The spatial patterns resolved by the AHN data underscore the capacity of high-resolution laser scanning to reveal fine-scale canopy--atmosphere interactions that are entirely missed by traditional land-use--based roughness representations.

Additional opportunities remain for integrating complementary remote sensing observations (e.g., multispectral vegetation properties) to enhance the dynamical fidelity of the roughness estimates. The proposed framework provides an observation-driven pathway for parameterizing surface roughness, offering substantial potential for improving land-use representations in wind-flow and chemical transport models such as LOTOS--EUROS.

Forest canopy height (FCH) is a critical biophysical parameter that characterizes forest structure and provides fundamental information for estimating above-ground biomass and carbon stocks. The Global Ecosystem Dynamics Investigation (GEDI) Level 2A (L2A) product offers accurate canopy height observations; however, its point-based nature constrains spatial continuity in FCH mapping. This study integrates the multimodal remote sensing datasets for continuous FCH mapping in Tanzania’s West Usambara (WUSA) forest, recognized globally for its rich biodiversity and ecological significance. Hence, remote sensing data, including Sentinel-1 polarizations (VV and VH), Sentinel-2 spectral bands and vegetation indices, and the SRTM digital elevation model (DEM), were integrated and matched with GEDI canopy height data used as reference for FCH modelling. The optimal feature set was derived by evaluating the performance of several feature selection and extraction methods, including Principal Component Analysis (PCA), Minimum Noise Fraction (MNF), Recursive Feature Elimination (RFE), Sequential Feature Selection (SFS), and the Selected K-Best approach using F-value and mutual information scoring functions. The feature set derived from RFE, comprising ten features from all data sources, demonstrated the highest accuracy and reliability in FCH modelling. Subsequently, four machine learning algorithms, including Random Forest (RF), Gradient Boosting Regressor (GBR), Support Vector Regressor (SVR), and Ordinary Least Squares (OLS), were evaluated for FCH modelling. Accordingly, RF achieved higher R² than GBR, SVR, and OLS, with differences of 0.9%, 8.7%, and 16.4%, respectively. Therefore, the RF model, as the most reliable model, was employed for FCH mapping across the WUSA forest.

Canopy height models (CHMs) are raster maps representing normalized tree canopy height above ground and are often used as co-products for estimating carbon storage, forest degradation, and biodiversity at regional to global scales. While airborne LiDAR delivers the most accurate canopy height (CH) measurements, its high cost and limited temporal coverage motivate the use of spaceborne (multispectral) imagery combined with machine learning.

In this study, we compare two distinct deep-learning approaches for continental-scale CHM estimation from 3 m PlanetScope imagery: (1) a CNN-based regression model (DeepLabv3+), and (2) a monocular depth-estimation model (Depth Anything V2) based on a foundation model. We train/fine-tune both models on a curated dataset of 16,973 pairs of airborne point cloud-derived CHMs and PlanetScope imagery of Australia using a stratified sampling scheme to ensure balanced representation of vegetation structural classes. We then evaluate their generalizability on independent validation sets across Australia, across different heights, and under limited-data scenarios.

Through extensive quantitative and qualitative analysis, we show that the DeepLab-based regression model outperforms Depth Anything across all evaluation metrics, partly because it can incorporate additional spectral channels. DeepLab also learns more effectively from less data. On our dataset, the conventional CNN-based regression model performs better than the fine-tuned foundation model.

To address the critical challenge of regional transferability for ALS-based above-ground biomass (AGB) models, we developed and applied a rigorous leave-one-region-out cross-validation (LORO-CV) framework. This protocol integrates a <1 SE “near-zero” bias filter to ensure models are not just accurate, but statistically free of regional bias.

With this framework, we compared two distinct modeling methods: a data-driven Best-Subset Selection (BSS) method and a Functional Regression (FR) method. The analysis was based on 163 field plots and co-located multispectral Titan ALS data from four regions in the Taiga Plains ecozone, Canada.

The BSS method identified a transferable linear model using height skewness, p95, and an intensity-weighted metric, which achieved 19.3% LORO-CV %RMSE and 2.0% mean absolute bias. Crucially, it passed our <1 SE bias screen in all regions. The FR model, relying only on height, achieved 22.4% LORO-CV %RMSE (4.1% bias) but failed the bias screen in two regions.

Our findings demonstrate that a systematic, bias-controlled data-driven method is effective for producing regionally transferable models. The results highlight the critical importance of ALS intensity metrics for this success, while also showing that the data-driven method currently surpasses the functional approach.

The role of forests in carbon sequestration and regulation is important to understand, given the alarming rate of global warming caused by greenhouse gases. Understanding the structural characteristics of trees can help assess the potential of forests for carbon storage. Light Detection and Ranging (LiDAR) has emerged as a powerful remote sensing tool that is capable of providing detailed three-dimensional information of the forest. The increasing availability of aerial LiDAR data has provided opportunities to estimate the forest biomass over a larger extent. This study utilizes the available LiDAR data from the provincial repository of geospatial data to estimate the diameter at breast height (DBH), which is a key parameter in existing biomass allometric models. LiDAR-derived tree metrics were integrated with the optical images to further differentiate the forest type to assess how it influences the aboveground biomass estimates in a heterogeneous mixed-wood forest. This research contributes to improving our understanding of LiDAR's potential for estimating DBH, an area that has not been explored much. It also demonstrates how existing global biomass allometric equations can be utilized in combination with remote sensing technology to provide a pathway to a larger extent and an efficient method of biomass estimation across diverse ecosystems.

Communication towers equipped with cameras are widely distributed across rural landscapes but remain largely unused for scientific observation. This presentation introduces an AI-driven framework that transforms such existing infrastructure into a high-frequency, real-time agricultural monitoring system, complementing traditional satellite and UAV remote sensing.

The proposed system resolves three fundamental challenges that hinder tower-based sensing: (1) precise georeferencing of highly oblique imagery through a quaternion-based spatial transformation; (2) automated delineation of cultivated parcels via a GIS-guided, iterative segmentation process integrating the Segment Anything Model (SAM); and (3) intelligent recognition of crop types, growth stages, and farming activities using a multimodal large language model that fuses time-series imagery with contextual field data.

Validated through deployments in varied agricultural regions of China, the framework demonstrates stable operation and parcel-level accuracy for continuous monitoring within 1–2 km of each tower. The results indicate a practical pathway toward scalable, cost‑efficient, and autonomous agricultural information acquisition at high spatio‑temporal resolution.

Rapid and reliable assessment of building damage after major earthquakes is essential for effective emergency response and recovery planning. This study formulates post-disaster building damage detection (BDD) as a binary image classification task (damaged vs. undamaged buildings) using multimodal satellite data and a unified ResNet-18 backbone to enable a controlled comparison of fusion strategies. The analysis focuses on the Mw 7.7 Myanmar earthquake of 28 March 2025 and integrates post-event COSMO-SkyMed Second Generation (CSG) dual-polarization (HH, HV) SAR imagery, Maxar optical data, OpenStreetMap (OSM) building footprints, and UNOSAT damage annotations. Three fusion paradigms are evaluated: Early Fusion (EF), Late Fusion (LF), and a novel Middle Fusion (MF) approach. The proposed MF framework introduces a Footprint-Guided Cross-Attention (FGCA) mechanism that uses building geometry as a spatial prior to guide feature-level interaction between SAR and optical representations. Five-fold cross-validation results show that MF consistently outperforms EF and LF, achieving higher precision, F1-score, and robustness across modality configurations. By jointly exploiting SAR structural sensitivity, optical detail, and footprint-based spatial context, the proposed Footprint-Guided Middle Fusion (FGMF) framework enables accurate and scalable building damage mapping from heterogeneous Earth Observation (EO) data.

Accurate and timely building damage detection (BDD) is crucial for disaster emergency response. Although deep learning-based change detection methods have made significant progress in remote sensing, their practical application in disasters still faces two major challenges: (1) Existing high‑accuracy models are typically computationally complex and difficult to deploy for real‑time inference on edge devices.. (2) Model performance heavily relies on large amounts of annotated data, but disaster data are extremely scarce. To address these challenges, this paper proposes a novel lightweight Local‑Global Interaction Network (LGINet) for efficient BDD. The core of LGINet is the proposed Local‑Global Interaction Unit (LGIU), which achieves efficient fusion of detailed and contextual features through a dual‑path architecture and channel‑wise cross‑attention mechanism. Furthermore, a Frequency Difference Enhancement Unit (FDEU) is proposed to generate more accurate damage features, and contrastive learning is employed to reduce the model’s sensitivity to weather conditions and its reliance on annotated data. Experimental results on the xBD and WBD datasets show that LGINet achieves F1-scores of 81.76% and 80.91%, respectively, with an inference speed of 47.83 FPS. It achieves the best balance between accuracy and efficiency, outperforming existing methods.

Recent armed conflicts have increased the need for reliable, spatially explicit damage mapping to support situational awareness, humanitarian assessment, and reconstruction planning. This contribution presents a hybrid change-detection framework for conflict-related urban damage mapping that combines AlphaEarth Foundations embedding change with Sentinel-1 SAR change indices. AlphaEarth provides semantically informed annual embeddings, while Sentinel-1 time series contribute all-weather sensitivity to structural change.

The study compares several embedding-based change metrics and combines the selected AlphaEarth indicator with SAR-derived change measures through simple scalar fusion rules. The proposed framework is designed to preserve the sharp sensitivity of SAR to abrupt structural changes while reducing part of the diffuse background response that often complicates single-source interpretation.

Experiments are conducted over war-affected urban areas in Ukraine, with illustrative examples from Bakhmut and Avdiivka. The results show that AlphaEarth and Sentinel-1 provide complementary information and that their fusion improves the spatial specificity of detected damage patterns. The contribution highlights the potential of combining foundation-model representations with radar time series for operational damage mapping in conflict settings.

Rapid and precise identification of disaster locations is essential for efficient emergency response and management. However, during the immediate post-disaster phase, the lack of timely and reliable information often impedes relief operations. Although satellite imagery and ground-based sensing systems provide valuable data, their effectiveness is constrained by factors such as time delays, high costs, and limited spatial resolution. At the same time, social media platforms such as X (formerly Twitter), Instagram, and Facebook have become valuable channels for real-time, crowd-sourced information. Users function as "human sensors," contributing extensive on-the-ground insights. Much of this content is visual—images that capture the effects of disasters with finer street-level detail and immediacy than textual posts. In this study, we propose a novel, deep learning-based image-matching framework designed to pinpoint the geographic coordinates of disaster events from social media images with street-level accuracy. The core of our approach is to match a query disaster image against a database of georeferenced Google Street View (GSV) imagery. The methodology consists of image pre-processing and feature enhancement; deep feature extraction and matching, and location inference and verification. The preliminary results on an external validation dataset are highly promising, demonstrating a high detection rate of ~90% with confidence scores above 0.9. The model proves resilient to key challenges such as partial occlusion and varied lighting, accurately segmenting multiple objects against complex backgrounds of damaged structures and flooded areas.

The paper presents a novel deep learning framework for automated disaster damage assessment using remote sensing imagery. It addresses the challenge of timely and accurate damage classification in the aftermath of disasters, aiming to improve emergency response and resource allocation. The proposed system leverages both pre- and post-disaster satellite images to assess building damage across four categories: no damage, minor damage, major damage, and destroyed. The central innovation lies in the development of a multi-modal attention mechanism, which integrates features from both pre- and post-event images to enhance damage detection. A lightweight ConvNeXT-Tiny architecture serves as the backbone, ensuring efficient processing while maintaining high performance.

Key contributions of this work include: (1) a cross-attention module that fuses multi-modal data, (2) an optimized preprocessing pipeline designed for large-scale datasets, and (3) novel data augmentation techniques that improve the model’s robustness. Experiments on a large-scale disaster damage dataset show the model achieves an impressive 94.90% classification accuracy, with strong performance in discriminating damage levels and resilience to incomplete or corrupted data. This framework represents a significant step forward in disaster response, offering a scalable solution for real-time damage detection. The research demonstrates the potential of combining remote sensing, multi-temporal imagery, and deep learning to expedite and improve disaster damage assessment, ultimately supporting more efficient emergency management.

Urban areas evolve rapidly, often increasing exposure to natural hazards, especially in seismically active regions such as southern Turkey. This contribution presents an AI-based workflow for multi-temporal analysis of urban expansion in the city of Osmaniye between 2015 and 2025. The methodology integrates Sentinel-2 multispectral imagery with a U-Net convolutional neural network trained on World Settlement Footprint (WSF) masks for binary segmentation of built-up versus non-built-up areas. After training on 2015 and 2019 data, the model was applied to the full temporal series to assess its generalisation capability and to quantify long-term urban growth.

Results show a substantial increase in built-up surfaces over the decade, with a temporary decline linked to the 2023 earthquake and a marked acceleration during the reconstruction phase. Beyond the quantitative trends, the spatial patterns identified by the model highlight how urban expansion has progressively shifted from the central districts toward peripheral zones, revealing both densification processes and outward sprawl. These observations provide valuable indications on how development pressures interact with seismic vulnerability.

The approach demonstrates the potential of AI and open satellite data for large-scale, reproducible monitoring of urban dynamics and for supporting risk-informed urban planning. Because it relies entirely on open-source datasets and tools, the workflow can be easily transferred to other hazard-prone regions, offering a scalable and transparent framework for assessing urban change, post-disaster reconstruction, and long-term exposure.

The AQpictures project, conducted under the ISPRS Scientific Initiatives 2025, addresses the emerging field of image-based estimation of fine particulate matter (PM2.5) concentrations in urban areas. PM2.5 represents a major public health concern, yet existing ground-based monitoring networks offer limited spatial coverage and satellite-derived products struggle to capture surface-level variability. Recent studies have demonstrated that visual attributes in outdoor images, such as sky colour, haze, and visibility, can provide useful indicators of PM2.5 concentrations. Building upon this premise, AQpictures aims to develop an open, reproducible framework for benchmarking and validating image-based air quality estimation methods.

The project first conducts a comprehensive literature review to classify existing approaches into four methodological categories: physics-based, machine learning, deep learning, and hybrid models. Based on this synthesis, a benchmark experiment is implemented for the city of Milan, combining a ten-month dataset of webcam images with co-located PM2.5 ground measurements. The workflow involves image preprocessing, feature extraction, and model evaluation using standard statistical indicators (R², RMSE, MAE). Preliminary tests include physics-based visibility models, feature-based regressors, and convolutional deep learning architectures.

All codes, datasets, and documentation are consolidated in an open-access GitHub repository to ensure transparency, reproducibility, and adaptability of methods across different environmental contexts. Early results confirm the feasibility of PM2.5 estimation from RGB imagery, though further investigations on multi-city datasets are planned to evaluate model transferability and robustness under varying urban and climatic conditions.

This paper presents MAMAI (Maternal Assistance and Monitoring through Artificial Intelligence), a standards-based framework designed to enable interoperable postpartum well-being monitoring using multi-vendor wearable devices. The proposed system addresses a key limitation in digital maternal health: the fragmentation of wearable ecosystems and the lack of integration with clinical infrastructures. MAMAI introduces a federated, edge–cloud architecture that allows wearable data to be processed locally while transmitting only summarized to the cloud.

A core contribution of this work is the integration of two complementary interoperability standards: the OGC SensorThings API for structuring IoT-based sensor observations, and HL7 FHIR for representing well-being indicators in clinically compatible formats. Through this dual-standard approach, heterogeneous wearable data—such as sleep patterns, physical activity, and heart-rate variability—are harmonized into standardized, platform-independent representations.

The framework further introduces a composite well-being score derived from normalized physiological indicators, enabling continuous and interpretable assessment of maternal health. A prototype implementation demonstrates the feasibility of the architecture, supporting end-to-end data ingestion, transformation, interoperability mapping, and visualization. Experimental results show efficient system performance with low end-to-end latency.

Overall, MAMAI provides a scalable and interoperable solution for integrating consumer wearable data into healthcare ecosystems, offering a foundation for next-generation maternal digital health systems and continuous postpartum monitoring.

Achieving the United Nations Sustainable Development Goal (SDG) 11.7.1—“providing universal access to safe, inclusive, accessible, and green public spaces by 2030”—underscores the critical role of urban green space in advancing global sustainability.Although extensive research has examined urban greenery from a traditional planar perspective, green spaces inherently possess vertical structure. Currently, systematic quantitative assessments of urban vertical greenery, residents’ actual exposure to vertical green space, and the associated inequalities remain limited. To address these gaps, this study integrates global population data with vegetation height information to construct an exposure-based analytical framework.We quantify spatial patterns of vertical greenery, residents’ green exposure, and exposure inequality across global urban areas, and further examine the drivers of inequality. Our findings reveal pronounced spatial disparities in urban greenery worldwide. On average, cities in the Global North exhibit approximately three times greater vertical greenery and nearly four times higher green exposure than cities in the Global South. African urban areas possess only one-sixth of the average vertical greenery and one-seventh of the exposure level observed in North America, while displaying roughly twice the inequality in green exposure, indicating much more uneven access to green resources. We also find that cities with higher average vertical greenery tend to experience lower exposure inequality, suggesting that increasing overall greenery can help promote more equitable access. These results provide new theoretical insights and policy-relevant evidence for advancing sustainable and equitable urban green development, supporting global progress toward sustainable development goals.

Walkability assessments for elderly pedestrians are often based on static representations of the built environment, overlooking temporal variations that influence walking conditions throughout the day. This study develops a network-based dynamic walkability framework that integrates static infrastructural characteristics with time-dependent environmental factors to capture spatiotemporal variability in pedestrian suitability. The approach combines sidewalk and arcade-based pedestrian networks with dynamic variables, including traffic, air quality index (AQI), temperature, humidity, shade, and lighting, evaluated at two time periods (12:00 and 17:00) across weekdays and weekends in three urban contexts in Tainan, Taiwan: a hospital area, a university campus, and a residential neighborhood. Results indicate clear spatial differences, with hospital and campus areas showing higher baseline walkability than residential areas. Dynamic analysis reveals temporal variation, with improvements ranging from approximately 3–8% in institutional environments to over 10% in residential areas. Segment-level results further show that temporal factors can alter pedestrian suitability, particularly in areas with limited infrastructure. Route-based validation demonstrates that the model generates alternative paths that prioritize safety and environmental comfort over the shortest distance. Compared to Google Maps routes, the proposed approach achieves higher average walkability, with improvements ranging from approximately 5% to over 15%, particularly in residential areas. These findings highlight the limitations of static and shortest-path approaches and emphasize the importance of incorporating temporal dynamics. The proposed framework supports time-sensitive routing and age-friendly urban planning strategies.

Existing building digital twins treat indoor environments as static geometric containers, ignoring the dynamic coupling between ventilation structure states and indoor environmental quality. Furthermore, managing indoor and outdoor spaces as separate data silos prevents the continuous assessment of occupant exposure across building boundaries. This paper proposes an environment-aware, indoor-outdoor integrated digital twin framework coupling geometric entity states with physical environmental fields for healthy mobility assessment. The framework utilizes a three-layer architecture. First, the Geometric-Semantic Layer provides a seamless LOD4 model with topologically stitched spaces, modeling ventilation facilities as first-class entities with mutable state attributes (Full Closed, Half Open, Full Open). Second, the Physical Field Layer maps mobile sensing data (PM2.5, CO2) onto semantic entities using a semantic-constrained method, treating walls and closed windows as aggregation barriers. Finally, the Behavioral Response Layer combines entity-level pollution values with pedestrian counts to compute a cumulative Crowd Exposure Index (CEI). Implemented on a Cesium platform, the framework was validated through a week-long university building experiment. Results show indoor PM2.5 in a fully enclosed study room averaged 61.2 μg/m³—1.6 times the outdoor level and 4.1 times the WHO guideline. This resulted in a CEI 12 times higher than in outdoor transit areas. Semantic correlation confirms the "Full Closed" window state primarily drives pollutant accumulation. This validates the framework's core geometry-physics coupling, demonstrating its potential to guide intelligent ventilation interventions and healthy building management.

Rapid industrialization and unplanned urbanization have increased air pollution levels across Indian cities, posing serious environmental and health challenges. This research presents a geospatial assessment of air pollutant behaviour across Nashik city by integrating multi-source remote sensing datasets and real observation datasets from Sentinel-5P, NASA POWER, and CPCB ground observations within a GIS-based analytical framework. Using ward-level mapping and spatial overlays, the study examines the distribution of key pollutants—PM2.5, PM10, NO2, SO2, and CO—and their relationship with environmental and anthropogenic parameters, including land use, road networks, wind direction, temperature, and vegetation density. The results consistently reveal high concentrations of PM2.5, ranging from a minimum of 52.4 µg/m³ to a maximum of 73 µg/m³, and PM10, a minimum of 87.3 µg/m³ and a maximum of 121.5 µg/m³, particularly along high-traffic corridors and industrial zones, which exceed the WHO standards. Correlations with meteorological and vegetative factors further highlight the influence of urban form and climatic conditions on pollutant dispersion. This integrated approach demonstrates how multi-source remote sensing and GIS tools can be effectively employed to identify emission hotspots, support evidence-based policy formulation, and strengthen urban environmental management strategies for sustainable development.

Air pollution is a critical environmental challenge, with nitrogen dioxide (NO₂) from vehicles and industries posing serious health and atmospheric risks. Traditional monitoring is limited, making satellite-based methods essential for large-scale assessment. Korba, Chhattisgarh is an industrial hub of coal mining and thermal power plants is a major pollution contributor. This study investigates the spatiotemporal dynamics, statistical behavior, and long-term trends of NO₂ concentrations over the Korba region from 2019 to 2024, utilizing Sentinel-5P TROPOMI-derived NO₂ column density and the Normalized Difference Pollution Index (NDPI). Year-wise NDPI patterns revealed a consistent pollution hotspot in the central-southern region, with the annual mean NDPI gradually increasing from 0.175 in 2019 to 0.191 in 2023. The monthly NDPI peaked in December-2024 at 0.525, indicating severe winter pollution. Statistical analysis showed moderate variability and a near-symmetric NDPI distribution with occasional spikes near industrial zones. Trend analysis identified a marginal but steady increase in pollution. Autocorrelation analysis revealed strong short-term persistence (lag-1 = 0.594), while spectral analysis identified a dominant annual frequency (0.083 cycles/month) with a peak power of 0.107, confirming the presence of strong seasonal variation and short-term persistence in NO₂ concentration. These results underscore the cyclic yet escalating nature of NO₂ pollution, with notable winter intensification. The findings emphasize the need for targeted emission control strategies and policy-level interventions to manage regional air quality. Future work should integrate ground-based validation and explore meteorological influences to improve predictive accuracy and guide sustainable environmental management.

The accumulation of C in peatlands generally depends on hydrologic conditions that maintain saturated soils and impede rates of decomposition. Boreal Peatlands have provided rich reservoirs of stored C for millennia. However, with climate change, warming and drying patterns across the boreal and arctic are resulting in dramatic changes in ecosystems and putting these systems at risk. As long as peatlands are functioning hydrologically, they will continue to sequester and store carbon. The ability to retrieve and monitor soil moisture from peatlands is of interest for a wide range of applications from hydrological modeling to understanding ecosystem vulnerabilities to increased drought, decomposition and wildfire to monitoring methane flux and peatland restoration. To develop soil moisture retrieval algorithms, we studied a range of boreal peatland sites (bogs and fens) stratified across geographic regions of North America from 2010 to 2024. We developed soil moisture retrieval algorithms from polarimetric C-band (5.7 cm wavelength) and L-band (24 cm wavelength) synthetic aperture radar (SAR) data. Both multi-linear regressions and gradient boosters (XGBoost, CatBoost and Explainable Boosting Machines) were developed. We found that integrating polarimetric SAR parameters that are sensitive to vegetation structure and parameters most sensitive to surface soil moisture in the models provided the best results. Data were withheld for model testing and coefficient of determination, RMSE, unbiased RMSE are reported.

The Peace–Athabasca Delta (PAD) is undergoing long-term ecological change driven by climate warming, hydro-regulation, and fluctuating flood–dry cycles. This study uses a harmonised 40-year Landsat composite time series (1984–2024) to assess vegetation, surface-water extent, and thermal conditions across the delta. An 11-year moving-window Mann–Kendall trend analysis was applied to NDVI, EVI, MNDWI, and LST, retaining only significant Theil–Sen slopes. Significant vegetation–water trends were combined into a 10-class framework that maps greening, browning, wetting, and drying across all landscape types, including ecotones. Parallel LST trends reveal reinforcing or contrasting thermal feedbacks. It provides a coherent basis for interpreting whether vegetation and hydrologic changes reflect ecotone expansion or contraction under thermal variability.

Mountain headwaters of the Eastern Slopes of Alberta (ES) are the primary source of freshwater of major easterly flowing basins in western Canada, supplying a significant volume of water to about four million people. However, increasing temperatures is altering mountain aquatic (open water areas, lakes, reservoirs, rivers, ponds, wetlands) and riparian vegetation (herbaceous and woody/shrub) ecosystems. The ES, Canada, has demonstrated landcover and process changes associated with climate warming, e.g., increases in the air temperature [1] have led to earlier snowmelt, and increased glacier wastage, resulting in higher river flows over a shorter period, which can result in expansion of open water areas during and following peak flow periods [2]. The impacts on wetlands are less visible or well known, and there is a need to evaluate spatial and temporal changes and trends in wetland loss, growth, or genesis across this mountainous ecosystem. Here, we provide a framework for quantifying and assessing multi-decadal wetland extents over the large spatial scale of the ES from 1984 to 2023. We used the historical Landsat archive to produce a remote sensing-based time series landcover classification over the last 40 years in the ES.

Functional Data Analysis (FDA) provides a powerful framework for representing temporal dynamics in remote-sensing data. Building on this concept, this study develops a transfer learning framework using a minimally trained Functional Principal Component Analysis (FPCA)-based feature extraction engine (“FPC engine”) to map dynamic wetlands at large scale. A small set of training locations from Ontario was used to train the FPC engine, which captures dominant seasonal backscatter patterns of open water, shallow water, and marsh-like vegetation. The trained engine was then transferred to the Prairie Pothole Region (PPR) to delineate dynamic wetland classes without extensive local calibration. This label-efficient design—supervised in selecting training locations but unsupervised in feature extraction—reduces field data needs while maintaining strong generalization. Validated results show that the transferred FPC engine effectively separates dynamic wetland classes across contrasting climatic and geomorphic conditions, supporting scalable and cost-efficient monitoring with Sentinel-1 SAR data.

Wetland mapping in boreal environments remains challenging due to complex vegetation structure, subtle and variable terrain gradients, diverse wetland types, and the proportion of treed wetlands. This study develops and evaluates a framework to remotely identify wetland types in Pukaskwa National Park (Ontario, Canada) by integrating multi-scale terrain metrics with multi-temporal Sentinel-2 spectral derivatives. Five years (2017–2021) of Sentinel-2 data were used to derive harmonic NDVI metrics, including linear trend, amplitude, and phase of the first Fourier component, capturing seasonal vegetation and hydrologic dynamics. These spectral predictors effectively delineated open water and non-treed peatlands but struggled in densely forested wetlands where canopy obscures surface moisture signals. To address this limitation, Gaussian scale-space analysis was applied to the Copernicus GLO-30 DSM, informed by FFT-based evaluation of terrain wavelengths (100 m–10 km), to generate multi-scale Local Relief Models and curvature metrics representing depressional and convex landforms. A hierarchical workflow masked open water using Sentinel-1, removed upland convex terrain using LRM-curvature rules, then applied Random Forest classification using field training data and combined spectral-terrain predictors. Accuracy assessment stratified by terrain context showed strong performance in low-lying depressional areas and suppression of false wetland detections in high terrain with local depressions. Reduced accuracy in relatively flat areas was attributed to DSM vertical uncertainty limiting detection of shallow depressions beneath dense canopy, resulting in reliance on optical separability that weakens under closed canopy but improves where tree cover is sparse. Overall, results demonstrate the value of combining Fourier-based temporal descriptors with multi-scale terrain analysis for boreal wetland mapping.

As an essential component in the atmosphere, black carbon (BC) can affect regional and global climate, air quality, and human health. Biomass burning is an important source of BC aerosols, and biomass burning in East Asia is rather active. In this study, we analyze the biomass burning over East Asia in 2010 using MODIS satellite fire data. A global chemical transport model, GEOS-Chem, is used to simulate temporal and spatial variations of BC aerosols and impact of biomass burning on these variations through two numerical experiments: one with all BC emissions while the other without the biomass burning emissions.

The results show that the 2010 biomass burning over East Asia frequently occurred in northeast China, north China, northern India and indo-China Peninsula. In China, biomass burning mostly happened in summer and fall, while in Southeast Asia, biomass burning happened in spring and winter. GEOS-Chem can reasonably reproduce the temporal and spatial variations of BC. The surface concentrations of BC in China are high in the North China and Southwest basins. Such a spatial pattern is similar in four seasons, with seasonality that BC concentrations are the highest in winter, followed by autumn, spring and summer. Sensitivity analysis shows that the biomass burning in East Asia contributed 8.6% BC concentrations in East Asia. Based on the EOF decomposition and correlation analysis, the BC concentrations due to biomass burning in some parts of East Asia was significantly increased through transport of BC in the first mode at 850 hPa in spring and winter.

The Chinese second-generation Directional Polarization Camera (DPC-2) onboard the GF-5(02) satellite provides global multi-angle, multispectral polarization observations, effectively bridging the gap between POLDER/PARASOL and SPEXone/PACE. Using one year of DPC-2/GF-5(02) measurements, land-ocean aerosol products are generated by fully exploiting polarization and angular information to enhance sensitivity to aerosol properties. Ground-based observations from the AErosol RObotic NETwork (AERONET) are used to evaluate the retrieval accuracy of Aerosol Optical Depth at 550 nm (AOD550), Ångström Exponent between 440 nm and 670 nm (AE440-670), and Single Scattering Albedo at 440 nm (SSA440), demonstrating the stability and reliability of the retrievals. For AOD550, the Root Mean Square Error (RMSE) and bias are 0.109 and -0.006 over land, and 0.071 and -0.001 over ocean. For AE440-670, the RMSE and bias are 0.488 and -0.151 over land, and 0.275 and -0.047 over ocean. For SSA440, the RMSE and bias are 0.044 and 0.003 over land, and 0.039 and 0.002 over ocean. Comparisons with mainstream satellite aerosol products indicate comparable and consistent accuracy. Overall, these results provide a coherent global characterization of aerosol distribution and properties, highlighting the strong potential of DPC-2/GF-5(02) for long-term aerosol monitoring and climate research.

This study involves remote sensing and Lidar-based data analysis to quantify the aerosol extinction profile under different urban patterns and seasons.

Direct and indirect climate forcing from Asian dust storms has been well documented, such as lifted dust aerosols becoming cloud-forming particles and changing radiation flux from surface to the top of atmosphere. However, whether such forcing becomes distinguished as related to dust origins remains unclear. Here we present a comparative analysis of historical dust storms in China originating in Mongolia and Xinjiang from 2016-2023, and determine their respective dominators by involving their individual and combined influence on dust storms. Most dust storms originated in Mongolia, with observed long-range transport and global scale development, in comparison to those originating in Xinjiang. During dust storms, cloud properties such as cloud droplet radius and cloud retrieval fraction liquid had nonlinear response, and a dominant role in 60.2% of the study area. Climate conditions such as surface thermal radiation and dewpoint temperature became dominated in periphery of dust storms. Xinjiang-originated dust storms, in contrast, were dominated by local aridity (65.2%). As the aridity decreased, dust storms were dominated by total precipitation, with increase from 0.5 up to 3.6, and the influence of surface heat flux decreased. Heat-flux-dominated regions encountered increased aridity, and the dominance of total precipitation was neutralized. These findings have important implication for global management and mitigation of Asian dust emissions.

The Arctic Observing Mission (AOM) is a satellite mission concept under study by the Canadian Space Agency (CSA), in partnership with Environment and Climate Change Canada (ECCC) and Natural Resources Canada (NRCan). AOM would use two satellites in a highly elliptical orbit (HEO) to enable frequent observations of meteorological variables, greenhouse gases (GHGs), space weather and air quality (AQ) over northern regions, reaching beyond the usable viewing range of geostationary satellites. These observations are important for operational activities, environmental monitoring and scientific research aligned with the Government of Canada priority of enhancing Arctic and northern situational awareness and security.

This study presents a hyperspectral remote sensing approach for monitoring global CO₂ point source emissions using China’s GF5 and ZY1 satellites. By applying the matched filter method in the 1.6 μm and 2.0 μm absorption band and the Integrated Mass Enhancement (IME) technique, this study successfully detects and quantifies emissions from multiple facilities within a single scene—demonstrated in a high-density industrial cluster in Xinjiang. Results show current systems can detect power plants with annual emissions above 2.90 MtCO₂, covering 6.74 GtCO₂/year globally across eight sectors. While power and chemical sectors are well captured, cement and petrochemical emissions remain poorly detected, highlighting the need for improved sensitivity to low-intensity sources.

Biomass burning releases a large amount of pollutants including carbon monoxide (CO), and generates secondary pollutants, e.g., ozone (O3). Both CO and O3 are major pollutants and can also significantly affect tropospheric chemistry. Understanding O3-CO relationship is important for evaluating transport and evolution of the pollutants in fire plumes. Here, we analyse the satellite remote sensing of fire count data from MODIS, satellite remote sensing of CO and O3 from AIRS, and the simulation of the global atmospheric chemistry model GEOS-Chem in the middle and lower troposphere during June and August of 2010. AIRS can capture fire-induced CO and O3 enhancements (ΔCO and ΔO3) well in fire-affected and fire-plume outflow regions. Two areas with high ΔCO and ΔO3 include central Africa and northwestern South America in the tropics, where the numbers of hotspots are the large in the MODIS fire data. AIRS CO and O3 in fire plumes are highly correlated in 850 hPa and 500 hPa. The GEOS-Chem simulation show CO and O3 enhancement in northwestern South America, but with lower ΔO3/ΔCO values. These findings highlight the importance of integrating satellite observations with atmospheric chemistry modelling on refining fire-affected air quality and tropospheric chemistry assessments.

InSAR Norway (InSAR.no) is one of the world’s first fully operational, open-access national ground-motion services. Jointly operated by NGU, NVE and the Norwegian Space Agency, with processing by NORCE on NGU’s high-performance computing cluster, it provides nationwide deformation time-series from Copernicus Sentinel-1 data. The service delivers more than five billion measurement points annually through a public web portal and is widely used for landslide mapping, infrastructure monitoring and climate-related research. It has transformed how Norway identifies and manages unstable slopes, supports early warning and infrastructure safety, and integrates satellite data with in-situ monitoring through networks of snow-protected corner reflectors.

Experience from InSAR Norway directly informed the European Ground Motion Service (EGMS) under the Copernicus Land Monitoring Service, which scales the same operational principles to continental level. EGMS demonstrates that harmonized, validated and open InSAR products can be maintained across national borders.

Building on these achievements, this paper outlines the concept of a Global Ground Motion Service (GGMS)—a federated system providing standardized, GNSS-anchored ground-motion data worldwide. Such a service would combine open satellite data, interoperable processing frameworks and regional capacity-building to support disaster-risk reduction and resilience globally. As the global community invests in disaster-risk reduction, an open GGMS could become one of the most tangible and enduring legacies of the Copernicus era.

Satellite geodesy has become a cornerstone for mapping tectonic deformation, fault activity, and seismic hazard through measurements of surface velocities and strain rates. Yet, in vast regions of diffuse continental deformation, such as the Alpine–Himalayan Belt, observational coverage remains limited. Historically, large-scale studies have relied on sparse GNSS networks, which cannot resolve shorter-wavelength deformation features in many areas. To address this gap, we processed Sentinel-1 radar acquisitions from 2016 to 2024 to generate transnational surface velocity fields and time series at 1 km resolution, spanning more than 11,000 km from southern Europe to eastern China and covering over 20 million km². Our solution integrates more than 220,000 Sentinel 1 SAR images with a newly compiled GNSS dataset, all referenced consistently to the Eurasian frame. From these velocities, we compute horizontal strain rates by taking spatial gradients, providing near-continuous deformation maps across the planet’s largest actively deforming zone. Horizontal motions and strain patterns are primarily tectonic, exhibiting a dual character: strongly localised along major faults yet broadly distributed elsewhere. In contrast, short-wavelength vertical signals largely reflect non-tectonic processes, especially widespread groundwater depletion. These new velocity and strain-rate products constitute foundational datasets, offering a detailed view of continental deformation at a transcontinental scale that feed into the Disaster Risk Management cycle.

The west coast of Canada occupies an active subduction zone and is the host of an often underestimated threat of volcanic eruption. This tectonically active region is the home of 348 known volcanic vents that have been active since the Pleistocene, 54 of which are Holocene in age or younger. The annual probability of any eruption has been estimated at 1/200, while the annual probability of a major explosive eruption has been estimated at 1/3333.

In 2021 the Geological Survey of Canada published a volcanic threat ranking study) which used a threat score assignment methodology developed by the United States Geological Survey. In this study, we describe how the results of this threat ranking guide the acquisition strategy of routine RCM SAR data over the highest threat volcanoes in and around Canada.

We describe the architecture of the fully automated, cloud-based processing system that routinely searches for fresh RCM SAR data, ingests and processes the raw data and displays processed InSAR data on a purpose-built interface for scientific analysis.

With the proliferation of the heavily automated InSAR measurements, human analysis of vast volumes of data becomes challenging. In this research, we also describe the application and performance of an open weight deep learning model trained specifically for the purpose of detecting magmatic unrest in InSAR data. We demonstrate a deformation detection threshold of 6.2 cm and a true positive rate of 0.98 using observations from a real magmatic unrest event in Reykjanes, Iceland through 2023-2024.

We provide updates on the NASA-ISRO NISAR synthetic aperture radar mission and the NASA OPERA project. NISAR launched in June 2025 and began science operations in November 2025. The mission status will be presented and products for different science applications shown. The OPERA project produces four different product streams to support agency information needs, with the Dynamic Surface Water Extent (DSWx), Surface Disturbance (DIST), and Surface Displacement (DISP) products already available, and algorithm development underway for a future Vertical Land Motion product. These are generated from a variety of sensor data, including harmonized Landsat/Sentinel-2, Sentinel-1, NISAR, and SWOT. Examples shown will focus on the DISP products, currently generated from Sentinel-1 data and with a new product line using NISAR data to roll out in early 2027.

Monitoring precursory ground deformation is essential for assessing landslide hazard in regions where hydrological conditions strongly influence surface stability. In Québec’s Saguenay–Lac-Saint-Jean (SLSJ) region, numerous surface failures have occurred in highly sensitive postglacial marine clays, where rainfall, snowmelt, and groundwater fluctuations act as dominant triggers. Although Persistent Scatterer InSAR (PSInSAR) enables regional monitoring of slow ground deformation, its utility for short-term prediction remains limited by the temporal gap between Sentinel-1 acquisitions. This study investigates whether hydrological time-series, when integrated with PSInSAR displacement trends, can be used to forecast the line-of-sight (LOS) displacement observed at the satellite acquisition immediately preceding documented failure events.

A dataset of 102 historical failures (2018–2024) was assembled and paired with 168 Sentinel-1 ascending scenes processed through the StaMPS PSInSAR workflow. Daily precipitation, air temperature, groundwater level, and terrain slope were compiled and temporally synchronized with LOS displacement time series.

An XGBoost regression model was trained to predict the LOS displacement at the subsequent Sentinel-1 acquisition, using an 80/20 train–test split and five-fold cross-validation. Model performance was evaluated using Pearson’s r, MAE, and RMSE. Results show strong predictive skill, with r = 0.82, MAE = 4.36 mm, and RMSE = 6.26 mm. Feature importance analysis highlights the dominant role of recent PSInSAR displacement and groundwater variability. These findings demonstrate the feasibility of integrating hydrological and InSAR time-series to forecast pre-failure surface displacement, supporting the development of satellite-based early warning strategies for hydrologically sensitive terrain.

This research validates social media geospatial tags using Sentinel-1A Interferometric Synthetic Aperture Radar (InSAR) data processed on Google Earth Engine, focusing on Hurricane Harvey as a case study. The study addresses critical uncertainties regarding the spatial reliability of crowdsourced disaster information, which has limited integration of social media data into operational disaster management frameworks.

Methodology: The methodology integrated 144,546 geotagged posts from Twitter, Facebook, and Instagram collected during Hurricane Harvey (August 25 - September 3, 2017) with Sentinel-1A SAR imagery processed on the Google Earth Engine cloud platform. InSAR analysis identified 1,247 square kilometers of flooded areas in the Houston metropolitan region. Spatial validation employed buffer zone analysis at 500m, 1km, and 2km distances, with temporal alignment matching social media timestamps to SAR acquisition dates.

Results: Results demonstrate that 68.3% of flood-related social media tags fell within actively flooded areas using 1km buffers, with accuracy increasing to 82.1% within 500m buffers, compared to only 12.7% random expectation. Temporal analysis revealed social media activity peaked 6-18 hours before peak SAR-detected flooding, suggesting potential early warning capabilities. The cloud computing paradigm reduced processing time from weeks to 4-6 hours, enabling near-real-time validation.

Conclusion: This study establishes that validated social media geospatial information can effectively augment satellite-based disaster monitoring systems, particularly during initial response phases when temporal resolution is critical. The integration framework demonstrates operational feasibility for multi-source geospatial data fusion in disaster risk reduction applications.

The paper presents an overview of the European Ground Motion Service (EGMS), a CLMS product that delivers continent-wide, high-resolution measurements of ground motion to users based on Sentinel-1 data. It explains the EGMS architecture, which integrates Persistent and Distributed Scatterer techniques to generate standardised products—Basic, Calibrated, and Ortho—allowing millimetric monitoring of land motion across Europe. The paper emphasises how EGMS fills a critical gap between localised ground measurements and global geodetic frameworks, offering harmonised datasets for hazard assessment, infrastructure management, and policy-making. Applications discussed include subsidence and uplift detection, landslide mapping, and analysis of critical infrastructure.

Looking forward, the paper outlines a potential evolution towards an expansion of the EGMS concept beyond Europe. This would enable standardised, freely accessible deformation data to support global hazard mitigation and climate adaptation. The paper concludes that while technically feasible, a global implementation will require strategic GNSS densification and international cooperation to ensure reliability and equitable access.

Accurate large-scale forest canopy height mapping is critical for biomass estimation and carbon monitoring, yet remains constrained by the limitations of individual remote sensing systems. This study presents a multisensor machine learning framework that integrates GEDI LiDAR with Sentinel-2, Sentinel-1, ALOS-2 PALSAR-2, and 3DEP terrain data to generate a 25 m resolution canopy height model (CHM) for the Northeastern United States in 2022. A key contribution is an adaptive GEDI relative height (RHad) strategy that selects optimal RH metrics based on canopy density, improving generalization across heterogeneous forest conditions compared to any single fixed RH metric. Independent validation against airborne LiDAR and USDA FIA plot data confirms that RHad achieves the highest accuracy and lowest bias of all configurations tested. The resulting regional canopy height map provides a reliable baseline for large-scale forest monitoring and future multitemporal analyses.

Timely and accurate detection of deforestation is essential for managing tropical forests, yet individual Earth observation sensors have inherent limitations. Multispectral imagery offers detailed spectral information on vegetation properties but is frequently hindered by cloud cover, while Synthetic Aperture Radar (SAR) imagery provides insights on vegetation structure independent of weather conditions but is sensitive to moisture variability and residual vegetation post-clearing. The complementary nature of these data has motivated multi-source fusion approaches, though most existing methods rely on offline processing or decision-level integration, limiting their real-time applicability.

This study generalizes a Bayesian Online Changepoint Detection (BOCD) framework based on the recursive estimation of the number of acquisitions since the last change to asynchronous, irregularly sampled Sentinel-1 SAR and Sentinel-2 multispectral time series. A dynamically weighted fusion mechanism is implemented, in which each sensor’s relevance reduces with increasing time since its last observation, according to a physical decay model. The resulting method, named ms-BOCD, enables interpretable, and Near Real-Time (NRT) detection of forest loss.

The ms-BOCD method is validated using MapBiomas Alerta reference data spanning deforestation polygons ranging from 0.1 to 50 hectares in the Brazilian Amazon.

Compared to $VH$-BOCD (BOCD using Sentinel-1 cross-polarization only) and the operational RADD and TropiSCO systems, ms-BOCD achieves a 25% improvement in detection performance and maintains 13% fewer false alarms than Global Forest Watch (GFW), a platform that aggregates multiple independent deforestation alert products. Overall, these results demonstrate the strong potential of multi-source Bayesian fusion for operational tropical forest monitoring.

This study assesses long-term forest change in Liberia’s Community Forest Management Areas for Conservation (CFMACs) and Protected Areas (PAs) from 2002 to 2024 using an integrated Landsat–Google Earth Engine (GEE) and an ArcGIS Pro workflow. Annual dry-season composites for three time periods were classified using a Random Forest model with 81.7% accuracy (Kappa = 0.781). Results show contrasting governance outcomes: CFMACs experienced modest forest gains from 2002–2014 and localized losses thereafter, while PAs exhibited larger overall gains but also greater cumulative forest loss, particularly along concession boundaries. Stability analysis revealed that PAs retained a higher proportion of Mature Forest over the 20-year period, whereas CFMACs showed more dynamic turnover and localized regrowth. The combined GEE/ArcGIS approach provides a scalable, transparent monitoring framework and demonstrates how governance type influences forest persistence, degradation, and recovery across Liberia’s tropical landscapes.

Accurate assessment of tree cover change is essential for monitoring deforestation, carbon emissions, and restoration progress. However, validation of global forest change products remains limited by the scarcity of consistent reference data. We present a benchmark dataset for tree canopy cover change evaluation across North America, derived from multitemporal airborne LiDAR data from the National Ecological Observatory Network (NEON). Using canopy cover maps from 2016–2022, we identified tree cover loss as a decrease of at least 20% in canopy cover persisting across multiple time steps. Thirty NEON sites spanning diverse biomes were included, forming a spatially and temporally robust reference for change detection.

We demonstrate the benchmark applicability by evaluating two global products: Forest Carbon Diligence (FCD) from Planet Labs, and the Global Forest Change (GFC) from University of Maryland. Across all sites, both products showed strong agreement with the LiDAR benchmark (r = 0.90 for FCD; r = 0.88 for GFC), though both underestimated change extent. Categorical metrics revealed higher precision than recall, indicating conservative detection thresholds relative to the benchmark.

This study establishes the NEON LiDAR-based benchmark as a valuable open resource for assessing and improving large-scale canopy cover change datasets. The approach highlights the importance of high-resolution, temporally consistent reference data for evaluating the accuracy of global monitoring products and guiding improvements in forest carbon accounting and conservation applications.

Under climate and human pressures, vegetation in Inner Mongolia exhibits complex fragmentation and degradation. Scientifically inverting its spatiotemporal dynamics is crucial for regional ecological restoration. To address the challenges faced by traditional cellular automata (CA) models in large-scale complex ecological transition zones—such as computing power bottlenecks and subjective transition rules—this study proposes a cloud-based vegetation degradation simulation and inversion framework (CA-VDS) via Google Earth Engine. By coupling Random Forest (RF) and an Improved Genetic Algorithm (IGA) with CA, the framework extracts nonlinear driving potentials and automates the optimization of bidirectional transition thresholds. Validation against the 2020 baseline shows CA-VDS effectively resolves manual parameter tuning limitations. Furthermore, it smooths the spectral fluctuations caused by short-term sporadic disturbances through the underlying spatial neighborhood mechanism, demonstrating its value in simulating potential ecological degradation risks and developmental trajectories. This work not only verifies the reliability of CA-VDS in analyzing complex nonlinear ecological processes, but also establishes a reliable parameter baseline and model paradigm for subsequent integration with CMIP6 and other multi-scenario data to conduct long-term future ecological predictions.

This study explores the application of Particle Swarm Optimization (PSO) to enhance vegetation indices (VIs) for the assessment of woody vegetation in a semi-arid savannah ecosystem. By optimizing VIs, the research aims to improve the discrimination between vegetated and non-vegetated areas, facilitating a more accurate random forest classification for habitat quality assessment. The optimization process preserves minimum VI values across different sensors to maintain lower bounds of reflectance, ensuring ecologically valid signals are represented, particularly in low-vegetated areas. Results indicate that maximum VI values increase post-optimization, enhancing sensitivity to canopy vigor, stress, health, and presence. The study highlights the effectiveness of UAV-derived indices, such as NDVI, NDRE, and SAVI, in capturing the dynamics of vegetation health and dryness, thereby contributing valuable insights into remote sensing methodologies for ecological monitoring.

The paper proposes the STF-Net (Spatial-Textural-Frequency Network) framework, designed to achieve a paradigm shift from traditional "change detection" to "suspected area identification," precisely identifying suspected abandonment areas and effectively suppressing pseudo-changes. The core of this framework lies in its fine-grained four-level annotation system and a three-stream parallel feature extraction architecture. The four-level annotation system includes "confirmed abandonment," "suspected abandonment," "non-abandonment change," and "no change," providing a robust data foundation for the model to learn the "suspected" concept, thereby compensating for the lack of "user-oriented" definitions in existing research. The three-stream parallel feature extraction architecture captures changes in geometric information (location, shape) via the spatial stream; quantifies the transition of surface texture from ordered to disordered, capturing structural degradation due to abandonment, through the textural stream; and analyzes periodic structural information in images, identifying the disappearance of periodic structures caused by cessation of cultivation, using the frequency stream. These three types of features are deeply fused, comprehensively utilizing information from different modalities, significantly enhancing the model's adaptability and identification accuracy in complex scenarios.

Estimating camera poses jointly from aerial and ground imagery remains difficult because large viewpoint changes reduce overlap, alter appearance, and weaken the geometric assumptions relied on by both classical photogrammetry and recent AI-based reconstruction models. This paper presents a controlled comparison between a classic photogrammetric approach represented by COLMAP and a cross-view fine-tuned end-to-end model based on Dust3R. Tests are carried out on a London building scene containing 10 aerial and 29 ground images. Fine-tuned Dust3R reconstructs the full image set, whereas COLMAP successfully registers 24 ground-level images. Because both reconstructions are defined only up to an unknown similarity transform and no ground-truth poses are available, we evaluate the shared subset through 7-DoF similarity transformation analysis rather than direct metric pose errors. After transformation, the translation RMSE of the shared camera centres is 10.0\% of the reconstructed scene diagonal in the fine-tuned Dust3R coordinate frame. We further compare pairwise geometric support using a unified fundamental-matrix RANSAC evaluation over 406 image pairs. The AI-based pipeline achieves substantially higher inlier ratios than photogrammetric pipeline under the same verification settings, indicating more successful cross-view orientation. The study contributes a clearer evaluation protocol for mixed aerial-ground pose estimation without ground truth, together with an empirical analysis of robustness, alignment behaviour, and current limitations of both pipelines.

We propose a generic method for epipolar resampling that is not tied to a specific camera model. We demonstrate the effectiveness of the approach on a central perspective, pushbroom and pushbroom panoramic camera models. We also devise an \textit{epipolarability index} that measures the suitability of an image pair for epipolar rectification, and provide a formal derivation of the ambiguity bound to epipolar resampling. An open-source implementation of the algorithm is available at github.com/micmacIGN/micmac

Thermal infrared (TIR) imaging provides surface temperature of the objects and reveals heat-transfer patterns of buildings, which supports applications such as insulation inspection, energy leakage, and thermal bridge detection. However, the TIR image dataset with reliable annotations for deep learning remains scarce, as the labeling process is time-consuming and tedious, and particularly challenging due to the low-texture and blurred features of TIR images. To address this challenge, we propose ThermalAssist, a geometry and gradient-aware framework designed to assist thermal anomaly labeling in TIR imagery. By combining sparse manual annotations with dense correspondence via flow-based propagation, the framework efficiently transfers labels across image sequences while preserving semantic consistency and boundary integrity. Experiments on the TBBR dataset demonstrate that ThermalAssist can transfer labels between images, achieving up to 21% higher F1-score and 35% higher precision compared to state-of-the-art tracking-based baselines. It also helps identify missing annotations and boundary inconsistencies for quality checks. This work provides a foundational tool for quality-assured thermal annotation pipelines and represents a key step toward more scalable, reliable, and intelligent labeling of thermal imagery.

Accurate image matching is essential for the precise orientation of airborne imagery, yet modern feature matchers are rarely evaluated on real aerial data with great temporal, seasonal, and radiometric changes. For this study, we introduce the AerialRefMatch dataset, which comprises 51 challenging aerial images and corresponding true-ortho reference data. We benchmark classical and deep learning–based matching algorithms on AerialRefMatch, considering two scenarios: matching original images and matching approx-orthorectified images generated using GNSS/IMU orientations. For each method, image-based ground control points are derived and used for single-image pose estimation; accuracy is assessed via independent checkpoints. Results show that directly matching on original images is very difficult: fewer than 14\% of images can be oriented with pixel-level accuracy. When approx-orthorectification is used, performance improves substantially. JamMa, SIFT, and SuperPoint+LightGlue achieve pixel-level accuracy for up to 30\% of images, with JamMa being most robust on difficult cases and SIFT-based variants being more precise on the easier ones. Deep detector-free models such as ELoFTR and RoMa are less accurate but more robust to the original images than other models. Overall, state-of-the-art deep learning-based matchers still struggle with large rotations, scale differences, and semantic differences, and strongly benefit from prior image orientation knowledge and lack sub-pixel precision.

Deep learning-based image matching faces a critical challenge when deployed on computationally constrained embedded aerial devices. Transformer-based architectures, particularly the scaled dot-product attention mechanism, incur high computational costs that limit inference speed for real-time applications. To address this bottleneck, we propose FastGlue, a sparse feature matching algorithm that adapts the LightGlue architecture through two targeted modifications: replacing the scaled dot-product attention with a ReLU-based linear attention module, and reducing the depth of the graph neural network. These changes reduce computational complexity while maintaining competitive matching performance. Evaluations on HPatches and MegaDepth-1500 benchmarks show that FastGlue achieves accuracy comparable to LightGlue while improving inference speed—from 20.05 ms to 17.05 ms on GPU, and from 840.45 ms to 665.85 ms on an RK3588 embedded CPU. Our work demonstrates that targeted architectural simplifications can yield meaningful efficiency gains for deep learning-based feature matching on resource-constrained platforms.

We present SCOP, an open-source and educational framework for generic photogrammetric bundle adjustment built in Python and powered by JAX automatic differentiation. SCOP removes the need for manual Jacobian derivation by expressing all projection models as pure mathematical functions with automatically computed exact derivatives.

The framework supports multiple camera geometries (pinhole, fisheye, equirectangular) and optimization methods (Gauss-Newton, Gauss-Newton-Armijo, Levenberg-Marquardt, Gradient Descent). Its modular architecture, separating cameras, images, and observations, allows easy extension to new sensors and constraint types, including GNSS positions, ground control points, and geodetic observations.

A hybrid computation pipeline combines JAX for differentiation with a Rust backend for sparse Schur complement elimination, achieving ~0.5 s per iteration on a real-world dataset with 79k unknowns and 181k observations. Following classical least-squares photogrammetry, SCOP provides rigorous uncertainty estimation through covariance matrices, normalized residuals, and reliability indices. With synthetic data tools and interactive 3D visualization, it enables transparent teaching and reproducible research.

Local feature matching in complex scenes is hindered by semantic ambiguity, where detectors often latch onto transient or repetitive patterns. We present TriCo-Net, which learns semantically aware and discriminative local features by enforcing a Triple Consistency (TriCo) principle across implicit semantics, scale, and spatial context. During training, an Implicit Semantic Strategy (ISS) distills cues from a segmentation teacher to modulate keypoint reliability and descriptor learning, while introducing no overhead at inference. A Scale-wise Semantic Harmonizer (SSH) aligns and fuses feature-pyramid levels to ensure cross-scale coherence, and a Global Context Propagator (GCP) broadcasts scene-level dependencies to resolve local ambiguities. On Aachen Day–Night v1.1, TriCo-Net achieves strong and consistent gains in visual localization, particularly under night conditions, and exhibits robustness to blur, noise, and large homographies. Ablations show complementary benefits from ISS, SSH, and GCP, with ISS contributing most at tight thresholds and at night. TriCo-Net narrows the day–night performance gap while maintaining mid-range throughput, offering a practical trade-off between robustness and efficiency.

Urban surface structure and component temperatures induce significant thermal anisotropy (TA), resulting in substantial differences in observed surface temperatures across varying viewing angles. Although previous studies have investigated the temporal dynamics of TA through observations and modeling, its spatial differentiation over heterogeneous surfaces remains poorly constrained. Resolving how surface heterogeneity influences TA is hindered by the coarse spatial resolution and limited angular sampling of current multi-angle satellite observations. Consequently, most mainstream thermal-anisotropy models were developed for simplified scenes and lack systematic evaluation of their applicability to complex urban environments. To address these challenges, we coupled the microscale 3D urban energy balance model (TUF-3D) with the state-of-the-art Discrete Anisotropic Radiative Transfer (DART) model. This approach allows for rapid and accurate TA modeling of hypothetical urban scenes with varying geometric structures and component temperatures, thereby quantifying the impact of surface heterogeneity on TA. Building height variability was used to represent geometric heterogeneity, while differences in building material properties were used to characterize component temperature heterogeneity. To evaluate , The results of a series of sensitivity experiments have validated the individual effects of geometric and component temperature heterogeneity on TA. From the perspective of component temperature, changes in average component temperatures result in a maximum TA difference of 7.29 K, while temperature variability alone contributes only 0.54 K. These findings suggest that assuming simplified scenes with uniform building heights or homogeneous component temperatures can introduce biases in TA simulations, potentially compromising the accuracy of models correcting for the angular effects of land surface temperature.

Understanding the thermal behavior of soil and vegetation separately is essential for interpreting land–atmosphere energy exchange, diagnosing ecosystem stress, and improving land surface modelling. However, conventional satellite LST products only provide a mixed radiometric signal, masking the distinct thermal responses of soil and canopy. This study introduces GloSVeT, the first global dataset that provides monthly surface soil and vegetation component temperatures at 0.05° resolution for 2003–2023. The dataset is generated using an enhanced multisource fusion framework that integrates multi-temporal MODIS observations with ERA5-Land skin temperature and vegetation structural information to retrieve physically consistent component temperatures. We summarize the data sources, modelling framework, and global implementation strategy, and present an independent evaluation using flux-tower networks with screened spatial representativeness. Validation results show strong agreement with in-situ measurements, with correlations typically above 0.9 and RMSE around 2 K for both soil and vegetation temperatures. Seasonal variations in performance reflect expected hydrothermal conditions, and a small cool bias is attributable to the temporal sampling of satellite observations. GloSVeT provides a new basis for studying surface energy partitioning, monitoring hydrothermal dynamics, and supporting ecosystem and climate model applications.

This study presents the development of a ground-based observation system designed for vicarious calibration of satellite sensors operating in the mid-wave infrared (MWIR) region. Conventional natural targets used in LWIR calibration lack spectrally stable emissivity in MWIR, motivating the need for dedicated reference targets and high-sensitivity measurement instruments. We introduce a thermally controlled ground reference target whose effective emissivity can be tuned by adjusting the ratio of water and metal surfaces using perforated plates of varying hole diameters. In parallel, an MWIR radiation thermometer employing lock-in detection was developed to enable accurate measurement of low-signal MWIR radiance from room-temperature targets. The system achieved measurement uncertainties down to 20–70 mK. A field campaign was conducted at the Goheung Aerospace Center using the integrated reference targets and radiation thermometer to validate performance under real environmental conditions. The results demonstrate the feasibility of applying controlled emissivity targets and lock-in-based MWIR radiometry to improve the accuracy of MWIR vicarious calibration frameworks.

A Method for Identifying Industrial Heat Sources

This study addresses the limitations of conventional urban shortwave radiation simulations in representing complex three-dimensional morphology. A parameterization approach for large-scale urban sky view factor was proposed, significantly improving computational efficiency and spatial adaptability. Based on this, an urban solar shortwave radiation transfer model was developed to quantitatively characterize the shading and reflection effects of building clusters. Furthermore, a novel remote sensing inversion method for urban surface reflectance and solar radiation parameters was introduced, enabling high-accuracy estimation of surface radiative properties and offering a new technical pathway for urban thermal environment and energy balance research.

Urban Heat Islands (UHIs) emerge from reduced vegetation, impervious surfaces, and anthropogenic heat emissions, leading to elevated surface temperatures in urban areas. Monitoring UHIs at fine spatial and temporal scales requires thermal data capable of capturing both urban heterogeneity and daily variability—conditions not satisfied by the native 1 km resolution of MODIS Land Surface Temperature (LST).

This study presents a regime-aware machine learning workflow to downscale daily MODIS LST to the native spatial scale of MODIS NDVI (231 m) over Bologna (Italy), using only MODIS-internal predictors and meteorological forcing. The approach adopts a two-stage architecture: a Ridge regression model estimates a day-level atmospheric bias, while a Random Forest reconstructs pixel-level residuals to recover fine-scale thermal variability from vegetation, land-cover, topographic, and atmospheric predictors. To account for atmospheric control, the dataset is partitioned into three thermal regimes (COLD, MILD, HOT), with independent models trained for each regime.

Pre-processing and data integration were performed in Google Earth Engine using MODIS LST (MOD11A1/MYD11A1), NDVI, SRTM-derived terrain variables, and built-up fraction from ESA WorldCover. Experiments show strong predictive performance (RMSE < 1 K; R² ≈ 0.90) and spatial patterns consistent with Local Climate Zones.

The MILD and HOT regimes provide the largest enhancement in spatial detail compared to the original MODIS product, while the COLD regime shows reduced performance, likely due to weaker surface–atmosphere coupling. Results highlight that atmospheric conditions play a dominant role in downscaling accuracy, exceeding the impact of model architecture. The framework enables scalable, daily UHI monitoring and supports heatwave analysis and climate-resilient urban planning.

Nighttime optical remote sensing provides valuable insights into natural and, in particular, human activities. This study evaluates the nighttime imaging capabilities of the Sentinel-2 mission using the only available nighttime acquisition not limited to ocean observations for dark signal calibration, covering the United Arab Emirates with Dubai in 2015. We checked the detection limit using granules over the Persian Gulf, extracted radiance spectra for different regions of interest, and analysed lighting types and temperatures. Results suggest a conservative nighttime detection limit of approx. 0.37 W/m²/um/sr for visible/near infrared bands, and 0.08 W/m²/um/sr for short-wave infrared bands. Sentinel-2’s high spatial resolution and multispectral bands, although designed for daytime observations, were capable of detecting and classifying bright visible/near and short-wave infrared emitters. Comparisons with hyperspectral EnMAP imagery acquired in 2025 validated the classifications and revealed changes in urban lighting over a decade. While limitations apply, this study highlights S2’s potential for nighttime remote sensing and supports considerations of nighttime capabilities for future satellite missions.

Multi-view image-to-point label transfer is an effective strategy for 3D semantic segmentation, but its performance largely depends on how predictions from multiple image observations are fused for each 3D point. Most existing pipelines rely on hard voting or handcrafted weighting rules, which do not explicitly learn the reliability of each view under varying geometric and image-quality conditions. In this paper, we introduce DeepChoice, a lightweight view-weighting module for image-guided 3D semantic segmentation. For each visible observation of a 3D point, DeepChoice exploits a compact set of visibility cues, including incidence angle, range, contrast, sharpness, signal-to-noise ratio, and saturation, to predict normalized per-view weights used to aggregate 2D semantic class probabilities into final 3D point-wise predictions. The method is sensor-agnostic, requires no meshing, and can be integrated as a replacement for standard multi-view fusion rules. Experiments on the full GridNet-HD benchmark show that DeepChoice improves over hard voting by 3.85 mIoU points and over mean-probability fusion by 1.26 points, while reducing the gap with the AnyView oracle upper bound. The largest gains are observed on thin and difficult classes such as conductors, pylons, and insulators. Furthermore, a complementary evaluation on the Images PointClouds Cultural Heritage}dataset shows that the proposed weighting strategy remains beneficial under a very different acquisition context and scene structure, yielding a 1.55 mIoU point improvement over hard voting. These results show that learning how to weight views is a simple yet effective way to strengthen image-guided 3D semantic segmentation pipelines. Code is publicly available at: https://huggingface.co/heig-vd-geo/DeepChoice.

Textured 3D meshes jointly encode geometry, topology, and appearance, yet their irregular structure poses significant challenges for deep-learning-based semantic segmentation.

While a few recent methods operate directly on meshes without imposing geometric constraints, they typically overlook the rich textural information also provided by such meshes. We introduce a texture-aware transformer that learns directly from raw pixels associated with each mesh face, coupled with a new hierarchical learning scheme for multi-scale feature aggregation.

A texture branch summarizes all face-level pixels into a learnable token, which is fused with geometrical descriptors and processed by a stack of Two-Stage Transformer Blocks (TSTB), which allow for both a local and a global information flow.

We evaluate our model on the Semantic Urban Meshes benchmark and a newly curated cultural-heritage dataset comprising textured roof tiles with triangle-level annotations with damage types.

Our method achieves 81.9\% mF1 and 94.3\% OA on SUM, and 49.7\% mF1 and 72.8\% OA on new dataset, substantially outperforming existing approaches.

Automatic pothole detection is important for improving road maintenance and transportation safety. While image-based pothole detection often struggles under poor lighting and weather conditions, point cloud data provides a robust alternative by capturing detailed surface geometry. Machine learning has demonstrated strong performance in point cloud classification. While traditional machine learning is simpler and relies on handcrafted features, deep learning models are more powerful, as they learn complex, high-dimensional patterns directly from the input data. While most existing work relies on deep learning models, which are time-consuming to train and require extensive labelled datasets, potholes can be well described by geometric features, making pothole detection well-suited for feature engineering. This paper compares traditional machine learning and deep learning approaches for pothole classification using point cloud data, to evaluate whether the added complexity and data demands of deep learning models are justified, or if traditional machine learning techniques are sufficient for accurate classification. A dataset with labelled pothole instances is created to train both models. The machine learning approach uses manually engineered geometric features as input to an ensemble classifier, while the deep learning model is trained on sampled data. Experimental results show that the machine learning approach outperformed the deep learning model. These results suggest that for this particular task, where informative domain-specific features can be manually engineered, the machine learning approach offers a more practical and efficient solution for real-world deployment, where labelled data may be limited.

Synthesizing realistic fac¸ade views of individual trees from nadir-view remote sensing data would transform large-scale forest

analysis, yet remains unsolved due to data scarcity and task ambiguity. We present the first conditional diffusion model to generate

structurally plausible fac¸ade views of individual tree crowns from single nadir-view LiDAR rasters, leveraging the FOR-species20K

benchmark dataset. Our approach integrates nadir projections with tree species and height within a U-Net-based denoising diffusion

framework. Experiments demonstrate that nadir imagery alone is insufficient, but conditioning on species and height enables

synthesis of visually realistic, species-specific fac¸ade views. The fully conditioned model achieves substantial gains in perceptual

(LPIPS: 0.184) and structural (SSIM: 0.576) similarity, outperforming nadir-only baselines by more than twofold. Our results

establish that ancillary attributes critically constrain the solution space, enabling diffusion models to infer plausible structures

from ambiguous nadir input. This work demonstrates a scalable path to enriching nadir-based forest inventories with synthesized

structural detail, reducing the need for resource-intensive ground surveys.

Metric monocular depth estimation has become increasingly important and is often used as a redundancy mechanism in autonom

ous driving, where accurate scene understanding is essential for safe decision-making. In this work, we evaluate three recently

proposed models that represent the state-of-the-art (Depth Anything, PackNet-SfM, and UnidDepth) using zero-shot testing on the

DrivingStereo dataset across diverse weather conditions, and benchmark their performance. Our analysis considers not only metric

depth accuracy metrcis but also each model’s ability to generalize under challenging environmental variations. While UniDepth

achieves notable improvements over Depth Anything and PackNet-SfM, our results show that substantial progress is still needed for

robust real-world deployment. To further assess its practical suitability for autonomous driving applications, we conduct a detailed

examination of UniDepth’s strengths, limitations, and failure modes.

We present the first Open-Set Recognition (OSR) workflow for environmental monitoring for Permanent Laser Scanning (PLS) setups, using a Deep Neural Network (DNN) solely trained on simulated data. Such monitoring systems were previously only trained with real-world data and under the closed-set assumption, because they are commonly designed to observe a specific and predefined phenomenon (e.g., beach erosion, rockfall activity, vegetation change, animal behavior). The use of real-world data requires manual labeling, which is tedious given the great amount of point clouds. For this reason, we use Virtual Laser Scanning of Dynamic Scenes (VLS-4D) in a PLS setup to investigate how knowledge from synthetic data can be applied to real-world PLS monitoring systems in open-set settings. We introduce a novel framework that enables Open-Set Recognition (OSR) for animal monitoring (e.g. honey bees) using PLS data. The DNN is fine-tuned exclusively on a simulated LiDAR point cloud time series of flying honey bees, and integrates OSR to handle unknown classes during real-world deployment (e.g., butterflies, leaves, wren, and hare). By leveraging deviations in feature embeddings of the DNN, our method reliably distinguishes the known honey bee class from previously unseen classes, supporting robust monitoring under persistent distribution shifts. This approach reduces the dependence on extensive manual annotation of real-world point clouds, while maintaining reliable classification performance. It also highlights the potential of synthetic training data and OSR for environmental monitoring with PLS systems.

This work proposes a semantic-enhanced dynamic spatiotemporal model that integrates temporal attention, dynamic graph learning, and semantic module to better capture the complexity of human mobility. By combining dynamic adjacency learning with geographic and semantic structures, the model identifies both physical and functional relationships between zones. Results on TELUS mobility data demonstrate that semantic-enhanced graph construction improves prediction accuracy and robustness, offering a more meaningful representation of urban mobility dynamics and providing a strong foundation for future mobility forecasting and city-scale analytics.

Urban quality of life (QoL) assessments often rely on objective spatial indicators such as infrastructure access, land use, and environmental conditions. However, these metrics may overlook how residents subjectively perceive their surroundings. This disconnect reflects a methodological gap in urban studies: the lack of frameworks that integrate both objective and perceptual aspects of urban quality. In response, this study introduces a Perception-Based Urban Quality of Life Index (PUQLI) derived from street view imagery and deep learning and compares it with a composite objective indicator built from 13 spatially measured indicators across seven QoL domains. Rather than replacing conventional QoL assessments, PUQLI is intended to capture the visual-perceptual or experiential dimension of urban quality as inferred from street-level imagery. Each indicator was normalized and spatially joined to a hexagonal grid system. Pearson correlation revealed only modest associations between PUQLI and the objective indicators, indicating that subjective and objective urban quality are related but not equivalent. A mismatch index was then computed to quantify perception–provision gaps, revealing statistically significant and spatially patterned divergences (t = –10.535, p < 0.0001). Positive mismatch clustered in mixed-use urban centers, whereas negative mismatch aligned with documented environmental and infrastructural stressors; together with the significantly negative mean mismatch, this indicates a structural perception–provision gap in which measurable provision does not always translate into favorable lived experience. These findings highlight the need to integrate subjective perception into urban quality assessment and position the mismatch index as a practical diagnostic tool for perception-informed urban planning.

This research develops a pedestrian behavior model that incorporates the existence and dynamics of pedestrian groups. Using high-precision laser sensor data collected in the atrium of a hospital, the research first defines spatiotemporal parameters representing interpersonal distance, relative speed, and walking direction between pedestrians. Based on these parameters, machine learning techniques, including Support Vector Machine (SVM) and Random Forest algorithms, were employed. The SVM demonstrated superior accuracy and stability, successfully identifying groups even under complex walking conditions. Building on these results, the pedestrian behavior model described by psychological stress factors, such as stress from other pedestrians, obstacles, and group dispersion, is improved to account for the behavior of pedestrian groups. Model parameters were calibrated using laser sensor trajectory data with individual attributes (sex, staff, mobility aid usage). The proposed model accurately reproduced observed walking trajectories, with errors within 80 cm for approximately 80% of pedestrians. Finally, the model was applied to evaluate pedestrian spaces by mapping spatial distributions of psychological stress. Pedestrian stress was highest around reception areas, while group dispersion stress was greater in low-density zones where groups tend to spread out. These findings demonstrate that incorporating group behavior enhances the realism and applicability of pedestrian models for evaluating and designing public spaces. Future work will focus on applying the model to diverse facilities and pedestrian environments.

This study presents a workflow that links roadside LiDAR sensing with the modelling of pedestrian crossing behavior. Using self-collected LiDAR data from an informal mid-block crossing in Munich, the workflow includes object detection, tracking, trajectory reconstruction, event extraction, and contextual feature engineering. Behaviour-based yielding and stepping-out moments are used to identify pedestrian decision moments, which are subsequently labelled as gap-accepted or gap-rejected according to gap-acceptance theory. For each decision moment, features describing pedestrian state, social context, and vehicle context are extracted from the reconstructed trajectories. A logistic regression classifier is applied as an interpretable baseline to estimate gap-acceptance decisions under varying traffic conditions. The preliminary results indicate satisfactory predictive performance and show intuitive coefficient patterns, highlighting the influence of vehicle time gaps, pedestrian standing position, and peer presence. Overall, the study demonstrates the effectiveness of LiDAR-derived trajectories as a behavioral sensing foundation for modelling pedestrian crossing decisions.

This study proposes a ring-based SpatialTransformer to learn how building uses at different distances from a railway station interact to generate pedestrian flow. Concentric ring buffers at 100-meter intervals up to 800 meters were defined around 100 randomly selected stations in Tokyo, treating each ring as a spatial token. Self-Attention was applied to learn inter-zone interactions directly from data, without prior structural assumptions. GPS-derived walking trip counts served as the target variable and Geographically Weighted Regression as the baseline. Across 30 independent trials, the SpatialTransformer consistently outperformed GWR in predictive accuracy. SHAP analysis revealed that mid-to-outer distance zone features dominate pedestrian flow prediction, while features from the 0-100m zone contributed little. The attention matrix showed that each distance zone attends most strongly to spatially distant zones, demonstrating that pedestrian flow is regulated by structural interactions across the entire catchment area rather than by any single zone in isolation. These findings challenge the compact city assumption that station-proximate development maximizes pedestrian flow, and suggest that land use distribution across the full walkable catchment area deserves greater consideration in urban planning practice.

Urban accessibility is a key factor in understanding spatial inequality, as it conditions residents’ ability to reach essential services and urban opportunities. This study analyses accessibility in the intermediate city of Los Ángeles, Chile, characterized by a centralized concentration of services and expanding peripheral residential areas. Accessibility to educational, healthcare, and commercial facilities was evaluated using approximate travel times generated through the TravelTime API, considering walking, public transport, and private vehicle modes. Travel times were calculated from the centroids of residential census blocks, and opportunity-based accessibility was assessed using travel-time thresholds to identify the range of accessible commercial establishments.The results reveal marked spatial disparities. Central areas exhibit the highest levels of accessibility due to the density and diversity of amenities, with walking emerging as the most efficient mode for short distances. In contrast, peripheral neighbourhoods show limited access to healthcare and educational facilities and depend largely on private vehicles to reach central services, despite having higher population densities. Commercial accessibility in these areas is primarily restricted to small-scale neighbourhood establishments. These findings indicate that accessibility is influenced not only by travel time and transport networks but also by the spatial distribution and variety of urban functions. The study highlights the usefulness of routing APIs as an alternative methodological tool for accessibility analysis in contexts where official mobility data are outdated or incomplete, offering valuable insights for urban planning and policies aimed at reducing spatial inequalities.

Rapid urbanization is accelerating the fragmentation of blue–green spaces and the degradation of ecosystem services, while widening inequalities in environmental exposure and access to ecological benefits. Taking the “Geo-information Town” as a case study, this paper develops an integrated 3D framework linking urban form, human behavior and spatial interactions. First, UAV oblique images are semantically segmented to identify blue–green–grey features and to jointly assess and filter image quality. Second, multi-source spatial data, including Gaode POIs, nighttime lights, urban land use, OSM road networks, vector base maps and Baidu heat maps, are used to characterize urban functions and vitality patterns related to catering, sightseeing, shopping and cultural–educational services. Third, social media check-in data from Xiaohongshu and Weibo are incorporated to capture residents’ subjective evaluations and place preferences for different spatial units. An XGBoost–SHAP modelling framework is employed to quantify the relationships between these subjective evaluations and blue–green–grey indicators, and to interpret the marginal contributions of different environmental and functional attributes. The results reveal how perceived landscape qualities and service functions jointly shape spatial attractiveness and human–landscape interactions at the neighborhood scale. Finally, we discuss future research on 3D indicator systems, fine semantic segmentation of blue–green spaces, multi-source big data fusion and perception–behavior–function coupling, providing methodological support for perception-oriented assessment of residential environmental quality and optimization of blue–green urban landscapes.

Active Mobility Accessibility Index (AMAI) quantifies the competitiveness of walking and cycling relative to driving using travel-time and distance ratios on identical sampled origin-destination pairs, reflecting network structure rather than destination choice. AMAI combines time parity and distance parity in a simple diagnostic score, using equal weights as a default specification for interpretation and policy use. Applied across the five Tyne and Wear local authorities, it demonstrates that cycling is more competitive than walking against driving. The median origin-level cycling AMAI is 0.820 and the median walking AMAI is 0.645. Parity remains limited where the share of origins at or above parity is 10.0% for cycling and 1.7% for walking. Initial API-based tests suggested that time-of-day effects are limited for the short local trips studied here, supporting development of a scalable in-house routing workflow for the main analysis. Validation against OA-level Census 2021 mode shares, with controls for terrain gradient and commute-distance composition, suggests that AMAI captures a relevant behavioural signal, while its main value lies in diagnosing local network competitiveness for policy and planning.

Understanding pedestrian behaviors is the foundation of simulation for space planning. However, conventional behavior modeling methods are insufficient for learning detailed interactions, and deep learning methods often lack interpretability. This study aims to develop a pedestrian trajectory modeling approach based on discovering causal relationships among pedestrians. The proposed method consists of two parts: analyzing causal relationships among pedestrians using statistical causal discovery methods and predicting trajectories using attention-based deep learning methods. The first part employs a semi-parametric method to identify the causal relationships underlying observed pedestrian behavior and construct a spatial-temporal graph based on these causal relationships. The second part primarily uses the graph attention network to learn interactions among pedestrians. The experimental results demonstrate that the proposed method achieves a good balance between prediction accuracy and interpretability, while also identifying limitations, including at low-density scenes and due to causal model assumptions.

Floods are natural hazards triggered by intense rainfall and are particularly destructive in low-lying areas such as floodplains. In flood-prone regions, effective disaster management relies on prevention, monitoring, and emergency response strategies. In this context, remote sensing, especially Synthetic Aperture Radar (SAR), has become indispensable for flood mapping and monitoring due to its ability to acquire data under adverse weather conditions and persistent cloud cover. Multi-temporal SAR imagery processed into RGB composites allows rapid visualization of inundation patterns, while the Geographic Object-Based Image Analysis (GEOBIA) approach improves the classification of flooded areas through the integration of backscatter thresholds and terrain elevation data. This study investigates the spatial and temporal dynamics of flooding in the Lower Tubarão River Sub-basin (LTRSb), southern Brazil, following an extreme precipitation event that produced 260 mm of accumulated rainfall between May 24 and 25, 2019. The Sentinel-1B SAR images, acquired pre- and post-event, were used to map flooded areas with an overall classification accuracy of 88%. The results indicate that three days after the event, flooding covered 140 km² (29%) of the LTRSb, predominantly affecting agricultural (86.3 km²) and pasture areas (47.6 km²). The flooded extent decreased to 62 km² after 15 days and to 15 km² after two months, with agricultural land consistently accounting for 97% of the flooded area. Urbanized areas (≈1 km²) were also impacted, indicating significant risks to infrastructure and public health. These findings highlight the importance of SAR-based flood monitoring for risk assessment and disaster management in hydrographic basins.

Post-seismic deformation reflects the crustal adjustment to stress perturbations induced by earthquakes and may lead to modifications of ongoing deformation patterns. While such effects are well documented for moderate-to-large events, their detectability and significance after low-magnitude earthquakes remain poorly understood. Here, possible deformation pattern modifications associated with the February 2022 ML 3.5 Brentonico earthquake in Northearn Italy are investigated using ground-deformation time series derived from the European Ground Motion Service (EGMS) Ortho products. The earthquake occurrence is treated as a temporal discontinuity, enabling the estimation of pre- and post-event vertical/horizontal differential velocities during 2019-2023. A dual-weighted interpolation scheme, accounting for both spatial proximity and time-series reliability, is applied to derive spatially coherent maps of kinematic modifications. The results reveal measurable and spatially organized changes in deformation patterns, including localized accelerations/decelerations and direction reversals. A clear spatial correspondence between differential velocity anomalies and mapped fault systems suggests that the earthquake acted as a localized stress perturbation, modulating pre-existing tectonic structures. This study demonstrates the capability of EGMS datasets to capture post-seismic deformation signals and highlights the importance of considering low-magnitude events in long-term deformation analyses.

Underground transportation networks are essential for mobility in densely populated cities, addressing urbanization challenges such as traffic congestion, noise, and air pollution. Ensuring the safety and reliability of these infrastructures requires structural health monitoring (SHM) systems, which detect faults, deterioration, and damage. While traditional in-situ monitoring techniques provide real-time data, they are often economically restrictive. Synthetic Aperture Radar Interferometry (InSAR) offers advantages for large-area, long-term monitoring and has been successfully applied to various infrastructures, including dams, bridges, highways, and subways.

This study investigates surface displacement along a 15.4 km metro line with 11 stations in the Gebze district of Kocaeli, Türkiye, using multi-temporal InSAR. Sentinel-1 SLC IW data acquired between January 2019 and October 2025 were processed using MiaplPy, generating 504 interferograms with a 6-day temporal baseline. The phase-linking workflow utilized Persistent Scatterers (PS), Distributed Scatterers (DS), and Statistically Homogeneous Pixels (SHP), combined with Combined Phase Linking (CPL) algorithm and SNAPHU for phase unwrapping, to obtain reliable displacement time series and mean deformation velocities.

Results indicate line-of-sight displacements ranging from –10 to 10 mm/year, with the highest movements near the first station. Time series analysis shows stability from 2019 to 2021, a sudden displacement from 2021 to mid-2022, and stabilization until 2025. Monitoring these deformations provides insights into construction-induced dynamics and enables early detection of potential risks. Incorporating additional data, such as lithological, soil, and geotechnical information, can enhance data-driven monitoring. Long-term deformation monitoring ultimately supports the development of sustainable urbanization strategies and contributes to safe, resilient underground infrastructure management.

While polarimetric interferometric SAR techniques provide a strong theoretical framework for forest height retrieval, their application using C-band Sentinel-1 data is challenging due to repeat-pass acquisition geometry and strong temporal decorrelation. In this study, we assemble a globally distributed dataset combining Sentinel-1 interferometric observations with spaceborne LiDAR forest height measurements from the GEDI and ICESat-2 missions. More than 1800 Sentinel-1 interferometric image pairs were processed and spatially matched with LiDAR observations across tropical, temperate, and boreal forest regions. Sentinel-1 Single Look Complex data were used to derive interferometric coherence and polarimetric–interferometric observables, enabling statistical analysis of their relationship with forest structural properties. The results reveal physical relationships between Sentinel-1 coherence and canopy height across multiple forest biomes, indicating that Sentinel-1 interferometric measurements, under near-zero spatial baseline conditions, retain measurable sensitivity to vegetation structure despite temporal decorrelation effects. These findings provide a conceptual basis for exploiting similar repeat-pass interferometric observations from new low-frequency SAR missions such as NISAR and and upcoming ROSE-L for forest height mapping. In addition, the assembled dataset provides a global benchmark for developing and evaluating data-driven approaches for forest height estimation using Sentinel-1 observation.

This contribution presents an integrated monitoring framework that combines Sentinel-1 multi-temporal InSAR (MTInSAR) with geological, hydrological, and operational datasets to analyse long-term ground deformation in the Epe cavern storage field, Germany. Using the SBAS approach, vertical and horizontal deformation components are derived for the 30 km² storage area, revealing a bowl-shaped subsidence feature and distinct east–west deformation patterns associated with cyclic cavern operation. The analysis incorporates geoscientific information from GeoBasis NRW, AGSI+, ELWAS, and BÜK/BK50 into a harmonized GIS environment. This enables correlation of SAR-derived deformation fields with cavern-pressure cycles, soil characteristics, and groundwater variations, providing process-oriented interpretation of subsurface–surface interactions. Deformation results will be updated to include data through October 2025. The study demonstrates how MTInSAR, combined with geoscientific knowledge, supports transparent deformation assessment and contributes to the development of geospatial digital twins for subsurface infrastructure, improving operational resilience and environmental risk management.

Accurate land-cover mapping in heterogeneous metropolitan regions requires robust methods capable of overcoming limitations of optical imagery, particularly under persistent cloud cover. This study investigates the potential of L-band ALOS-2 PALSAR-2 SAR data for operational land-cover classification over Istanbul, Türkiye, by integrating advanced machine-learning algorithms, feature-selection strategies, and hyperparameter optimization. Four classifiers Random Forest (RF), XGBoost, LightGBM, and a shallow Artificial Neural Network (ANN) were evaluated using full-polarization SAR observations and a reference dataset derived from Sentinel-2 composites, orthophotos, and LPIS parcel boundaries.

Pre-processing included radiometric calibration, Lee filtering, terrain correction, and extraction of GLCM texture metrics from HH and HV channels, yielding an initial 20-feature set reduced to 18 through correlation and variance filtering. A LightGBM-driven Recursive Feature Elimination (RFE) procedure identified an optimal subset of ten features. Model optimization employed Bayesian hyperparameter tuning (TPE) under stratified 5-fold cross-validation to ensure reproducibility and generalization.

Results show that LightGBM achieved the highest accuracy (OA = 85.1%, κ = 0.81), followed by XGBoost (83.6%), RF (81.4%), and ANN (78.9%). Water surfaces were consistently the most accurately classified class, while confusion primarily occurred between urban and bare surfaces. Hyperparameter tuning improved F1-scores across all models, and reducing the feature stack to ten variables enhanced performance without loss of class separability. HV-derived texture features, particularly Entropy and Contrast, provided the highest discriminative power.

The study demonstrates that optimized feature selection and systematic hyperparameter tuning significantly enhance SAR-based land-cover classification, offering a transferable workflow for large-scale metropolitan mapping.

Mining induced deformation in underground metal mines poses a threat to surface infrastructure, underground access and environmental safety, and needs reliable spatially continuous monitoring. In this contribution, a long-term interferometric SAR analysis over an underground hard rock metal mines (Mine-B) in Khetri Copper Belt, India using TSX/TDX high resolution SAR data is carried out. Coherent small baseline DInSAR time series (CSB-DTS), stacking DInSAR, and single reference PSI chain are implemented. Stacking DInSAR derived average LOS deformation velocity and single-reference PSI derived velocity are obtained for mine-B for dataset of January 2023- December 2024. The obtained results verify that Mine-B is substantially stable while having a persistent fast and slow deformation concentrated inside and around the trough in SoZ-2 in Mine-B. The workflow shows how the combination of CSB-DTS, stacked DInSAR and PSI can facilitate the operational subsidence monitoring and long-term stability evaluation in complex mining environments and gives and gives an indication for the future integration with in-situ measurements and numerical models.

Spatially accurate fire danger information is critical for reliably predicting fire ignition probability, spread potential, and behaviour. However, Canadian fire management agencies mainly predict fire danger using weather stations, which only collect observations at explicit spatial points and cannot accurately model the fine-scale spatial variability of moisture across large and remote areas. This study predicts and maps the drought code, a variable representative of the moisture of deep, slow drying, compact organic matter across the landscape of British Columbia using ALOS-2 polarimetric SAR. A random forest model predicted the drought code of target areas with high accuracy to values derived from weather stations. The model was applied to forested areas across a time-series of ALOS-2 images on a grid-by-grid basis at a one square kilometer resolution and predicted the occurrence of fine-scale differences in drought code associated with differences in topography and elevation. The development of this drought code prediction model will allow fire management agencies to predict spatially accurate, fine-scale differences in drought code across the densely forested and highly mountainous landscape of British Columbia, improving fire behaviour and fire prediction systems.

Monitoring water levels in small and remote reservoirs is critical due to the climate crisis and rising water demands. Traditional in-situ gauge networks often provide sparse or inconsistent coverage, especially in remote regions. Satellite altimetry provides a global alternative, but it is frequently limited by long revisit times or coarse footprints unsuitable for smaller water bodies. Existing SAR-based methods face inherent limitations: amplitude-based approaches rely on accurate external Digital Elevation Models, whereas interferometric techniques are affected by coherence loss and phase-unwrapping ambiguities. To address these limitations, this research introduces a novel approach for estimating water level changes using SAR amplitude data without relying on prior morphological information. By modeling the coastal zone as a set of distinct planar slopes, the method relates the vertical water level change to the horizontal shoreline shift specific to each slope, observed as changes in the satellite range direction. The stack's standard deviation image is used to identify low-slope areas, where the horizontal response to water level variations is most pronounced. In these regions, area-based image matching is applied to quantify displacements within the coregistered stack. Finally, a least-squares estimation is used to determine temporal water level changes and local coastal slopes. The method was validated on Lake Trasimeno, Italy, using a stack of 30 Sentinel-1 images acquired in 2022. Comparisons with in-situ gauge data demonstrated high reliability, achieving an accuracy of 4 cm and a Normalised Median Absolute Deviation of 9 cm. The preliminary results are promising, while further experiments are currently underway.

This paper investigates wavelength-adaptive baseline design for spaceborne repeat-pass SAR tomography (TomoSAR) through a controlled simulation framework comparing representative X-band and C-band configurations. The study focuses on how radar wavelength influences the trade-off among vertical resolution, temporal decorrelation sensitivity, sidelobe behaviour, and baseline sampling efficiency. Using a discrete TomoSAR forward model, several experiments are conducted to analyse reconstruction performance under identical aperture, varying coherence conditions, different baseline sampling strategies, joint aperture-spacing design scans, and noise perturbations. Quantitative results show that X-band provides a clear intrinsic resolution advantage under coherent conditions, particularly for closely spaced scatterers, but this advantage degrades more rapidly under temporal decorrelation. C-band, while offering lower nominal resolution, exhibits more stable performance across coherence loss, wider design-space tolerance, and stronger robustness in noisy conditions. The comparison of uniform, minimum-redundancy, and irregular baseline patterns further indicates that baseline optimization is more critical for X-band than for C-band. The study moves beyond the general statement that “X-band is higher resolution whereas C-band is more robust” by providing experiment-based and frequency-dependent baseline design guidance. The findings support practical acquisition planning for future repeat-pass TomoSAR missions and contribute to a more quantitative understanding of wavelength-dependent sampling design.

In order to preserve their quality, SAR data are usually used in their native plane, the slant range. However, it is sometimes necessary to link ground data and radar data. Today, ground range or terrain-corrected data are frequently used for this purpose. An alternative to this approach is to project the data into slant geometry, which allows both the superimposition and co-registration of data from different sensors and the preservation of the resolution and phase of the SAR data.

An Observational Definition of the Absolute Phase in Radar Interferometry

The absolute phase in radar interferometry is required for topography and displacement estimation but lacks a general definition. The conventional definition states that absolute phase is proportional to the range difference between primary and secondary acquisitions. This definition is appropriate for simple targets such as point targets, but it cannot be directly applied to general targets.

Here, a universal observational definition of the absolute phase is proposed. It applies to any mode and does not require any assumptions about scattering mechanisms. For differential interferometry, the absolute phase is obtained by temporally unwrapping the phase as the intermediate secondary acquisition time progresses from the primary to the secondary acquisition time. This definition requires a continuous series of interferometric phase measurements along a pre-specified absolute path. The absolute phase matches the wrapped phase modulo 2π and agrees with the conventional definition for point targets.

This contribution discusses several implications of this general definition. The absolute phase of a complex target need not, and sometimes cannot, be proportional to the range difference. An example involving a permafrost landform demonstrates that the absolute phase following a cyclic change is nonzero and cannot be interpreted as a range difference. Another consequence is that the absolute phase in an interferogram can show 2π discontinuities, even when the interferogram itself is continuous and the coherence high.

This general definition enables thorough evaluation of InSAR processing chains and supports interpretation of observations.

Because of its high reproductivity, PolSAR is a suitable sensor for change detection in urban areas. Although many methods of change detection have been proposed, methods focused on polarimetric states transformation are rarely adapted. Through eigenvalue calculations, polarimetric basis which maximizes the polarimetric sensitivity can be calculated, and if this basis is fixed, quantitative change detection is available. The result from this method shows the obvious change in the target area, which is ‘Umekita project 2nd’ in Osaka city. However, changes outside the target area were also larger so that the change detection was not very effective from a relative viewpoint. To solve this problem, algorithms which surpass unnecessary changes in urban areas should be developed, and deeper understanding of scattering mechanisms in urban areas is needed.

The Amazon region has experienced increasing pressure due to the expansion of mining, especially illegal alluvial mining, driven by rising gold prices and a lack of economic opportunities. In Ecuador, this activity has contributed significantly to deforestation and the alteration of water systems, affecting river stability and water quality. The increase in suspended sediments and the modification of river channels generate ecological, economic and social impacts, including production losses and increased vulnerability of riverside communities. In this context, monitoring through remote sensing and Geographic Information Systems has become an essential tool for assessing river dynamics and the effects of illegal mining in the Amazon biome. This study analyses changes in land use and land cover in the Nangaritza River, considering an intense rainfall event that occurred in 2025. Cloud-free mosaics were generated using Sentinel-2 images, spectral indices were calculated, and supervised classification using Random Forest was applied to establish seven coverage categories. The results show a notable expansion of mining areas and sand deposits, accompanied by a reduction in forest cover. The transition matrix revealed significant losses of forest transformed into mining soil and turbid water, as well as an increase in sedimented areas downstream. The analysis of river dynamics identified five critical areas of mining expansion associated with increased sedimentation, turbidity, and morphological alterations to the riverbed. These changes reflect the growing anthropogenic pressure on the river and the need to strengthen monitoring systems to mitigate environmental impacts.

This study investigates the rapid transformations in land use and land cover (LULC) within Lahore District, a phenomenon with profound implications for ecological sustainability and land-use governance. Analysing these dynamics is crucial for minimizing adverse environmental impacts and promoting sustainable urban development. The primary objective is to assess historical LULC patterns over 30 years (1994–2024) and to simulate potential changes for the years 2034 and 2044 using an integrated CA-Markov modeling approach supported by GIS techniques. Landsat imagery from multiple sensors (TM and OLI) was processed through supervised classification methods, achieving classification accuracies exceeding 90%. The temporal analysis revealed marked changes, notably a substantial increase in built-up areas by 359.8 km², alongside reductions in vegetation cover (198.7 km²) and barren land (158.5 km²). Water bodies exhibited minimal variation throughout the study period. Future LULC scenarios generated via the CA-Markov hybrid model demonstrated strong predictive performance, as evidenced by a kappa coefficient of 0.92. The projections indicate continued urban Expansion primarily at the expense of green and undeveloped areas. These findings emphasize the pressing need for sustainable land management practices and provide a robust decision-support framework for urban planners. By integrating predictive modeling into planning policies, this research helps align developmental objectives with environmental conservation in rapidly urbanizing regions like Lahore.

This study evaluates the residential expansion of Guayaquil between 2016–2020 and 2020–2024 using Sentinel-2 imagery and supervised classification with the Support Vector Machine (SVM) algorithm. Given the rapid land transformation in tropical coastal cities, the research applied the Built-up Area Extraction Index (BAEI) to enhance the detection of built-up surfaces and distinguish them from vegetation and bare soil. The integration of BAEI and SVM allowed the development of an accurate, replicable, and low-cost approach to monitor the city’s urban growth.

Three periods of Sentinel-2 images with low cloud coverage were processed, re-projected, classified, and validated. Classification included four thematic classes (residential, vegetation, bare soil, and water bodies) using 1,000 training samples distributed across the city. All classifications achieved over 85% overall accuracy and a Kappa index of 1.0, confirming the model’s robustness in heterogeneous urban environments.

Spatial analysis of land-use transitions revealed that residential growth is concentrated in peripheral sectors such as Ciudad Santiago, Mucho Lote 2, Mi Lote, Trinipuerto, and areas near Narcisa de Jesús Avenue. Results indicate a strong tendency toward contiguous expansion, forming residential corridors along major road networks and the Guayas River. However, dispersed peripheral nuclei highlight challenges for service provision and environmental sustainability.

Overall, the combination of Sentinel-2 imagery, BAEI, and SVM proved highly effective for detecting built-up areas in tropical contexts, offering a scalable methodology for monitoring urban expansion in Latin American cities.

Hedgerows, a traditional agroforestry practice, are declining in Europe, threatening biodiversity and climate control. To support high-quality agricultural carbon credit certification, a method for automatic hedge detection using Pleiades Neo 30cm satellite imagery was developed.

Two methodological approaches were tested in three French study areas with varied landscapes: (i) a classic image segmentation using NDVI, Green Cover Fraction, and LiDAR-derived Digital Height Model, and (ii) a foundation model retrained on 150 annotated tiles with diverse landscape and satellite acquisition configurations. The methods were compared using quantitative (Intersection Over Union, Omission & Commission errors) and qualitative indicators.

The foundation model demonstrated superior hedge detection and robustness across different landscapes. A ground truth dataset based on stratified random sampling and equal allocation was created to allow the quantification of its accuracy using standard accuracy metrics. It achieved a precision of 0.89 and a recall of 0.83 for the hedge class. It effectively adapted to the morphological and ecological diversity of hedges, with few commission errors primarily due to confusion with isolated trees or linear vegetation, and omissions mainly in discontinuous or degraded hedges.

The study confirms the relevance of Pleiades Neo for detecting thin-scale elements like hedges, the effectiveness of foundation models with limited reference data, and their potential for large-scale hedge mapping. Future work aims to incorporate more spectral bands and expand the model's training to detect hedgerows across the European Union under various satellite acquisition contexts, paving the way for operational tools in agricultural carbon credit valuation.

This study focuses on county-level areas with high-density wind farm distribution in the Xing'an League of Inner Mongolia, China. Using high-resolution satellite imagery from 2016 to 2024, land cover information within and around wind farms was extracted through visual interpretation, and the spatiotemporal dynamics of land cover in these areas were analyzed. The results indicate that: (1) From 2016 to 2024, land cover change in the study area was primarily driven by wind farm expansion, which increased cumulatively by 130.87 km² (+260.35%) and exhibited the highest dynamic degree among all land cover categories (LK = +32.54%/yr). (2) Grassland was the most severely impacted land cover type, with 78.74 km² converted to wind farm land, accounting for 59.66% of the total newly established wind farm area, while cultivated land and forest land contributed 20.06% and 18.33%, respectively. (3) As wind power expanded, the land cover composition within wind farms shifted from a cultivated land–grassland balance toward grassland dominance. (4) Areas subjected to temporary disturbance from wind farm construction activities tended to recover progressively, with cultivated land exhibiting a faster recovery rate than grassland.

Addressing the United Nations Sustainable Development Goals, particularly sustainable cities and communities (SDG 11), and

the protection of terrestrial ecosystems (SDG 15), is closely linked to understanding patterns of urbanization. Rapid urban growth

significantly influences ecosystem functions, including transportation, housing, and economic development. Monitoring this growth

and analyzing performance patterns are essential for supporting decision-making and guiding urban planning and management. This

study presents an automatic approach for monitoring urban expansion by applying the Random Forest machine learning classifier

from 2015 to 2025 using Google Earth Engine. The method exploits spectral indices not only for unsupervised classification but

also for training the Random Forest classifier, thereby ensuring a fully automated workflow. The proposed approach is applied

to Gharbia Governorate, a region which lacks surrounding desert margins and is instead entirely composed of fertile agricultural

land, to monitor urban expansion in three-year intervals. The proposed study, which achieved a kappa coefficient exceeding 0.96

across all study periods, revealed a gradual decline in agricultural land from 75.5% in 2015 to 72.7% in 2025. These outcomes offer

valuable insights to support evidence-based planning and promote sustainable land use management.

Cropland abandonment (CA) is a major land-use change with important environmental and socio-economic implications. This study evaluates cropland abandonment detection using multi-temporal Landsat features and a spatially independent validation framework, comparing the performance and spatial behaviour of Random Forest and XGBoost classifiers. A set of temporally aggregated spectral indices (NDVI, BSI, NDBI, and MNDWI), including multi-year trends and variability measures, was integrated into a 56-band composite dataset. Training and validation samples were generated using 100 × 100-pixel windows centred on land-use parcels, with overlapping areas between different reference classes explicitly excluded to avoid label ambiguity. To reduce spatial autocorrelation, the data were split into separated training (1,582.6 km²) and testing (719.2 km²) areas within the Savinjska statistical region in Slovenia. Random Forest (RF) and XGBoost (XGB) classifiers were trained and evaluated using spatially separated validation data. Classification performance was assessed using overall accuracy, user’s and producer’s accuracy, and F1-score. Results indicate that XGB achieved a higher overall accuracy (0.705) compared to RF (0.670) and exhibited strong sensitivity in detecting cropland abandonment, while RF produced more conservative and spatially stable estimates of abandoned cropland area. Spatial error maps and area-based comparisons reveal systematic differences between the two classifiers, particularly in their tendency to overestimate abandonment extent. The findings highlight the importance of spatially explicit validation strategies, careful reference data preparation, and multi-temporal feature design for robust cropland abandonment mapping. The main contribution lies in the systematic assessment of model behaviour, spatial error patterns, and area estimates under strict spatial separation of training and testing data.

Land Use/Land Cover (LULC) classification is essential for understanding the spatial distribution of Earth’s surface and for supporting sustainable environmental and economic development. Recent extreme events in Central Europe have emphasized the link between LULC change and disaster vulnerability, highlighting the need for accurate monitoring. Advances in satellite technologies, particularly Sentinel-1 and Sentinel-2, combined with deep learning (DL) methods, have significantly improved LULC mapping. Convolutional Neural Networks (CNNs) excel at spatial feature extraction, Long Short-Term Memory (LSTM) networks capture temporal dependencies, and Convolutional LSTM (ConvLSTM) models integrate both spatial and temporal information.

This study evaluates the comparative performance of DL architectures for LULC classification in the Harz Mountains, Central Germany—a region experiencing notable forest cover loss. We assess 2D-UNet using two temporal processing approaches, examine the effect of attention mechanisms in 3D-UNets, and explore multiple integrations of ConvLSTM layers within UNet architectures. Our goal is to identify the most effective strategy for capturing spatio-temporal dynamics in LULC datasets, contributing to improved monitoring and management of vulnerable landscapes.

Monitoring land-use and land-cover (LULC) dynamics in rapidly urbanizing regions is critical for sustainable environmental planning. Dynamic metropolitan areas with rapid urbanization, such as Istanbul, Türkiye, are experiencing significant land-cover changes, among which deforestation is one of the most critical. This study presents a Google Earth Engine (GEE)-based framework to monitor LULC changes in Istanbul from 2016 to 2025 by fusing Sentinel-2 optical imagery, Sentinel-1 SAR and topographic data. From these datasets, a feature set—including spectral bands, vegetation indices, SAR backscatter metrics, and topographic variables—was derived and used to train a Random Forest (RF) baseline model on 2016 Land Parcel Identification System (LPIS) reference data. The baseline model was then applied across the time series to assess its temporal transferability, overcoming the limitation of up-to-date ground-truth data. The baseline model achieved an overall accuracy of 72%, calculated using a validation dataset derived from the LPIS reference data. Feature importance analysis revealed that structural variables—particularly DEM and SAR metrics—were the primary contributors to the classification, used in combination with optical features. Time-series results indicate a cumulative decline of 231 km² in agriculture and 379 km² in forest cover during the study period, inversely corresponding to urban growth. The results of the study highlight that, although applying a single-year model without independent annual validation data causes certain uncertainties—arising from methods, sensors, or topography (e.g., misclassifications)—the proposed framework is highly practical for monitoring deforestation and urbanization trends in complex landscapes.

Accurate and efficient land-use mapping provides intuitive spatial information, which helps to rationalize the planning and deployment of land resources, and provides a basis for urban planning, agricultural development, environmental protection and other aspects. This study utilizes the Google Earth Engine platform and the Resnet-50 method to explore the spatial distribution of land use in Daxing District, Beijing in 2023, by combining point of interest (POI) data, nighttime light data, Sentinel-1 data, and Sentinel-2 data. The results of the study show that the accuracy of land use mapping using different data is different, and the accuracy of the Resnet-50 method is better than that of the Random Forest method. Making full use of the band features and index features of Sentienl-1 data and Sentinel-2 data, nighttime light data and POI data can improve the accuracy of land use mapping results. Among them, the land use mapping accuracy of the proposed method is the highest, with an OA of 88.11% and a Kappa coefficient of 0.83. Ranking the importance of different features found that VH band in January-March has the most important effect on the land use mapping results and the residential land in the POI data has the least important effect on the land use mapping results. This study provides a feasible reference program for efficiently and accurately obtaining land use mapping data for a large study area.

Reliable urban risk assessment requires accurate and up-to-date information on building characteristics, particularly construction year and height, which are often incomplete or unavailable in existing databases. This study presents a cloud-based methodology for the automatic estimation of these parameters using multispectral and very high-resolution Earth Observation (EO) data. The proposed approach integrates temporal analysis of multispectral satellite imagery (Sentinel-2 and Landsat) with photogrammetric processing of very high-resolution stereo imagery (Pléiades). Building construction year is estimated by detecting temporal changes in spectral indices using spline-based modeling and discrete-difference analysis, achieving an accuracy of better than ±3 years. Building height is derived from digital surface models generated from satellite stereo imagery, with a mean accuracy of less than 2 m relative to LiDAR reference data (~1.40 m). The methodology was implemented in a cloud computing environment (Google Earth Engine and Google Colab) and tested in the City of Zagreb, Croatia. Validation results show robust performance, with an F1-score of 0.819 for construction year estimation and strong agreement between EO-derived and LiDAR-based height values. The results demonstrate the potential of EO-based methods for scalable, reliable extraction of building information, thereby supporting improved urban risk assessment and decision-making.

Zhejiang Province is experiencing rapid land use/land cover (LULC) transitions driven by urban expansion, infrastructure development, and increasing environmental pressures. Understanding historical dynamics and future trajectories of these changes is essential for informed regional planning and ecological management. This study analyzes land use changes from 2000 to 2020 and forecasts future patterns for 2025 to 2040 by integrating multi-temporal land use data with key spatial drivers, including elevation, slope, aspect, Normalized Difference Vegetation Index (NDVI), and proximity to roads and built-up areas. Change detection results reveal substantial declines in croplands and green spaces alongside rapid urban expansion, particularly around Hangzhou and Shaoxing and along major transportation corridors, reflecting an early phase of accelerated urbanization from a relatively small baseline. Future land-use dynamics were simulated using a hybrid Convolutional Neural Network - Long Short Term Memory (CNN-LSTM)-Cellular Automata (CA)-Markov framework that captures complex spatiotemporal dependencies and neighbourhood interactions under physical and anthropogenic constraints. Model projections indicate a more moderate growth regime from 2025 to 2040, with urban land increasing by 1.7%, croplands decreasing by 2.2%, and modest gains in water bodies (1.9%) and forest cover (1.1%), suggesting landscape saturation and policy-influenced land management. Validation using the observed 2025 land use map demonstrates strong predictive performance, achieving an overall accuracy of 86% and a Kappa coefficient of 79%. The results provide spatially explicit insights to support balanced development and enhanced ecological resilience.

Land use/land cover change (LULCC) analysis is essential for understanding environmental transformation and guiding sustainable territorial planning. Remote sensing offers a valuable source of information for monitoring these changes, but the accuracy of thematic maps depends heavily on the classification algorithm applied. This study compares the performance of three widely used supervised Machine Learning (ML) algorithms (Random Forest (RF), Support Vector Machine (SVM), and Artificial Neural Networks (ANN)) for LULC mapping in the Greater Guayaquil region, a tropical area with persistent cloud cover.

A mosaic of Landsat-9 images from 2023 was processed in Google Earth Engine, followed by the selection of representative training and validation samples. The algorithms were implemented in R Studio, and accuracy was evaluated through confusion matrices and external comparison with MapBiomas Ecuador.

Four LULC classes were defined: Forest, Crops, Vegetation-free areas (urban/bare soil), and Water. Results indicate that SVM achieved the highest performance, with 93% overall accuracy and a Kappa coefficient of 0.91, followed by RF (92%; κ = 0.89) and ANN (90%; κ = 0.86). SVM also showed the highest spatial agreement with MapBiomas (>90%). Discrepancies were concentrated in rapidly changing urban–agricultural boundaries.

The superior performance of SVM is attributed to its capacity to model non-linear class boundaries in complex tropical landscapes. Despite expectations that RF would perform best based on previous literature, SVM proved more effective for this specific AOI. The study confirms that Landsat-9 combined with supervised ML models, particularly SVM, offers a robust and cost-effective approach for environmental monitoring and land-use planning in data-limited regions.

This study analyzed the advance of deforestation associated with artisanal mining in the Tapajós Environmental Protection Area (APA), north of Jacareacanga, Pará State, Brazil, for 2017 and 2024. The National Institute for Space Research (INPE) monitors deforestation in the Amazon using remote sensing, and the Tapajós APA stands out among protected areas for high rates of mining-related deforestation. Sentinel-2A images were used to generate the Normalized Difference Vegetation Index (NDVI) and RGB composites, assigning the 2017 NDVI to the red channel and the 2024 NDVI to the green and blue channels. Auxiliary data, including active mining processes in 2024 and the hydrographic network, were integrated for analysis. This approach enabled the identification of two distinct spectral responses: (i) cyan areas corresponding to regions that were non-vegetated in 2017 but exhibited regenerated vegetation in 2024; and (ii) red areas corresponding to regions that were non-vegetated in 2024. The results enabled visualization of the spatial progression of deforestation, particularly along drainage networks and in relation to active mining areas, revealing a pronounced expansion associated with artisanal mining across multiple waterways, with an upstream progression consistent with alluvial gold and cassiterite deposits. The data corroborate deforestation alerts issued by the Real-Time Deforestation Detection System (DETER) and the Deforestation Alert System (SAD)/Imazon, indicating the continuity of artisanal mining pressure in the Tapajós APA. The methodology demonstrated efficiency in detecting environmental changes and can be replicated in other areas under mining pressure, contributing to territorial monitoring and environmental management

Wildfire activity in Canada has increased significantly in recent decades, shifting to larger, more frequent fires and the emergence of megafires (>10,000 ha) across various ecozones. These events typically exhibit complex spatial patterns of burn severity, including larger and more homogeneous patches of high severity. The burn severity patterns and their drivers in megafires remain unclear, in particular, across diverse ecozones. Remote sensing indices such as the Relativized Burn Ratio (RBR) provide an effective means of quantifying burn severity at large spatial scales. This study uses RBR to evaluate nine megafires (each >50,000 ha) representing the 95th percentile and above of fire size within varying ecozones between 2016 and 2022. These fires were used to develop two random forest models: one predicting RBR and another predicting the within-fire z-score of RBR. Within-fire standardization of RBR was conducted to see whether it alters the relative importance of environmental drivers. In the RBR model (OOB R² = 0.75), regional variables such as ecozone and fire ID, along with drought code, were dominant predictors. In contrast, the z-score model (OOB R² = 0.68) emphasized fuel characteristics, including biomass and canopy closure, with additional contributions from elevation and drought-related variables. These results suggest that broad regional and fire-regime controls exert a stronger influence on burn severity than local fuel conditions at the megafire scale. Standardizing burn severity within fires reduces this regional signal but does not improve predictive performance, highlighting the importance of accounting for regional variability in large-fire dynamics.

Urban Heat Islands (UHIs) represent one of the most pervasive manifestations of human-induced modification of the land surface. They arise from the replacement of natural surfaces with impervious materials, reduced evapotranspiration, waste heat emissions, and altered aerodynamic roughness, collectively causing cities to exhibit elevated temperatures relative to surrounding rural areas.

The Local Climate Zone (LCZ) framework introduced by Stewart & Oke (2012) provides a standardized, physically based classification system for describing urban and natural landscape types according to their surface structure, cover, and thermal properties. Unlike traditional land-use/land-cover schemes, LCZs are explicitly designed for urban climate studies and allow for consistent comparison of urban form, function, and thermal behaviour across cities worldwide.

While LCZ maps of Bologna already exist within the WUDAPT protocol, they are characterized by a declared not high level of accuracy. So, the present work aims to produce the first detailed and reliable LCZ thematic map for the Municipality of Bologna, using higher quality, remotely sensed, input data.

To assess the impact of multi and hyperspectral imagery on the classification results, a Sentinel-2 and a PRISMA image were considered for the study.

Overall, this study provides for a first time a detailed and accurate LCZ map for the Municipality of Bologna and confirms the value of combining UCPs with both multispectral and hyperspectral data.

This study was carried out within the Space It Up project funded by the Italian Space Agency, ASI, and the Ministry of University and Research, MUR, under contract n. 2024-5-E.0 - CUP n. I53D24000060005.

This study focuses on predicting the evolution of internal urban spatial features, a dimension often overlooked in research that prioritizes urban expansion. Using the Yangtze River Delta as the case study, the work integrates multimodal remote sensing data—including high-resolution optical imagery and SAR data—to capture detailed land-use patterns, structural textures, and functional–transportation relationships. These fused datasets support the Futureland model, which applies parallel computing and Generalized Logistic Regression to simulate future spatial configurations with high efficiency and accuracy.

The results suggest that from 2030 to 2050, major cities such as Shanghai, Suzhou, Wuxi, Changzhou, Jiaxing, and Hangzhou will develop into more connected and compact urban clusters. Transportation networks and functional areas are expected to evolve in general alignment, while localized deviations reflect the complexity of internal urban dynamics. Land-use types are projected to undergo new spatial combinations and reorganizations, indicating improved continuity and diversity within the urban structure.

By systematically revealing trends in land-use evolution, transportation–function coupling, and urban form transformation, this research provides a clearer understanding of future urban spatial development. The proposed predictive framework offers valuable guidance for urban planning, governance, and sustainable regional development.

Global shifts in vegetation patterns highlight the need for effective monitoring as climate conditions intensify. Remote sensing provides valuable tools for detecting stress across landscapes. This study examines whether thermal satellite and airborne data can detect early stress linked to temperature, drought, and fire history. Within Australia, eucalypt species are increasingly vulnerable to climate-driven canopy dieback. Prolonged drought and extreme temperatures increase the risk of dieback. As eucalypts regulate leaf temperature through transpiration, when water is limited, leaf temperatures rise and can be detected using thermal imagery. Therefore, our research questions are:

• Can temperature-related stress patterns be identified?

• Does fire history affect long-term stress?

• Can thermal changes indicate dieback events?

• Does topography shape canopy stress?

Study Area:

Scott Creek Conservation Park in South Australia contains diverse native vegetation and steep terrain. The canopy is dominated by two stringybark eucalypt species, with areas burnt in the 2021 fire and unburnt controls.

Data and Methods:

Thermal satellite data from 2019–2025, including land surface temperature and water stress indicators, were analysed alongside local climate records. Airborne hyperspectral, LiDAR, RGB, and thermal imagery (50 cm) were processed to derive canopy structure, topography, and thermal patterns. A supervised classification was used to assess canopy condition.

Preliminary results:

Indicate that after a fire, high and moderate levels of vegetation stress increased and persisted into the following year. Vegetation in fire-affected areas showed no significant improvement in WUE during the recovery period. suggesting that fire-affected vegetation remained physiologically stressed despite visible regrowth.

The study investigates the long-term environmental and social impacts of asbestos mining in the Troodos Mountains of Cyprus, where chrysotile extraction between 1907 and 1988 left a lasting legacy of contamination and landscape degradation. Using multi-temporal aerial photographs, Sentinel-2 satellite data, and field observations, the research analyses land use transformations and vegetation recovery processes in the Amiandos mine area. A land use transfer matrix and Normalized Difference Vegetation Index (NDVI) were applied to assess ecological regeneration and detect spatial patterns of recovery.

To address ongoing environmental health risks, pre-trained deep learning models based on convolutional neural networks (CNNs) were used to identify asbestos-cement roofing in high-resolution aerial imagery. The results indicate measurable reforestation since the 1990s, but also reveal remaining asbestos waste deposits and deteriorated roofing materials posing persistent hazards to local communities.

The integration of remote sensing, vegetation indices, and deep learning methods provides a comprehensive approach to understanding environmental inequality in post-industrial landscapes. This framework supports the development of inclusive and data-driven restoration strategies consistent with the European Union’s environmental health goals. By combining spatial intelligence with machine learning, the study demonstrates the potential of remote sensing to monitor ecological recovery and mitigate asbestos-related risks in Cyprus and similar post-mining environments.

Urban heat and fine particulate matter (PM₂.₅) pollution are critical challenges for sustainable cities, but their high-resolution spatial and temporal patterns are not well understood. This study develops a multi-year 250 m downscaling framework to map the synergy between land surface temperature (LST) and PM₂.₅ in Wuhan, China. Using machine learning–based residual correction, annual, summer, and winter mean PM₂.₅ concentrations in 2015 and 2020 were downscaled from 1 km TAP data to 250 m grids. Correlation and spatial autocorrelation analyses were applied to reveal the spatial patterns of LST–PM₂.₅ interactions. The downscaled PM₂.₅ achieved high accuracy (R² > 0.80), and the heat–pollution relationship showed strong spatial heterogeneity. From 2015 to 2020, synergistic zones changed in the Urban area, consistent with the growth of impervious surfaces. These results provide a fine-scale spatial basis for understanding the coupled dynamics of urban heat and air pollution, supporting integrated strategies for climate adaptation and urban air quality management.

The coastal waters of Shanghai, situated at the confluence of the Yangtze River Estuary and the northern Hangzhou Bay, form a typical high-turbidity aquatic environment influenced by sediment discharge from the Yangtze River and strong tidal dynamics. Extensive urbanization and coastal development have led to the proliferation of hardened shoreline structures in this region, which have altered natural hydrodynamic conditions and sediment transport patterns, contributing to ecological issues such as wetland degradation. In recent years, Shanghai has initiated ecological restoration projects aimed at rehabilitating healthy coastal ecosystems. These restoration efforts, involving geomorphic reshaping, may directly disturb and modify sedimentary environments. However, their impact on suspended sediment concentration (SSC)—a key environmental parameter—across large spatiotemporal scales remains unclear. Traditional in-situ monitoring methods are inadequate for capturing such large-scale dynamic variations, whereas satellite remote sensing provides an effective alternative. This study utilizes multi-source remote sensing data to develop an inversion model for SSC suitable for Shanghai's coastal waters, systematically analyzing the influence of different shoreline types on sediment distribution. The findings illustrate how ecological restructuring of shorelines affects the spatial and temporal variations of SSC, thereby providing a scientific basis for optimizing coastal management strategies and assessing the effectiveness of ecological restoration efforts.

This study investigates the critical role of Urban Morphological Blocks (UMBs) in shaping Land Surface Temperature (LST) patterns across seasons and cities. Through a comparative analysis of Beijing, Wuhan, and Fuzhou, China, we integrated multi-source remote sensing and 3D building data to define UMBs based on building height and density. Employing robust statistical models (Geographical Detector and Random Forest Regression), we quantified the driving forces behind LST variations. Our results consistently identified Low-Rise, High-Density blocks as the primary heat contributors, while High-Rise blocks exhibited cooling effects. Crucially, we found a strong seasonality in dominant drivers: surface biophysical parameters (e.g., vegetation, impervious surfaces) governed LST in warm seasons, whereas 3D architectural morphology (especially building height) became paramount in winter. Furthermore, factor interactions revealed synergistic effects, with the combination of block type and vegetation yielding the highest explanatory power. These findings underscore the UMB as a vital unit for urban climate analysis. The study provides actionable insights for planners, recommending targeted mitigation in high-risk blocks, promotion of thermally efficient building forms, and the adoption of season-specific strategies to enhance urban resilience against heat stress.

Accurate, long-term precipitation data is essential for reliable drought monitoring. This study addresses heterogeneous uncertainties in existing precipitation datasets across mainland China by developing two modifiable weighting schemes: a Cheng-Kling-Gupta Efficiency weighted-ensemble model (CWEM) and Bayesian Model Averaging (BMA). These methods were used to merge seven monthly precipitation datasets into new weighted products (BMAEP and CWEP). The precision and drought monitoring utility of these fused products were evaluated against the benchmark Multi-Source Weighted-Ensemble Precipitation (MSWEP) product using gauge data. Results show that the new weighted schemes outperform individual datasets and MSWEP. Specifically, BMAEP-2P achieved a superior composite CKGE index of 0.828, CWEP-4P attained a higher correlation coefficient (CC) of 0.905, and CWEP-2P excelled in relative bias (0.579%) and root mean square error (20.755 mm). Furthermore, BMAEP or CWEP performed optimally for drought monitoring across all sub-regions of China at multiple time scales (1-24 months), with average highest CC and probability of detection values reaching 0.919 and 0.844, respectively. Contribution analysis identified CPC as the dominant factor enhancing model performance. The study demonstrates that CWEM and BMA methods effectively generate superior precipitation datasets for drought monitoring applications.

Fine particulate matter (PM₂.₅) poses serious risks to environmental quality, human health, and sustainable development. However, existing studies seldom achieve long-term, pixel-level PM₂.₅ projections or comprehensively evaluate scenario-based predictions under different development pathways. This study proposes an integrated method for projecting PM₂.₅ distribution at a pixel-level under multiple scenarios by incorporating land-use simulations, land surface indices, and spatial dependence effects. Using multi-temporal Landsat series, ground-based PM₂.₅ observations, and socio-economic data, we generated land-use projections under the Shared Socioeconomic Pathways (SSPs) using the Futureland model. Corresponding land surface indices (NDVI, NDBI, NDWI) were derived and used within a spatial lag model to predict PM₂.₅ concentrations for the Yangtze River Delta (YRD) from 2010 to 2030. Results indicate that under SSP1, characterized by sustainability, forest area and NDVI increase, leading to a significant decline in PM₂.₅ levels. Conversely, under the fossil-fueled SSP5 scenario, urban expansion drives up NDBI and PM₂.₅ concentrations. These findings demonstrate that increased green space and reduced fossil fuel reliance are crucial for improving air quality. The proposed method provides decision-makers with actionable insights for policy formulation and highlights the environmental importance of land-use planning. Future work will integrate dynamic meteorological models and conduct uncertainty assessments. This scenario-based projection framework can be applied to support sustainable urban and environmental management in rapidly developing regions.

The upper troposphere and lower stratosphere (UTLS) plays a vital role in the global climate system. Central China, particularly Hubei Province, is located directly downstream of the Tibetan Plateau, making it a key region for observing long-range aerosol transport. In this study, the vertical structure and transport mechanisms of UTLS aerosols over Hubei are investigated using a multi-year (2016-2018) satellite dataset. We applied rigorous Cloud-Aerosol Discrimination (CAD) scores and quality control procedures to nighttime CALIOP profiles to minimize cirrus cloud contamination. Our results show that the UTLS background over this region exhibits low aerosol loading during non-monsoon seasons. However, a substantial aerosol enhancement occurs in summer. The monthly mean extinction coefficient at 14-16 km reaches a peak of 8.05 × 10-3 km-1 in August, with a particulate depolarization ratio of ~0.25, indicating the presence of non-spherical particles related to the Asian Tropopause Aerosol Layer (ATAL). To investigate the physical drivers of this seasonal variation, we integrated MERRA-2 meteorological fields and HYSPLIT backward trajectories. The analysis reveals a dual transport mechanism: First, intense local deep convection driven by the East Asian Summer Monsoon (EASM) vertically pumps boundary-layer pollutants into the upper atmosphere. Second, the Asian Summer Monsoon Anticyclone (ASMA) and the westerly jet stream advect aged aerosols horizontally from the Tibetan Plateau to Central China. These findings provide direct observational evidence of how regional monsoon systems synergistically modulate stratospheric aerosol loading.

All living organisms significantly impact air quality, which is vital for the Earth's ecosystems. Air pollution has increased in the Indian subcontinent, mainly due to harmful gases and particles from industrialization and urban development. CNN Health reported that as of February 25, 2020, 21 out of the 30 cities with the worst air quality globally are in this region, with six in the top ten. Urgent research and pollution control measures are needed, especially in urban areas where human health and the environment are most affected.

The Air Quality Index (AQI) measures pollution levels from key pollutants like PM10, PM2.5, ozone, sulfur dioxide, and others, using a scale from 0 to 500. While pollution control boards collect ground data, the limited number of sensors in large cities can be a challenge. Satellite imagery enhances coverage, although in situ data remains essential in many areas.

This research aims to connect satellite remote sensing with air quality monitoring by determining air quality indices for pollutants in Dobson units. In situ sensors measure concentrations in micrograms per cubic meter, while satellites use molecules per square meter (Dobson units). Different techniques, like regression analysis, are used to develop location-specific indices for urban and suburban areas.

The focus of this research is on methodology rather than final conclusions, highlighting the importance of accurate real-world representations through reliable data and atmospheric models.

Air pollution remains a major environmental challenge in Kathmandu District, driven by rapid urbanization, increasing emissions, and meteorological influences. This study examines the spatial and temporal variability of PM₂.₅ and PM₁₀ from 2019 to 2024 by integrating satellite-derived Aerosol Optical Depth (AOD), ground-based measurements, and advanced statistical and machine learning techniques. Two regression approaches—a simple linear model using AOD and a multivariate model incorporating temperature, relative humidity, wind speed, wind direction, and planetary boundary layer height (BLH)—were evaluated using R² and RMSE metrics. The multivariate model consistently outperformed the simple linear regression, demonstrating improved predictive capability and was validated using PM data from the US Embassy monitoring station at Phora Durbar. Seasonal analysis showed pronounced pollution peaks in winter, with PM₂.₅ levels ranging from approximately 165–167 µg/m³, while summer exhibited the lowest concentrations (~51 µg/m³). PM₁₀ showed moderate seasonal variability with a notable decline during spring. The study also identified the influence of wildfire events and meteorological conditions on episodic pollution spikes. Despite limitations related to satellite resolution and uneven ground monitoring coverage, the integration of remote sensing, meteorological parameters, and machine learning proved effective for estimating particulate matter concentrations. Overall, the results highlight distinct seasonal pollution patterns and underscore the value of combined observational and modeling approaches for improving air quality assessment in Kathmandu District.

This study aims to systematically investigate the influence of spatial extent and spatial resolution on the estimation of soybean LAI and AGB and model transferability during the peak of the growing season. The research objectives are to: 1) assess and compare the predictive performance of Stepwise Multiple Linear Regression (SMLR) and Random Forest (RF) models for estimating LAI and AGB across different spatial extents and spatial resolutions; 2) evaluate the transferability of these models across spatial extents and resolutions to determine their robustness under varying scale conditions. Our results demonstrate that RF model outperformed SMLR and presented the highest LAI estimation accuracies across the three nested spatial extents with RMSE of 0.52m2/m2, 0.33m2/m2, and 0.31m2/m2, respectively, explaining 86%, 91%, and 91% of LAI variability at 1m2, 25m2, and 100m2 extents, respectively. Similarly, the RF model had the overall best accuracies with RMSE of 67.13g/m2, 76.98g/m2, and 58.03g/m2, respectively, explaining 83%, 86%, and 84% of soybean AGB variability at 1m2, 25m2, and 100m2 extents, respectively. Moreover, the results showed that the accuracies of both models increased for both LAI and AGB estimation at larger scales. We found that RF models outperformed SMLR in estimating soybean LAI and AGB at 3m resolution (LAI: R2=0.86, RMSE=0.39m2/m2, rRMSE=6.19%; AGB: R2=0.82, RMSE=59.09g/m2, rRMSE=18.09%) and 10m resolution (LAI: R2=0.92, RMSE=0.28m2/m2, rRMSE=4.36%; AGB: R2=0.80, RMSE=59.95 g/m2, rRMSE=18.35%), respectively. Further, the transferability of RF models showed weaker performance when applied to estimate soybean LAI and AGB at higher (or smaller) spatial extents and coarser (or finer) image resolutions.

This study looks to use fine resolution Synthetic Aperture Radar (SAR) for crop classification of small scale fields. The study site is the University of Guelph's Elora Research Station and looks to conduct crop classification with a random forest on four plots divided into 28 fields of 7 x 14 m in size. Four crop types are planted in each field which include, alfalfa, corn, soybeans, and winter wheat. The datasets used for the analysis are 4 SAR scenes taken during the May to July growing season with two of the plots used as training sets, and the other two as testing. The dataset is provided by the Finnish microsatellite company, ICEYE, with the data products being 0.5-meter resolution VV images. Additional textural information known as Grey Level Co-occurrence Matrix (GLCM) are processed from the SAR scenes and added to the random forest. The analysis was conduced at the pixel level and a 70-30 training and test split is used, with the final output map being aggregated to display the most populated classes present in each separate field. Results of the study show that only 6 out of 56 fields were wrongly classified. Corn had a producer accuracy (PA) of 0.93 and a user accuracy (UA) of 0.97, and oats with a PA of 0.85 and a UA of 0.88. Soybeans had a moderate performance with a PA of 0.87 and a UA of 0.63, and alfalfa performed the worst with a PA of 0.54 and a UA of 0.88.

The study of marine macrophytes is becoming increasingly important due to the threat of climate change to intertidal environments and the potential of macrophytes as nature-based climate solutions. Laminaria digitata (finger kelp), Saccharina latissima (sugar kelp), and Zostera marina (eelgrass) are three marine macrophytes whose habitats are known as blue carbon ecosystems due to their outstanding carbon sequestration capabilities. These species are found throughout the Eastern Shore Islands, Nova Scotia, an Area of Interest (AOI) for ecological and biological importance identified by Fisheries and Oceans Canada. Hyperspectral satellite imagery has been little explored as a solution to mapping marine macrophytes in comparison to other remote sensing data, including multispectral imagery and airborne hyperspectral imagery. To test the efficacy of hyperspectral satellite imagery for mapping marine macrophytes in cold temperate regions, we mapped finger kelp, sugar kelp, and eelgrass using a PRISMA image, near Sheet Harbour, NS, within the Eastern Shore Islands AOI. The results show that machine learning classifiers can use hyperspectral imagery to differentiate marine macrophytes, but it is more challenging to differentiate between species with very similar reflectance spectra, such as finger kelp and sugar kelp. The classification accuracy also decreases at deeper depths, where the benthos-reflected signal is diminished. Further investigation is needed to determine the value of narrow hyperspectral bands for species level mapping; initial results suggest that hyperspectral imagery can achieve improved discrimination of spectrally similar species of submerged aquatic vegetation compared to multispectral imagery of the same spatial resolution.

Forest damage is an increasing global concern, particularly as climate change intensifies the frequency and severity of both abiotic and biotic stressors. Early detection of stress-induced damage is essential for effective forest management, yet conventional methods remain labour-intensive and slow. A significant knowledge gap persists regarding how abiotic stress, such as drought, interacts with latent fungal pathogens that can shift from asymptomatic to aggressive under unfavourable conditions.

Multispectral imaging has demonstrated strong potential for detecting physiological disturbances in tree canopies, including pest outbreaks, but its capacity to identify pathogen-specific damage remains poorly explored.

In this study, we investigate whether UAV multispectral drone imagery can detect canopy damage linked to fungal pathogens in Norway spruce (Picea abies). Research was conducted in two contrasting trials in southern Sweden, representing optimal versus drought-prone growth conditions. Across the 2023–2024 growing sea-sons, tree vitality, needle condition, and phenology were monitored and paired with fungal community data to classify reference trees by pathogen type and stress response.

Weekly drone flights provided multispectral imagery that was radiometrically corrected, canopy-segmented, and processed to derive vegetation indices and individual-tree crowns. Using reference trees as training data, statistical models will assess damage patterns and vitality loss.

We expect to detect and distinguish stress signatures arising from combined biotic–abiotic interactions. And validate the Eurich et al. damage model in older trees.

Early- and in-season crop mapping provides vital information for precision agriculture. It is still a challenge for early- and in-season crop mapping because of the limited available images and similar spectral information. This study aims to enhance early- and in-season crop mapping by developing a Genetic Programming (GP) method to construct customized crop features. GP automatically generated candidate features for the target-crop using early- or in-season images, selected programs with substantial value disparities between target and non-target crops through the fitness function, and finally outputted the customized feature after the evolutionary process. These customized features were then compared with commonly used spectral bands and vegetation indices to evaluate their effectiveness for early- and in-season crop mapping. The results proved that the customized crop features had significant advantages in both early- and in-season crop mapping. The early-season accuracy in April after crop planting was 3.97% to 9.53% higher than spectral features and vegetation indices. Based on the classification for the in-season crop mapping, the customized crop features maintained the best performance. Advantages of customized crop features include the ability to automatically select effective bands of useful months without requiring expert knowledge, the ability to catch and enlarge the subtle spectral differences with the early- and in-season images, and the little information redundancy compared with spectral features and vegetation indices. It can be concluded that the customized crop features are outstanding for early- and in-season crop mapping.

Efficient nitrogen (N) management is a key factor for sustainable potato production, as over- or under-fertilization can significantly affect yield, quality, and environmental outcomes. This study explores the potential of unmanned aerial vehicle (UAV) multispectral imagery and machine learning (ML) to predict in-season potato nitrogen status under field conditions in Manitoba, Canada. A DJI Mavic 3M equipped with four spectral bands (green, red, red-edge, and near-infrared) was used to capture canopy reflectance at 15 m altitude during the 2023 and 2024 growing seasons. Vegetation indices (VIs) such as NDVI, GNDVI, CIgreen, TCARI, and SRRE were extracted from orthomosaics and evaluated for their relationships with petiole nitrogen concentration (PNC). Feature selection methods including Recursive Feature Elimination (RFE), Boruta, and Partial Least Squares Regression (PLSR) were applied to enhance model efficiency. Random Forest (RF), Support Vector Machine (SVM), and Gradient Boosting Regression (GBR) algorithms were compared for prediction accuracy. RF combined with RFE achieved the highest performance (R² = 0.57), confirming its robustness to multicollinearity and nonlinear relationships. The results highlight the strong relevance of CIgreen and red-edge indices to N variability and demonstrate the potential of UAV-based spectral sensing integrated with ML for precision nitrogen management in potato systems.

Satellite remote sensing remains central to global mangrove forest mapping, yet the effectiveness of existing products is often limited by coarse spatial resolution and insufficient locally representative training data. These constraints are particularly evident in many African coastal regions, where access to very high-resolution satellite imagery and field observations is scarce. Deep learning–based super-resolution offers a promising alternative by enhancing the effective resolution of freely available imagery. This study evaluates the utility of applying semantic segmentation to super-resolved Sentinel-2 and Sentinel-1 data for mangrove mapping in two ecologically distinct regions: Tanzania and Ghana. Using analysis-ready data from Digital Earth Africa, temporal median composites of Sentinel-2 VNIR, red-edge, and SWIR bands, together with Sentinel-1 VH and VV images, were generated at 10 m resolution. A modified ESRGAN model was trained to produce imagery with a five-fold increase in spatial resolution. Both the original and super-resolved datasets were used to train a U-Net–based binary segmentation model, supported by training labels derived from Global Mangrove Watch data, Google Earth imagery, drone surveys, and fieldwork.

Results indicate that super-resolved imagery substantially improves the accuracy and precision of mangrove classifications relative to the original-resolution images. The enhanced spatial detail supports the detection of small mangrove patches, complex shoreline features, and local degradation patterns, yielding more complete estimates of mangrove extent. Incorporating Sentinel-1 backscatter further improves mapping accuracy. The study demonstrates that deep learning–based super-resolution can overcome key limitations of open-access satellite data, enabling more reliable, fine-scale mangrove mapping.

Operational quantification of forest aboveground biomass remains one of the most demanding components of Verified Carbon Standard (VCS) project monitoring, largely due to the need for repeated large-scale field inventories. To reduce costs and enable automated updates, CERFO developed for Ecotierra a hierarchical modelling system integrating field measurements, drone photogrammetry, and Sentinel-1/2 imagery. The system is designed to generate biomass updates autonomously every 1–3 years.

In 2024, seventy-two field plots were installed, and plot-level biomass was computed using regional allometric equations. Drone acquisitions from RGB and multispectral sensors produced high-resolution structural and spectral predictors (>300 variables). A machine learning ensemble (AutoGluon), trained on a stratified split of 68 plots, achieved strong accuracy (R² = 78.5%, relative bias = 1.7%). Bias-corrected drone predictions (via empirical quantile mapping) were then used as pseudo-observations for the satellite modelling stage. The Sentinel-based model, combining optical and radar indices, reached R² = 67.8% with a relative bias of 0.79%, demonstrating the value of multi-sensor integration.

A comprehensive uncertainty analysis using one million Monte Carlo simulations confirmed the stability of aggregated results (T² = 84.2%, mean bias = –0.22%), meeting MRV requirements for carbon reporting. The final operational product is a fully automated processing chain that retrieves satellite images, performs preprocessing, computes predictors, applies the trained model, and exports biomass maps.

This approach provides a robust and scalable solution for continuous biomass mapping and forest carbon monitoring. Future improvements include expanding field sampling across broader ecological gradients to reduce uncertainty in underrepresented environments and further strengthen model generalization.

Accurate estimation of aboveground biomass (AGB) is critical for understanding carbon dynamics and forest structure at regional and global scales. Waveform LiDAR, with its ability to capture detailed vertical profiles of vegetation, has proven highly effective for AGB estimation. However, spaceborne waveform LiDAR missions such as NASA’s GEDI face limitations due to sparse sampling, necessitating integration with complementary remote sensing datasets for continuous coverage. This study develops a comprehensive framework to evaluate the contribution of multispectral optical imagery (Sentinel-2) and dual-polarized SAR data (Sentinel-1 C-band and ALOS PALSAR L-band) in reconstructing forest structure across multiple canopy layers in a tropical forest in French Guiana. Using LVIS waveform LiDAR as a reference and an AGB map derived from LiDAR–SAR fusion, Random Forest models were trained to predict LiDAR waveform metrics at relative heights (RH10–RH98), followed by SHAP analysis to quantify feature importance. Results reveal that satellite data exhibit greatest sensitivity at mid-canopy levels (RH55–RH85), with SWIR bands outperforming other optical features, particularly during the dry season when canopy moisture is reduced. SAR features, especially cross-polarized channels, provide consistent contributions across biomass ranges, though their effectiveness declines in very dense forests (>350 Mg/ha).

Forest structural complexity is a key component of forest ecosystems and generally reflects the combined characteristics of tree height, diameter at breast height (DBH), canopy cover, tree spacing, and species composition. While spaceborne LiDAR systems such as GEDI provide near-global full-waveform observations for deriving the Waveform Structural Complexity Index (WSCI), the discrete distribution of GEDI footprints limits their spatial continuity. This study addresses this challenge by integrating GEDI-derived WSCI samples with multisource remote sensing data to enable large-scale mapping of forest structural complexity. We developed and compared machine learning models (RF, SVR) and a deep learning architecture (ConvNeXt) to evaluate their ability to predict WSCI from multisource remote sensing data. The results show that the deep learning framework, supported by multisource remote sensing data, effectively overcomes the discrete footprint limitation of GEDI and enables spatially continuous mapping of forest structural complexity at the regional scale. The ConvNeXt model demonstrated clear advantages, reducing RMSE and MAE to 0.55 and 0.43 (compared with 0.60/0.49 for RF) and improving IoA and correlation to 0.73 and 0.61, thereby enhancing the reliability of regional-scale complexity mapping despite the sparse GEDI footprint distribution. This provides a practical and scalable pathway for large-scale forest structure characterization and supports regional forest monitoring and management. Future work will include expanding the study area, incorporating additional field measurements, integrating terrain-related variables for improved modeling under complex topography, and exploring multi-year datasets to assess temporal dynamics in forest structural complexity.

By integrating remote sensing observations, topography, and crop growth parameters, a multimodal deep learning model named Soil Spectral–Spatial Cross-Attention Network (Soil-SSNet) is proposed. Soil-SSNet includes a spectral sequence convolution module to capture dynamic spectral features, a spatial attention module to address surface heterogeneity and uncover the response relationship between salinity and natural environmental variables, and a multi-head cross-attention mechanism that uses spatial features as Query to guide the selection of spectral-index responses. Compared to traditional machine learning models such as Random Forest (RF) and Support Vector Regression (SVR), the overall accuracy of Soil-SSNet improves by approximately 40%. After incorporating multi-source covariates (water conditions, crop growth status, and topographic factors), the model’s accuracy further increases by about 25%. With the addition of the cross-attention mechanism, accuracy improves by another 35%, significantly enhancing the fusion capability of spectral and environmental information and achieving soil salinity inversion with higher accuracy and stronger generalization. Finally, spectral sensitivity analysis reveals that the 705–750 nm and 1580–2350 nm bands contribute the most to salinity inversion. Mechanism analysis further uncovers a significant coupling effect among salinity, crop growth, and topography: vegetation growth characteristics reflect the intensity of salt stress, topographic factors dominate the redistribution pattern of water and salt, and soil moisture dynamics determine the accumulation and dispersion patterns of salinity. In summary, Soil-SSNet not only improves the accuracy and interpretability of soil salinity inversion in saline-alkali farmland but also provides quantitative evidence for understanding the environmental processes and mechanisms of salinization.

Tanzania hosts one of Africa’s largest Protected Areas (PA), yet deforestation remains widespread in surrounding unprotected landscapes. Assessing the effectiveness of PAs requires analytical approaches that account for environmental and accessibility biases inherent in PA placement. This study presents a remote-sensing-based spatial modelling workflow that integrates Global Forest Change (GFC) forest-loss time series (2012–2022) with terrain, accessibility, and demographic covariates to quantify avoided deforestation attributable to protection.

Biophysical and anthropogenic variables influencing forest-cover change, including elevation, slope, distance to roads, settlement density, and population distribution, were harmonised to a 30 m grid and combined with protected area boundaries from the World Database on Protected Areas. To address spatial biases, Propensity Score Matching (PSM) was applied to match protected forest pixels with statistically similar unprotected pixels, reducing confounding effects and enabling a credible counterfactual baseline. A binomial logistic regression model was then fitted to the matched dataset to estimate the likelihood of deforestation under different conservation categories.

Results show that protected forests were, on average, about three times more likely to avoid deforestation than comparable unprotected forests. National Parks and Game Reserves demonstrated the strongest outcomes, being nearly ten times more effective at avoiding deforestation. Nature Forest Reserves were around three times more effective, while Forest Reserves and Game Controlled Areas showed more modest effects, being roughly twice as likely to avoid deforestation.

The analysis is transparent, reproducible, and scalable, demonstrating how Earth observation and spatial causal inference can strengthen national forest monitoring, support conservation planning, and inform policy processes.

Climate change is increasingly disrupting agricultural ecosystems, particularly in Canada, where rising temperatures and altered precipitation patterns are impacting crop resilience. Alfalfa (Medicago sativa L.), a key forage crop valued for its productivity and nutritional quality, is especially vulnerable to winter stress due to reduced cold tolerance and increased damage from thaw cycles. This study presents a remote sensing-based framework for estimating fractional vegetation cover (FVC), a critical indicator of crop health and ecosystem stability. By integrating Sentinel-2 satellite imagery with high-resolution UAV data, the approach leverages machine learning algorithms, including random forest (RF) and gradient boosting (GB), to efficiently predict alfalfa FVC. UAV-derived RGB orthoimages provide detailed spatial reference data, minimizing the need for extensive field surveys. The proposed method demonstrates the potential of combining multi-source remote sensing with ML to capture complex vegetation dynamics and improve monitoring accuracy. Although both models showed high potential in estimating the alfalfa FVC, RF outperformed GB in terms of all evaluation criteria. The resulting FVC maps provide actionable insights for early spring field assessments, enabling the timely identification of damaged areas and supporting informed crop rotation decisions. The proposed framework was tested across multiple alfalfa fields in 4 provinces of Canada, including Quebec, Ontario, Nova Scotia, and Manitoba, demonstrating robust performance under varying environmental conditions. Its scalability and adaptability make it suitable for broader applications in precision agriculture and climate-resilient crop monitoring.

The Chambal–Yamuna Badland Zone (CYBZ) is among the most severely degraded semi-arid landscapes in India, where persistent gully development reduces soil productivity and threatens long-term environmental stability. This study examines how vegetation dynamics and surface deformation have evolved in the CYBZ over the past 25 years by integrating long-term optical and SAR-based remote sensing observations. MODIS NDVI data (2000–2024) processed through Google Earth Engine were used to track vegetation greening and browning patterns, while Sentinel-1A/B SAR datasets (2017–2024) were analysed using Persistent Scatterer Interferometry (PSI) and the Small Baseline Subset (SBAS) approach to quantify deformation linked to active erosion.

The NDVI time series shows a clear and statistically significant rise in vegetation cover across the region, with the strongest greening occurring in the eastern badlands, particularly after 2015. This widespread improvement aligns with increasing rainfall and indicates a gradual transition from highly eroded terrain to more vegetated and potentially stabilised surfaces. InSAR results reveal minimal ground deformation within major gullies, suggesting that gully erosion during the study period has been low. Seasonal fluctuations observed in PS displacement curves correspond to vegetation cycles rather than ongoing surface lowering. Negative deformation signals are predominantly associated with agricultural zones adjacent to the badlands.

Overall, the combined use of MODIS NDVI and Sentinel-1 InSAR provides a robust framework for monitoring ecological recovery and erosion dynamics in geomorphically fragile landscapes. The findings highlight increasing vegetation stability and reduced gully activity, offering new insights into the contemporary evolution of the Chambal badlands

Accurate prediction of wheat yield is crucial for ensuring food security through the use of machine learning techniques. This research aims to forecast wheat yield in Pakistan by integrating five remote sensing indices, including Green Normalized Difference Vegetation Index, Normalized Difference Vegetation Index, Enhanced Vegetation Index, Soil Adjusted Vegetation Index, and Atmospherically Resistant Vegetation Index, with five climatic variables: maximum Temperature, Minimum Temperature, Rainfall, Soil Moisture, and Windspeed alongside the drought index, Standardized Precipitation Evapotranspiration Index. Ten model combinations are created within two wheat season scenarios: Full Seasonal Mean and Peak Seasonal Mean. Employing two nonlinear ML algorithms, Random Forest and Support Vector Machines, as well as two linear models, LASSO and Ridge, the study aims to determine the most effective combination and ML algorithm in both scenarios. Results indicate that in SC1, the RF model combination (GNDVI + SPEI + WS + SM) outperformed other models (R2 = 0.75, RMSE = 2.40, MAE = 1.98). Similarly, in SC2, the RF regression surpassed SVM, with the model combination demonstrating the highest performance, achieving R2 = 0.78, RMSE = 2.25, and MAE = 1.88, followed by (NDVI + Tmax + Tmin + PPT + PET + WS + SM; R2 = 0.75). Notably, the linear LASSO model also exhibited comparable performance to RF, achieving R² values of 0.74–0.69 in both scenarios. The findings support SC2 for yield prediction, underscoring the significance and potential of ML methodologies in timely crop yield prediction, establishing a robust foundation for ensuring regional food security.

This study evaluates the utility of AlphaEarth Foundation (AEF) embeddings, a pre-trained geospatial foundation model, for regional wetland mapping in Stockholm County, Sweden. Accurate and up-to-date spatial information is crucial for planning, but traditional methods are challenged by the heterogeneity and variability of wetland environments.

Our research assesses how AEF's 64-dimensional feature vectors, summarizing multi-sensor satellite time series (Sentinel-1, Sentinel-2, Landsat) at 10m resolution, perform when integrated with established remote sensing variables (topographic, hydrological, and LiDAR derivatives) within standard machine-learning workflows (MLP). The methodology employs a two-step hierarchical classification based on the BIOTOP SE inventory: a Level-1 land-cover prediction (Huvudklass) followed by binary wetland identification within suitable classes.

Preliminary results demonstrate the potential of this approach. The Level-1 classification showed strong performance for certain classes (e.g., klass6 F1-score: 0.98). For the binary classification within klass4, the model achieved a robust F1-score of 0.87 for the target Wetland subclass (Precision: 0.90, Recall: 0.88). This work highlights the possibility of adapting global pre-trained satellite embeddings with traditional remote sensing inputs using light machine learning models for practical, policy-relevant environmental applications, such as updating national biotope inventories.

Food insecurity remains a major challenge in Rwanda, particularly in rural regions where stunting and anemia rates remain high. Because agriculture is dominated by smallholder farms (0.1–0.5 ha) with fragmented fields and frequent intercropping, accurate crop type mapping is both essential and difficult. Traditional machine learning approaches struggle to model the spatial–temporal variability of such landscapes, whereas CNN-based models require large annotated datasets that are costly to obtain.

We introduce GLSTM-MLP, a hybrid framework that integrates LSTM and MLP classifiers with precomputed Haralick descriptors to efficiently encode spatial context. By combining spectral bands (SB), radiometric indices (RI), and elevation data, the model decouples spatial and temporal dependencies, enabling robust crop type classification even with limited training samples.

Using 3 m PlanetScope time series imagery and drone-based ground-truth data from two Rwandan villages, we evaluated GLSTM-MLP against MLP, RF, and SVM across three feature scenarios: (i) SB + RI; (ii) SB + Haralick features; and (iii) SB + RI + Haralick + elevation. We further compared performance with 2DCNN-LSTM and 3DCNN.

GLSTM-MLP consistently outperformed all baselines, achieving F1-scores of 91%, 91%, and 93%, compared with 87–91% for MLP, 89–91% for RF, and 86–89% for SVM. While 2DCNN-LSTM and 3DCNN underperformed in this data-scarce setting (F1 < 85%), highlighting the advantage of integrating domain-driven feature engineering with sequential modeling.

These results demonstrate that combining temporal dynamics with engineered spatial context provides a practical, data-efficient pathway for accurate crop type classification in heterogeneous, smallholder-dominated farms in SSA, even under limited ground-truth availability.

Accurate monitoring of tree health is important for ensuring sustainable and efficient orchard management in precision agriculture. We evaluated a modified Mask R-CNN deep learning framework for assessing apple tree health using multispectral UAV imagery. The model was tested with four backbone architectures (ResNet-50, ResNet-101, ResNeXt-101, and Swin Transformer) on three image combinations: RGB, 5-band multispectral imagery, and three principal components (3PCs) derived from five spectral bands and twelve vegetation indices. Among all configurations, the Mask R-CNN with a ResNeXt-101 backbone trained on 5-band multispectral imagery achieved the highest performance, reaching an F1-score of 85.70%. In comparison, PCA-based 3-component inputs performed lower (F1 = 82.75%), indicating that while dimensionality reduction reduces computational cost, it may also discard critical information relevant to vegetation health.

Holm oaks (Quercus ilex) are a keystone species of the Mediterranean savannas in the southwest of the Iberian Peninsul, which are of high ecological and socioeconomic value. This ecosystem is increasingly threatened by Seca, a decline process of oaks driven by abiotic factors and the pathogen Phytophthora cinnamomi. Monitoring tree vitality is therefore essential, and structural traits such as diameter at breast height (DBH) provide early indicators of stress-related growth reduction.

LiDAR remote sensing enables efficient derivation of these metrics, but existing methods involve trade-offs: terrestrial laser scanning offers high detail but limited coverage, while airborne and UAS-LiDAR cover larger areas but often lack sufficient point density. Portable SLAM (Simultaneous Localization and Mapping) LiDAR systems offer a promising alternative, providing flexible, high-resolution data collection across broad areas.

This study assesses the potential of a portable SLAM system to derive holm oak structural attributes. In July 2025, approximately 450 trees across 17 ha in Majadas de Tiétar (Spain) were scanned. In a first attempt, based on an Outer Hull Model, DBH was derived for 9 trees by fitting convex hulls to point cloud stem slices extracted at 1.3m height. Initial validation against field measurements showed strong agreement (R² = 0.971; RMSE = 3.33 cm).

These first results demonstrate that portable SLAM LiDAR can reliably capture stem structure and support large-scale monitoring.

A study focusing on investigating how hyperspectral data can be used towards enhancing landcover mapping accuracy in the Drylands ecosystem, so as to support evidence-based decision-making, strengthen restoration planning, promote conservation efforts and enhance national and international reporting frameworks.

The study looks at spectral separability analysis to quantitatively show how best PRISMA hyperspectral data can distinguish land-cover classes being mapped in the study area as compared to Sentnel-2.

The study also presents supervised classification analysis results performed using Random Forest classifier on Sentinel-2 and PRISMA original image, PCA transformed image and MFN transformed image and comparing their accuracy levels.

Finally, the study looks at spectral un-mixing to be able to quantify in terms of abundance, which landcover class is present in each pixel and to what proportion.

This study advances precision viticulture by developing a scalable hyperspectral and GeoAI framework for grapevine cultivar discrimination. Using proximal spectrometry and satellite hyperspectral imagery, the research demonstrates the methodological and feature-level transferability of spectral information from in-field spectrometry to spaceborne data. Ten machine learning and deep learning algorithms were evaluated, with Support Vector Machines (SVM) and a 1D Convolutional Neural Network (1D CNN) achieving the highest accuracies. A novel Partial Least Squares (PLS) ensemble feature selection approach reduced data dimensionality by 95%, identifying key red–NIR, Green and SWIR spectral regions for cultivar mapping. Transferring these features to pansharpened PRISMA hyperspectral satellite imagery yielded high classification accuracies (>80%) at 5 m resolution, confirming the operational potential of hyperspectral GeoAI for vineyard characterisation. The findings establish a foundation for scalable, satellite-driven cultivar mapping to support site-specific management and digital viticulture practices.

TBA ...

The Ecuadorian Amazon faces growing socio-environmental pressure from gold mining, which threatens biodiversity and ecosystem integrity. In Zamora Chinchipe, a province that hosts more than 600,000 ha under protection (18% of Ecuador’s continental protected areas), mining expansion reveals a critical tension between conservation and extraction. This study evaluates the spatial distribution and intensity of gold mining in the buffer zones of six protected areas, using data from MapBiomas Ecuador and Geographic Information System (GIS) techniques.

Mining areas within 5 km of each protected area were extracted from MapBiomas LULC maps and analysed through Kernel Density Estimation (Epanechnikov function, 500 m cell size, 2500 m radius). The results reveal heterogeneous mining pressure, with hotspots concentrated in Cerro Plateado, Podocarpus, and El Zarza, often within 1.6–5 km of official boundaries. Spatial correlation shows that 89% of hotspots lie within 500 m of watercourses and 78% in slopes between 15°–35°, highlighting the geomorphological and hydrological dependency of mining activities.

Conversely, areas such as Yacuambi and Tiwi Nunka show minimal pressure, where local governance and indigenous territorial control have effectively limited extractive expansion. These results demonstrate that governance factors are as critical as physical conditions in determining conservation outcomes.

The integration of MapBiomas data and KDE offers a replicable, low-cost tool for monitoring mining dynamics, providing spatial evidence to strengthen protected area management and inform sustainable territorial planning in the Amazon region.

We are presenting an atmospheric modeling and inversion pipeline for the FINCH (Field Imaging Nanosatellite for Crop residue Hyperspectral mapping) hyperspectral imaging CubeSat, built by the University of Toronto Aerospace Team Space Systems division. Such a pipeline will allow us to validate our spectral unmixing pipeline under possible FINCH imaging conditions, and permit scientifically useful data collection. FINCH, through innovations in crop residue cover mapping, will further enable sustainable agricultural practices.

Tree hydraulic functioning varies strongly over the diurnal cycle as transpiration, stomatal conductance, and xylem tension shift with radiation and vapor pressure deficit (VPD). Capturing these within-day dynamics remotely is essential for interpreting stress signals from airborne or satellite sensors, yet most remote-sensing studies treat forest condition as static. For conifers such as Norway spruce (Picea abies), even brief midday hydraulic limitation can contribute to long-term drought vulnerability and bark-beetle susceptibility, underscoring the need for diurnal monitoring.

Optical indicators such as the Photochemical Reflectance Index (PRI) track rapid photosynthetic-efficiency changes via xanthophyll-cycle activity but often saturate in dense canopies and are sensitive to geometry and structural effects. PRI’s two-band design also underrepresents the broader green-shoulder region (520–550 nm), where carotenoid–chlorophyll interactions more reliably reflect stress in evergreen species.

To address these limitations, we apply a family of green-shoulder indices (GSCR) that integrate information across multiple narrow bands in the 520–550 nm plateau. These indices capture both rapid xanthophyll dynamics and slower pigment adjustments linked to declining water status. Recent UAV studies show that GSCR metrics are highly sensitive to early physiological stress in spruce.

This study extends GSCR use from static health detection to tracking diurnal hydraulic processes. We (1) test whether spectral indices can reproduce the within-day trajectory of spruce physiological activity, including midday depression, and (2) quantify relationships between spectral indices, leaf water potential, and sap-flow velocity. By combining drone-based hyperspectral imaging with high-frequency hydraulic measurements, we establish a framework linking pigment dynamics to diurnal hydraulic status at the crown scale.

Ground-level air temperature (Tair) is an essential variable for climate monitoring, agricultural management, and hazard prevention. Conventional ground-based measurements often fail to capture the fine-scale spatial variability, especially in regions with complex terrain. Land Surface Temperature (LST) remote sensing offers a complementary solution, providing spatially continuous and temporally frequent observations. This study evaluates the potential of MODIS and Landsat-8 LST products to estimate Tair in a heterogeneous agricultural landscape. We developed spatiotemporal regression models linking satellite-derived LST to ground observations from meteorological stations over five years (2018–2022). MODIS data provided high temporal coverage through 8-day composites, while Landsat-8 offered higher spatial resolution LST via the Statistical Mono-Window algorithm. The models were validated using Leave-One-Out Cross-Validation, achieving high predictive accuracy for MODIS-based Tair estimation (R² = 0.981, RMSE = 1.1 °C), whereas Landsat-8 captured finer spatial variability (R² = 0.859, RMSE = 3.4 °C). Our results demonstrate that integrating multi-resolution LST products enables accurate, dense mapping of Tair, supporting operational forecasting for precision agriculture. The study also discusses limitations related to land-cover heterogeneity, temporal representativeness, and potential extensions using spatial correlation methods or radar-derived crop-structure information.

This study uses UAV imagery and videos to monitor cattle behaviour in a rotational grazing system in Thunder Bay, Ontario, Canada.

Heat waves and droughts are recurring global phenomena that profoundly influence terrestrial ecosystems. This study assessed the impacts of drought and heat waves on the functioning of tropical evergreen (Kerela, South India) and moist deciduous (Barkot, North India) forest sites using meteorological and satellite vegetation products, including the FAPAR, SIF, GPP, and ET during 2007– 2018. The SPI, calculated from ERA5 daily data, was used to identify drought occurrences in space and time, while the Mann–Kendall test detected historical trends. Heat waves were characterized using hourly maximum temperature data from ERA5, following the criteria of IMD. Temporal anomalies were quantified using Z-scores, along with the Mann–Kendall trend test and Theil–Sen’s slope analysis. Moist deciduous showed consistent and pronounced declines in productivity and moisture related variables during droughts, particularly during monsoon season, indicating strong sensitivity to water stress. Evergreen forests exhibited more variable responses, with weaker and less consistent drought signals during the pre-monsoon season and mixed responses in vegetation variables even during monsoon drought conditions. Heat wave impacts also varied across forest types. Evergreen forests showed contrasting responses depending on the timing of heatwave events, with early and mid-heatwave phases associated with reductions in productivity, while late heatwave events showed relatively positive or mixed responses among vegetation indicators. In moist deciduous, heatwaves resulted in more consistent negative anomalies in productivity-related variables, although some years exhibited contrasting behaviour in SIF relative to other indicators. The findings underscore the heterogeneous response of forest ecosystems to extreme climatic events.

Accurate prediction of soil organic matter (SOM) in saline semi-arid regions is vital for sustainable land management. This study evaluates EnMAP hyperspectral imagery combined with machine learning (ML) for SOM estimation using Random Forest (RF) regression applied with three feature selection (FS) algorithms. Embedding RF-FS, recursive feature elimination (RFE), and Competitive Adaptive Reweighted Sampling (CARS), alongside five spectral pre-processing techniques have been analysed. The first derivative (FD) transformation significantly enhanced model performance, outperforming other processing methods. FD combined with RF-FS demonstrated the highest accuracy (PCCC = 0.697, R² = 0.446, RMSE = 0.825) compared to other FS methods. In contrast, all feature-selection approaches applied to the original reflectance showed substantially lower performance, with OR-RFE, OR-RFFS, and OR-CARS yielding PCCC values of only 0.617, 0.610, and 0.547, respectively, and consistently higher RMSE values near 0.90. Frequency analysis identified key informative bands in the SWIR-2 (2207–2445 nm) and visible–NIR (418–801 nm) regions, aligning with known organic matter absorption features. These results demonstrate that integrating derivative spectroscopy with robust feature selection substantially improves SOM prediction in challenging semi-arid environments, providing a effective framework for operational remote sensing of soil fertility.

UAV-based lidar systems operate at relatively low power compared to airborne platforms, making them especially sensitive to environmental conditions that influence return strength, canopy penetration, and ground-surface detection. Daytime acquisitions are affected by solar illumination, which introduces photon noise and reduces signal-to-noise ratio, while nighttime missions eliminate this interference and should theoretically enhance intensity and penetration. However, nighttime humidity, dew, and shallow fog can attenuate or scatter the laser pulse, complicating the expected benefits. This study evaluates the combined influence of illumination and atmospheric moisture on UAV-lidar performance across dense and wet vegetation.

We conducted paired day and night flights (Hesai XT32M2X lidar) at 40 m, 75 m, and 100 m over two contrasting peatland sites in eastern Canada: the humid, densely vegetated Alfred Bog (Ontario) and the drier Quinces Bog (Nova Scotia). Independent GNSS checkpoints were used to assess positional accuracy without constraining point-cloud processing.

Nighttime flights at the dry site yielded the clearest benefits, including increased third-return frequencies, stronger intensity values, and slightly improved ground-surface accuracy. In contrast, nighttime flights at Alfred Bog were affected by fog and canopy-level moisture, which produced increased scattering, duplicate points, and reduced ground detection, particularly at lower altitudes. These results show that while nighttime acquisition can improve data quality, its effectiveness depends strongly on humidity and surface wetness.

Overall, both illumination and environmental conditions must be considered when planning UAV-lidar missions. Nighttime flights offer advantages under dry conditions but may degrade outcomes in humid environments where fog or dew is present.

Koshalya‑Jhajhara (K-J) watershed in north India, tributaries of Ghaggar River, covering an area of 134.92 km2 in north India was assessed using multisource geospatial datasets. Satellite imagery, ASTER digital elevation model, Survey of India topographic maps and ancillary thematic data were processed to derive key maps including land use and land cover, geology, geomorphology, drainage density and slope. A temporal comparison of land use for 1999–2000 and 2015–2016 reveals rapid urbanization and shrinking forest cover. Built‑up area expanded from 7.12 km2 to 24.84 km2 while forest area contracted from 109.35 km2 to 96.78 km2, indicating substantially increased water demand and reduced natural recharge capacity. Groundwater potential was assessed by integrating geology, geomorphology, drainage density, slope and land use/land cover through an Analytic Hierarchy Process based multi‑criteria evaluation. The resulting Groundwater Potential Zone map shows that the majority of the watershed is stressed: 61.83 km2 falls in the poor category and 37.87 km2 in the very poor category for groundwater availability. These findings highlight critical zones where recharge and demand‑management interventions are urgently required. Surface water enhancement options were identified through geospatial suitability analysis. Fourteen sites were delineated for check dams and fifteen sites for percolation tanks, selected to maximize recharge potential, minimize sedimentation risk, and complement existing drainage patterns. When implemented, these structures will increase local infiltration, raise groundwater tables, and reduce peak surface runoff, thereby improving water security for domestic, agricultural and ecological needs. Measures have been recommended for sustainable land and water management in the watershed.

The spectral range of the Field Imaging Nanosatellite for Crop residue Hyperspectral Mapping (FINCH), developed by the University of Toronto Aerospace Team’s Space System Division, poses significant challenges for hyperspectral unmixing to determine crop residue cover fractional abundances. The severe lack of standard indices necessitates the use of complex, data-driven unmixing models. Complex unmixing models require dense, well-distributed manifolds, whereas ground-truth datasets are sparse and expensive. To better leverage the information content in existing ground-truth datasets, this study presents a framework that decouples interpolation and abundance-mapping estimation via an intermediary conditional data generator, contrasting with traditional single-model unmixing pipelines. This decoupling enables the inclusion of physical prior knowledge about spectra, which is not natively accessible to unmixing models, thereby artificially expanding the dataset to serve as a Monte Carlo approximation of the population risk. To test this hypothesis, we have proposed two generator models: a 1D Conditional U-Net with Conformer Layers (GD-Streamline) and a Dual-Path Transformer Conditional Variational AutoEncoder (TCVAE), and two unmixing models: Multi-Layered Perceptrons (MLP) and Fourier Neural Operators (FNO). The results of the study indicate a need for rigorous integration and introduction of spectral priors within the scope of the proposed decoupling framework to prevent domain and mode collapse and hallucination by the generator models; otherwise, this leads to falsely approximated data manifolds, ultimately resulting in unsatisfactory and out-of-distribution unmixing performance.

Savannahs are vital ecosystems whose sustainability is endangered by the spread of woody plants. This research targets the accurate mapping of fractional woody cover (FWC) at the species level in a South African savannah, using EnMAP hyperspectral data combined with Sentinel-2 multispectral imagery. Field annotations were intergrated with very high-resolution multispectral drone data to produce land cover maps that included three woody species. The high-resolution labelled maps were then used to generate FWC samples for each woody species class at the 30-m spatial resolution of EnMAP. Several machine learning regression algorithms were tested for FWC mapping on multi-seasonal and/or multi-annual EnMAP and Sentinel-2 imagery. Highest accuracy rates were achieved when incorporating data from both the dry and wet seasons, and for most experiments, data acquired across more than one year. The achieved results demonstrated the suitability of our approach for accurately mapping FWC at the species level and highlighted the synergistic potential of EnMAP and Sentinel-2 data for monitoring savannah ecosystems.

This study investigates the sensitivity of LiDAR, optical, and L-band SAR features for estimating forest above-ground biomass (AGB) in Istanbul’s Belgrade Forest by integrating GEDI lidar measurements with Sentinel-2 and SAOCOM 1A data. Using GEDI L4A AGBD footprints as reference, the authors derived 27 multisource predictors, including spectral indices, SAR backscatter, and polarimetric decomposition parameters. Four machine-learning models—MLP, Kernel Ridge, Lasso, and Elastic Net—were trained and evaluated using a 70/30 train–test split and repeated k-fold cross-validation. Results show limited but notable predictive capability, with R² values ranging from 0.15 to 0.20. The MLP achieved the highest accuracy (R² = 0.20), which can be attributed to its ability to model nonlinear relationships between biomass and multispectral features. Feature-selection experiments reveal that Sentinel-2 red-edge and SWIR bands, along with vegetation indices such as NDVI_red, LSWI, and Cire, consistently provide the strongest contribution to biomass estimation. SAR features contributed less significantly. The study concludes that optical features currently outperform L-band SAR for biomass modelling in this environment and recommends incorporating topography, multitemporal SAR, and additional machine-learning approaches to improve AGB prediction accuracy in future work.

Rapid urbanization in the plain river network of Changshu City has led to prominent water quality degradation and eutrophication risks. Traditional in-situ monitoring is constrained by sparse sampling and high costs, while conventional remote sensing approaches struggle with accurate water body extraction and stable parameter inversion in turbid, fragmented rivers. This study establishes a targeted remote sensing monitoring framework using 2024 multi-source data (Gaofen-2, Sentinel-2) and field measurements. An optimized modified Normalized Difference Water Index (mNDWI) combined with a spatially weighted adaptive threshold algorithm is adopted to precisely extract complex river networks. Based on Pearson correlation analysis, sensitive spectral bands and band combinations are screened for four key indicators: Chlorophyll-a (CHL-a), Total Nitrogen (TN), Total Phosphorus (TP), and Secchi Depth (SD). Statistical regression and weighted Principal Component Analysis-Random Forest (PCA-RF) models are developed for quantitative inversion, and their accuracy is verified using cross-validation with R² and RMSE. A weighted modified Carlson Trophic State Index suitable for plain river networks is applied for eutrophication assessment, and the single-factor pollution index method combined with the worst-factor principle is adopted to conduct comprehensive water quality evaluation in accordance with the national surface water environmental quality standards. The integrated inversion–evaluation–mapping workflow realizes spatially continuous water quality analysis, providing a reliable and region-adapted technical solution for remote sensing monitoring and scientific management of water environment in plain river network areas.

Mowing event temporal localization in dense satellite time series is crucial for monitoring agricultural practices and supporting sustainable land use policies. This study presents an innovative approach using a foundational model (FM), Prithvi-EO-2.0, tailored for Earth Observation time series, to precisely detect and temporally localize mowing events in grassland parcels. Unlike traditional methods that primarily identify mowing occurrence or frequency, this work advances the temporal pinpointing of individual mowing events, addressing challenges related to sparse annotations and diverse agro-climatic contexts. The methodology leverages high-resolution Harmonized Landsat and Sentinel-2 (HLS) optical data, treating time series as sliding temporal windows to capture rapid vegetation changes. The FM backbone is combined with a trainable localization head to predict the precise timing of mowing events, supported by a postprocessing step to reduce false detections. The dataset used includes over 450 newly annotated parcels from Central Greece, enabling robust training and comprehensive evaluation with metrics such as F1-Score, Precision, Recall, and Temporal Distance between predicted and actual events. Preliminary results demonstrate a significant improvement in localization performance, with a 6% increase in F1-score and an average temporal deviation of 2.7 time steps from ground truth. Ablation studies validate the impact of temporal window length, model architecture, and postprocessing on performance. The study highlights the strong generalization capabilities of FM-based approaches despite limited fine-tuning data, paving the way for enhanced agricultural monitoring using multi-temporal satellite data.

Macadamias are the fastest-growing fruit tree crop in South Africa, but the industry is met with challenges due to changing environmental conditions exacerbated by climate change. One of the challenges increasing in frequency and intensity is out-of-season flowering events. These events result in serious problems for orchard management, harvesting practices, and orchard sanitation. Understanding macadamia phenology is, therefore, important and should be investigated as timely phenology changes are crucial in the agricultural sector, particularly the timing of flowering phenology.

Multispectral remote sensing has been successful in quantifying flowering using conventional vegetation indices. However, based on the canopy distribution of macadamia flowers occurring predominantly below the dense evergreen canopy, the use of multi-spectral vegetation indices needs to be complemented to ensure the dependability of the phenology assessments. Synthetic aperture radar data could potentially address these limitations by facilitating the monitoring of within-canopy structural changes throughout the phenology evolution. Furthermore, to ensure validation of flowering phenology signals captured by satellite sensors, the use of unmanned aerial vehicles offers an intermediate level of observation. Therefore, this study aimed to investigate macadamia phenology through the integration of multi-sensor, multi-scale remote sensing data to advance the detection of flowering dynamics in macadamia, located in Barberton, Mpumalanga, South Africa.

This study highlights that macadamia flowering can be detected from a remote sensing perspective, despite the limitation of the flowers being inconspicuous, underscoring the value of integrating optical and synthetic aperture radar data to improve flower detection.

Mapping perennial crops in tropical regions remains challenging due to high spectral complexity, frequent cloud cover, and phenological overlap between different types of vegetation. This study evaluated the potential of the Harmonized Landsat-Sentinel (HLS) dataset to identify perennial crops in the municipality of Jacupiranga, São Paulo, Brazil, an area representative of the Atlantic Forest mosaic. A hierarchical classification was applied using the Random Forest (RF) algorithm on temporal compositions of 2024 NDVI, NDWI, and BSI indices, structured into three analytical levels: (1) natural vegetation versus anthropic areas, (2) perennial crops versus other uses, and (3) banana versus peach palm. Accuracies ranged from 0.86 to 0.98 and F1 ranked between 0.86 and 0.95. The most influential variables were concentrated in transitional periods of the annual cycle, reflecting subtle changes in canopy moisture and vegetative vigor rather than a clear distinction between dry and wet seasons, which are not well-defined in this tropical humid environment. The final maps indicate that approximately 25% of Jacupiranga is agricultural land, of which 4,320 ha correspond to perennial crops, with 80% occupied by banana plantations. The results demonstrate the potential of HLS open data to generate accurate multiscale mapping of perennial crops in complex tropical landscapes, supporting digital agriculture and sustainable management in family farming regions.

This study identifies the primary drivers of natural land conversion in Chile's 343 municipalities from 1999 to 2024. Using comprehensive spatial data, the analysis reveals that urban and road density are the most significant drivers of this change. In contrast, higher human development acts as a mitigating factor. The strength of these drivers is also shown to be highly dependent on local precipitation patterns. These findings provide a critical, evidence-based foundation for targeted land-use policy and ecosystem conservation in Chile.

To address the limitations of traditional fixed-point sampling for monitoring water chlorophyll concentration and improve the inversion accuracy of complex inland waters, this study took the Naoli River Nature Reserve as the research area and conducted research based on Sentinel-2 images and measured chlorophyll point data. First, the study performed combined calculations on multispectral bands to generate multiple derived bands, and selected "B4+B5+B6" as the optimal band combination through the coefficient of determination (R²). Then, using this combination as input, various empirical models were constructed and evaluated using multiple indicators. The results showed that the univariate cubic function model had the highest R² and the smallest multiple error indicators, which was significantly better than other models, and successfully realized the spatial inversion of chlorophyll concentration in the study area. This study reveals the complex nonlinear relationship between chlorophyll and band combinations, provides a high-precision inversion technical scheme for water chlorophyll, offers data support for algal bloom early warning and water quality fluctuation tracking, and provides scientific references for the optimization of river basin management measures and ecological protection decisions.

As climate change is increasingly affecting marine and terrestrial ecosystems, researchers, resource managers, and coastal communities are using satellite-based remote sensing, such as Sentinel-2 multispectral imagery, to monitor coastal environments at large scales. The ability to automatically define the position of the coastline from imagery, referred to as “coastline extraction”, is a valuable tool in extending monitoring of coastal ecosystems, such as kelp forests and eelgrass meadows, to regional and provincial scales. In this work, we present a new dataset for water segmentation, and thus coastline extraction, consisting of manually annotated Sentinel-2 images acquired at low tide, specific to the Pacific coast of British Columbia (BC), Canada. We then evaluate three methods for coastline extraction: an adaptive thresholding method, and two convolutional neural networks trained on firstly, a global dataset and secondly, our newly created BC dataset. The model trained on the BC dataset achieved the highest accuracy across standard image segmentation metrics and in coastline positional error measured relative to a manually defined reference coastline. Additionally, very-high resolution unmanned aerial vehicle data collected at validation sites with comparable tide levels to the Sentinel-2 dataset imagery showed that training on BC specific data decreases pixel misclassification, and therefore coastline positional error, due to the presence of subtidal and intertidal algae and vegetation at various validation sites in the study area.

Climate fluctuations influence many aspects of agriculture including crop growth, soil conditions, distribution of fertilizer and water resources. These climatic fluctuations thereby pose significant challenges for agricultural productivity worldwide. However, the availability of agricultural datasets to study the impact of various adverse climatic conditions on different crops remains limited. To address the data availability limitation for agro-climatic impact study, this paper introduces an Explainable Climate-Aware Generative AI framework (XCA-GenAI). The framework combines a Conditional Tabular GAN (CTGAN) to generate realistic synthetic datasets, a Random Forest (RF) regressor to predict crop yield and stress-related parameters, and a SHAP-enabled “what-if” simulation module that evaluates and explains crop responses under varying temperature and rainfall conditions. The proposed framework is employed to generate synthetic representations of ten climatic variations ranging from Very Hot–Dry to Very Cool–Wet using SF24 dataset. Crop-specific predictive models then estimate how change in climatic condition alters crop density, pest pressure, and frost risk. Further, explainability analysis provides interpretable insights of climate impact across multiple crops represented in the dataset. Comprehensively, this work introduces a climate-aware agricultural decision-support framework to aid farmers and agronomists for informed decision making under varying climatic conditions.

Effective monitoring of cotton fields, especially at the regional level, while also detecting diseases in individual plants, remains a significant problem in precision agriculture. This paper presents a combined framework for monitoring cotton in Pakistan, using satellite remote sensing and artificial intelligence-based leaf image classification. Multi-temporal Sentinel-2 imagery from the 2022 kharif season was used to map cotton fields and evaluate canopy condition during the growing season.

Cotton fields were mapped using a Random Forest classifier with an overall accuracy of 93% and a Kappa coefficient of 0.82. The estimated cotton acreage of 65,269 ha nearly matched official figures. The crop state inside the mapped cotton area was then evaluated using a Fused Health Index constructed from NDVI, EVI, NDMI, NDRE, and SAVI. The results showed geographic variability in canopy condition, with 24.5% of the region falling into the low-health class, 50.9% in the moderate-health class, and 24.6% in the high-health class. A Vision Transformer model achieves 97% accuracy in classifying RGB images of cotton leaves into eight diseases and conditions. The satellite analysis identifies where stress is concentrated at the district scale, while the image-based model gives symptom-level diagnostic help. Together, these results suggest that combining remote sensing and artificial intelligence can improve timely cotton monitoring and allow more targeted field management.

Sand mining is the practice of removing sand from lakes, rivers, and streams using various techniques, including dry and wet pool mining, bar excavations, skimming, and scalping. Excessive and illegal mining activities can lead to severe environmental hazards, including deforestation, water pollution, soil pollution, and air pollution. Fallacious mining practices can also lead to the exhaustion of resources on open lands and riverbeds. In this study, geospatial techniques have been utilised to investigate and audit sand mining practices surrounding a city in North India, where resources are extracted for the city's development. The present study was conducted with the objective of identifying unauthorised mining activities in allotted sites as well as nearby areas, mapping and verifying mining operations in relation to approved mining plans and environmental clearances, and estimating the minor minerals extracted at the mining site. Satellite images from Sentinel-2 and Google Earth, along with the coordinates of the lease area and EIA reports for the site, served as the data sources for the research. Violations of the rules, such as flow obstruction, mining along riverbanks, and mining outside the lease area, were observed through the use of remote sensing images. Furthermore, it can be concluded from the present study that satellite-derived analysis offers a time and cost-effective means of inspecting mining areas.

Repeated airborne laser scanning (ALS) offers new opportunities to quantify individual-tree structural dynamics over time. In this study, we analysed annual to multi-year changes in height growth and crown structure of mature Norway spruce in southern Sweden using repeated ALS acquired in 2016, 2017, 2019, and 2022. Individual trees were delineated from normalized point clouds, and tree height increment, maximum crown radial extension, crown projection area, and crown-boundary metrics were derived to evaluate temporal structural change. To interpret the results in relation to recent climate extremes, the observation period was divided into pre-drought (2016–2017), during-drought (2017–2019), and post-drought (2019–2022) phases, and structural changes were expressed on a yearly basis. Tree height increased significantly in all periods. Annualized median height growth was 24.9 cm yr⁻¹ before drought, 26.5 cm yr⁻¹ during drought, and 16.8 cm yr⁻¹ after drought. In contrast, annualized maximum crown radial expansion was limited before drought (2.4 cm yr⁻¹), peaked during drought (19.0 cm yr⁻¹), and was nearly absent after drought (0.17 cm yr⁻¹). Crown-boundary metrics further suggested an upward shift of the upper crown and a reorganization of the middle crown over time, although lower-crown estimates were more uncertain. Driver analyses showed that height growth was mainly related to tree size before and during drought, whereas post-drought growth became more influenced by stand competition. Overall, this study shows that repeated ALS can be a useful tool for analysing crown structural dynamics during and after drought, providing a promising basis for monitoring how tree architecture responds to stress.

The objective of this study is to build an assessment model of nitrogen balance from reservoir to valley fields linked to reservoirs by combining spatial information processing, chemical analysis and bioanalysis. To this end, the biomass of rice and weeds was detected from hyperspectral images mounted on UAVs, the growth processes of rice and weeds were understood with GIS, and the biomass of rice and weeds was estimated separately to estimate the amount of nitrogen fixation in rice.

This study showed the possibility of determining the distribution of humus from the distribution of carbon and nitrogen in a paddy field by using a UAV hyperspectral sensor, a random forest. Furthermore, by qualitatively assessing the contribution of soil micro-organisms to nitrogen fixation in rice using soil microbial diversity and activity values (BIOTREX), a model was constructed to enable an assessment of the nitrogen cycle derived from organic nitrogen supplied by the reservoir, and the nitrogen balance was estimated.

We showed that the nitrogen balance can be evaluated from the balance of soil-derived nitrogen and humus-derived edible nitrogen in reservoirs, rainfall, and rice paddies by chemical analysis. By combining this with BIOTREX, it was shown that when humus-derived edible nitrogen is high and BIOTRX values are high, the change of organic nitrogen in humus to inorganic nitrogen is promoted. It was shown that this could be used as an indicator of the need for fertilizer inputs and as a method for assessing the potential of agricultural land.

Accurate crop maps are important for agricultural monitoring in water-limited regions because they provide spatial information for crop inventory assessment, land management, and resource planning. In the Navajo Nation, crop classification is challenging because agriculture is influenced by arid environmental conditions, limited water availability, and unevenly distributed cultivated land. This study evaluates a crop-classification workflow for a selected agricultural Region of Interest (ROI) within the Navajo Nation using Sentinel-2 imagery, the USDA/NASS Cropland Data Layer (CDL), and the CDL confidence layer in Google Earth Engine. High-confidence CDL pixels (confidence ≥ 95%) were used to construct pseudo-reference samples for the 2017 and 2022 growing seasons, and a 3 × 3 neighborhood homogeneity filter was applied to reduce local label uncertainty. Spectral predictors derived from Sentinel-2 imagery included the Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), Green Chlorophyll Vegetation Index (GCVI), and Land Surface Water Index (LSWI). A Random Forest classifier was implemented separately for each year using an 80% training and 20% testing split. The resulting classifications achieved overall accuracies of 87.30% for 2017 and 90.88% for 2022. These results show that confidence-screened CDL samples combined with multi-temporal Sentinel-2 features can support reliable crop classification within the selected ROI under limited reference-data conditions and provide a practical basis for agricultural monitoring in the Navajo Nation.

The remote sensing (RS) and geographic information systems (GIS) technologies combined with artificial intelligence (AI) enable more efficient and sustainable agricultural ecosystems. In recent years, the use of the machine learning and the deep learning models trained over the geospatial data have emerged as a pivotal catalyst for sustainable and smart agriculture initiatives. The unmanned aerial vehicle (UAVs; drones) has become as a transformative force in the context of crop health monitoring, disease detection and yield predictions combined with supervised and unsupervised machine learning that helps to revolutionizing the motoring, deep analysis and timely decision making.

This study presents the integration of remote sensing technologies (e.g., UAVs, drones) combined with data-driven artificial technologies to help the farmers in precision agriculture for cotton framing for plant health monitoring against pest infestation, micro irrigation and crop yield prediction. A UAV-based dataset for cotton crops is prepared from a region in Pakistan. The prepared dataset is evaluated through multiple experiments using classical supervised machine learning algorithms for the classification; Naïve Bayes (NB), Support Vector Machine (SVM), and Random Forest (RF). These classification algorithms helped to classify the cotton crop health; healthy or unhealthy. The experimental results indicate that the RF algorithm outperforms the other applied machine learning methods, in terms of its accuracy and precision.

The research focuses on the use of advanced remote sensing techniques to fight the effects of climate change in urban areas, with particular reference to heat islands. The proposed methodology, applied to the metropolitan city of Naples, is based on the analysis of very high-resolution satellite images from the WorldView-3 constellation, combining panchromatic and multispectral data through pan-sharpening to obtain detailed maps of urban vegetation, including smaller green spaces such as flower beds, tree-lined avenues, private gardens, green roofs, which are often overlooked because they are difficult to map in a sustainable and widespread manner. Through the calculation of spectral indices (NDVI, MSAVI2, GNDVI, NDRE), the study has enabled not only the precise geolocation of photosynthetically active areas, but also the monitoring of their health status by comparing satellite datasets acquired in June and September 2023. The results highlight marked water stress during the summer period, manifested by a reduction in the average values of the indices. These results constitute a valuable decision-making tool for resilient urban planning and the implementation of Nature-based Solutions, and demonstrate the sustainability and replicability of the methodology in other territorial contexts.

Accurate estimation of wildfire-driven carbon loss in boreal forests requires spatially explicit prediction of both aboveground and belowground combustion, yet most existing approaches remain region-specific and are rarely evaluated for transfer across fires or geographic domains. Here, we develop and evaluate a transferable modelling framework for plot-level aboveground and belowground carbon combustion using field observations linked to remotely sensed burn severity, vegetation structure, biomass, climate, terrain, peat occurrence, and fire-weather predictors.

Models were trained in western Canada and evaluated using fire-wise hold-out data within the training region and independent transfer domains in Alaska and Québec. To avoid optimistic performance estimates, all tuning and validation were conducted using grouped cross-validation at the fire-event level. Predictor formulations were defined a priori to represent alternative ecological hypotheses about combustion controls. Predictors included Canadian Fire Weather Index components (FFMC, DMC, DC, BUI and FWI), calculated using MODIS-derived burn dates and 7-day antecedent means. After recursive feature elimination, compact non-collinear predictor subsets were retained for modelling.

Predictive performance varied more strongly among predictor formulations than among model families, indicating that ecological representation exerts greater influence on transferable combustion modelling than algorithm choice. For aboveground combustion, the strongest model achieved R² = 0.31 and RMSE = 682.7 g C m⁻². Belowground combustion was more difficult to predict and was best represented by a climate-augmented nonlinear structure. Transfer to Alaska was weakest for both responses, and high-combustion observations were systematically underpredicted, highlighting uncertainty associated with rare extreme burns.

FINCH is a student-led satellite mission whose novel sensor and cost effective form seek to provide crop residue mapping at a much lower cost and aid in smart-agriculture initiatives. To achieve this, crop residue must be accurately quantified using the limited reflectance range of 900 nm to 1700 nm. Hence, novel unmixing methods must be developed. Two datasets were evaluated: a laboratory-acquired dataset and a simulated, atmospherically propagated dataset. Multiple unmixing methods were tested, including Linear Regression, a Bayesian Linear Dirichlet model, K-Nearest Neighbors, Random Forest, and deep learning approaches such as a Multi-Layered Perceptron. Strong performance was achieved on the laboratory dataset, with the Multi-Layered Perceptron achieving an R2 for crop residue of 0.8436, total R2 of 0.8935, and an RMSE of 0.0909 when plotting true to predicted abundances, demonstrating the feasibility of accurate unmixing in controlled conditions. However, performance decreased substantially on the atmospherically propagated dataset, likely due to nonlinearities and other stark differences between datasets that limit transfer learning. These findings indicate that while the lab results are highly promising, additional atmospheric measurements and model adaptations are necessary to achieve full confidence in FINCH’s predictions. Further testing and validation will be critical to establish robustness and guide the development of operational unmixing methods for determination of optical design and imaging requirements.

Understanding vegetation phenology in high-altitude regions is critical for assessing ecosystem responses to climate variability (Cleland et al., 2007). The Indus River Watershed in Ladakh (69,548 sq.km) spans elevations from 953m to 8,546m with diverse vegetation types adapted to extreme conditions. This study addresses the research question: How does vegetation phenology vary annually across 2018–2023? We employ satellite-based NDVI analysis to quantify phenological patterns, map spatiotemporal vegetation dynamics, and identify climate-driven changes in this data-sparse, high-altitude region.

The harvest mouse, Micromys minutus (Pallas, 1771) is the smallest rodent in Japan and now listed in the Red Data Books of Tokyo, 2 prefectural capitals, and 28 prefectures in Japan due to drastic decline of grasslands. For the harvest mouse, the height and density of the tall grass species where nesting occurs are considered particularly important. However, it has been difficult to continuously and extensively acquire information on the three-dimensional structure of herbaceous vegetation. With recent development of UAV technology, UAV data are beginning to be applied to the analysis of herbaceous vegetation. For acquiring three-dimensional information via UAV, methods include using LiDAR sensors or generating 3D point cloud data from aerial photographs using SfM. This study evaluates whether UAV LiDAR or UAV SfM is more suitable for estimating the height of tall grass species such as Japanese silver grass (Miscanthus sinensis), which serve as important nesting sites for the harvest mouse. As a result of analysis, the proposed method was found to be effective to estimate grass height regardless of whether UAV LiDAR or UAV SfM is used. However, when comparing the accuracy of canopy height estimation using UAV LiDAR data alone, UAV SfM data alone, and combined UAV LiDAR and SfM data, combined UAV LiDAR and SfM data found to perform best. Maximum canopy height was found to be best estimated using the combination of median of hand-measured five maximum canopy height values and maximum height calculated using the combined UAV LiDAR and SfM data.

Light pollution is an emerging global environmental issue driven by the intensified use of artificial light at night, with emissions increasing by nearly 50% in the last decades. In Ecuador, the rapid urban and industrial expansion of Guayaquil has led to a significant rise in nighttime radiance, raising concerns about its effects on nearby protected forests such as Cerro Blanco, Papagayo, Estero Salado, and Prosperina.

This study evaluates the impact of light pollution on local flora and fauna using VIIRS nighttime satellite imagery for 2014, 2019, and 2024. Average radiance values were processed in Google Earth Engine and classified into low, medium, and high pollution levels. Species occurrence data from iNaturalist and GBIF were integrated to identify taxa exposed to elevated light levels.

Results reveal a marked increase in light radiance, especially in areas adjacent to urban growth. In Cerro Blanco, radiance has intensified since 2017, disrupting natural light–dark cycles. Nocturnal and endemic species (such as Engystomops guayaco and Sylvilagus dauleensis) were identified among the most exposed, with potential alterations in reproductive, foraging, and behavioural patterns.

The study demonstrates that artificial light is encroaching upon protected ecosystems, threatening biodiversity and compromising ecological processes. The findings underscore the urgent need for conservation strategies that reduce light emissions, promote sustainable lighting technologies, and preserve natural darkness in nocturnal habitats. This work provides critical insights for the management of protected areas in Guayaquil and contributes to the broader understanding of light pollution impacts in megadiverse regions.

With the continuous intensification of marine plastic pollution, monitoring the transport and dispersion of microplastics has become a critical global concern. However, predicting microplastic concentrations remains highly challenging due to the lack of direct satellite signatures and the complex non-linear physical mechanisms governing their dispersion. This study develops an interpretable machine learning framework to monitor surface microplastic concentrations in the West Pacific from 2018 to 2020. We profoundly integrated Japanese AOMI in-situ microplastic observations, ERA5 meteorological/wave reanalysis, and Euro-Argo subsurface profile data utilizing a 3D Inverse Distance Weighting (3D-IDW) spatiotemporal interpolation algorithm. A Random Forest (RF) model was subsequently trained, achieving robust predictive accuracy (R² = 0.64, 0.76, and 0.87 for 2018, 2019, and 2020, respectively). Crucially, we incorporated SHapley Additive exPlanations (SHAP) to overcome the "black-box" limitations of traditional ensemble models. The SHAP analysis explicitly revealed a distinct, year-by-year regime shift in dominant environmental drivers: microplastic distribution was primarily governed by stable hydrographic and biological conditions in 2018; by dynamic wave forcing (e.g., long-period swells and Stokes drift) in 2019; and by extreme meteorological events (e.g., typhoon-induced terrestrial flushing) in 2020. Ultimately, this physics-informed framework successfully elucidates the dynamic transition of microplastic transport mechanisms between hydrographic–biological dominance and meteorological–physical forcing, providing vital scientific support for targeted pollution mitigation and coastal resilience planning.

Satellite data is crucial in regions lacking ground monitoring stations and is helpful in identifying areas likely to have high concentrations of pollutants harmful to human health. As these cities expand and grow, the quality of life and conditions will also be changing. The study seeks to determine the correlation between land surface temperature (LST), elevation, enhanced vegetation index (EVI), rainfall, particulate matter (PM2.5), nitrogen dioxide (NO2) and ozone (O3) both day and night during the months of October-December, January-February, June-August in the year 2019, 2020, 2024 and 2025. The results indicate a varied strength in relationship between variables in each season, day and night. During the day the highest negative correlation is obtained between elevation and carbon monoxide, while during the night the highest negative correlation is obtained between elevation and LST in all periods analysed. Results from the study thus indicate that it is critical to study the spatial and temporal variations of aerosols and temperature over varied conditions in a geographical region.

This study analyses the spatio-temporal dynamics of urban heat island (UHI) in Pune, focusing on the impact of Land use land cover (LULC) changes on the thermal environment. Using satellite imagery from the summer and winter seasons of 2015 and 2024, LULC, land surface temperature (LST), and geospatial indices were analysed at citywide and ward levels. Results indicate that from 2015 to 2024, the urban area increased by 12.77 km2, with the highest urbanization over Hadapsar. Between 2015 and 2024, Pune's mean LST and UHI increased by 8.18°C and 2.65°C in summer but dropped by 5.19°C and 0.54°C in winter. At a ward scale, during both seasons, the highest alterations in LST (UHI) were experienced at Sangam wadi ward. Among geospatial indices, NDMI was the most significant regulating LST across both seasons and years. Ward-level analysis for 2024 shows that a 1% rise in latent heat can lower UHI by 0.3°C in summer and 1°C in winter. Human thermal discomfort in the city is in the less comfortable category, with wards like Sangam wadi showing an increased discomfort across both seasons. The outcomes of this research can serve as a basis for decision-making to improve the resilience and sustainability of the region.

Wetlands are among the most dynamic and ecologically important ecosystems, yet they remain one of the least temporally monitored environments globally. Existing wetland datasets provide only static or low-frequency snapshots, making it impossible to track rapid hydrological fluctuations, disturbance events, and long-term degradation processes. To bridge this gap, we introduce GWD30, the first-ever global wetland dynamics dataset with near-daily temporal frequency (4-day interval) and 30-m spatial resolution, covering the period 2000–2024.

GWD30 is generated using a seamless data cube and a dynamic sample generation approach that converts static training labels into full time-series dynamic labels using temporal–spectral pattern embedding. A two-stage classification system combining machine learning and knowledge-guided refinement produces a globally consistent wetland taxonomy with 14 detailed classes.

This dataset enables unprecedented monitoring of wetland ecosystem behaviour across regions, timescales, and climate zones. GWD30 opens new opportunities for ecological modelling, biodiversity monitoring, hydrological analysis, climate research, and global conservation planning.

Deep-space exploration depends heavily on remote sensing as its primary data source, with the Moon and Mars serving as the main targets for scientific investigation and future human expansion. In these harsh planetary environments, rovers have become the essential platforms for surface exploration and sample acquisition. To support the development of next-generation rover systems, high-fidelity simulation environments are crucial. They enable safe, efficient, and repeatable testing of rover mechanics, perception, localization, and mapping algorithms under realistic planetary conditions, reducing mission risks and accelerating system development. This paper provides a comprehensive comparative analysis of existing lunar and Martian simulation environments, assessing them in terms of scene fidelity, rendering engines, supported robotic platforms, and intended application tasks. Building on this analysis, we introduce a generalized and reproducible workflow for constructing high-fidelity planetary simulation environments grounded in authentic remote sensing data products. Finally, we demonstrate the fidelity and practical utility of a state-of-the-art planetary simulation environment through a set of targeted validation experiments, followed by a discussion of key findings and future directions for the development of next-generation planetary simulation platforms.

The ability to perceptually interpret complex planetary surface environments is essential for successful robotic or crewed exploration. In this research, we present an online semantically-rich 3D information system that offers an immersive, high-fidelity simulation environment, accurately reproducing lighting and terrain conditions to support multi-disciplinary investigation of planetary surfaces. Built for both desktop and VR environments, it allows users to transition from conventional desktop analysis to fully immersive exploration, where spatial perception and cognitive engagement are significantly enhanced. Using candidate landing sites at the lunar south pole as case studies, we evaluate the performance of the proposed online semantically-rich 3D information system. Preliminary results indicate that the system enables users to interpret complex surface data more efficiently and intuitively than conventional observation methods.

Forage lichen fractional cover mapping using multi-spectral remote sensing (RS) data is challenging, especially over rocky landscapes where there is a high spectral correlation between lichens and non-lichen features. Given this, it is deemed that the use of airborne or satellite hyperspectral imagery may improve lichen mapping. In this study, we report the first results of using AVIRIS-NG hyperspectral imagery and UAV images to estimate forage lichen fractional cover (Cladonia spp.) in a rocky Canadian shield landscape where non-lichen bright features were prevalent. To estimate forage lichen fractional cover, we conducted a regression approach based on deep multi-layer perceptron (MLP) models whose number of hidden layers and neurons were determined using exhaustive grid search procedures. The three MLP models were trained and tested on four scenarios with different hyperspectral compression AVIRIS-NG band images and WorldView-3 (WV3) data of three sites. Our experiments showed that mapping lichen fractional cover using the 5 m AVIRIS-NG surface reflectance imagery was more accurate (i.e., higher R2 and lower RMSE values) than the one using a 4-band WorldView-3 (WV3) image with a spatial resolution of 2 m in most cases.

The catastrophic failure of the Córrego do Feijão Tailings Dam I in Minas Gerais, Brazil, on January 25, 2019, resulted in approximately 270 fatalities, underscoring the potential risks posed by tailings dams and the necessity for stringent monitoring of these structures. We utilized Sentinel-1 SAR data to generate deformation time series via Interferometric Synthetic Aperture Radar (InSAR) and to retrieve surface soil moisture (SSM), enabling pre- and post-failure analyses. InSAR analysis revealed significant pre-failure deformation within the tailings impoundment behind the dam crest, with a line-of-sight velocity reaching −69 mm/yr. In contrast, the post-failure period showed no significant ongoing deformation, indicating a relative stabilization. SSM showed a progressive increase in near-surface moisture, peaking on January 22, 2019, three days before the collapse. Although increased near-surface moisture alone cannot be a sign for liquefaction, continuous saturation might foster conditions prone to failure. We also proposed a spatiotemporal Graph Attention Network–Gated Recurrent Unit (GAT-GRU) technique to predict deformation time series derived from InSAR. The proposed GAT-GRU technique exhibited efficacy in predicting deformation trends by modeling spatial and temporal dependencies within the InSAR-derived time series. Overall, this study emphasizes the potential of InSAR, soil moisture analysis, and predictive models as reliable and complementary tools for managing tailings dam safety.

This paper presents a method and results for detecting, mapping and modelling the progression of wildfire in Australasia, Europe and North America within seconds. Initial detections are achieved using the BRIGHT algorithm (Engel et al., 2020, 2021). BRIGHT uses 10-minute Geostationary satellite observations, to dynamically threshold the satellite observation time stamp comparing it to the 28-day bioregion average at each respective timestamp. This produces a wildfire location (hotspot) and an estimate of FRP (Fire Radiative Power), Engel et al., 2022; Chatzopoulos-Vouzoglanis et al., 2022, within 20-45 seconds. Once detected, grouped fire locations are passed onto a fire behaviour simulator Spark / Inferno (Miller et al., 2015), to deliver a comprehensive bushfire analytics model framework which predicts fire behaviour.

At present this product is available in real time for Australia and is available on-demand for Europe and North America.

Low-lying and densely populated coastal cities are not only crucial areas for human survival and rapid development but also highly vulnerable regions sensitive to climate change. In recent years, tropical cyclone-induced flooding has emerged as a major hazard threatening the sustainable development of coastal cities. At the same time, rapid land use changes in these urban areas have significantly altered the original landscape structures and land use patterns, becoming key drivers of escalating flood risks. Therefore, when mapping and assessing urban flood risks, it is essential to comprehensively account for the combined effects of climate change and urbanization. This study uses Shanghai, a typical coastal city, as a case study to propose an integrated framework for simulating and evaluating coastal flood hazards while incorporating land use changes. The framework realizes the numerical simulation of flood disasters in coastal cities based on physical processes by coupling the SFINCS fast flood inundation model, the land use change model and the Delft 3D storm surge numerical nested model. The results indicate that by 2100, urban land use changes will expand the inundation area of a 1,000-year tropical cyclone flood by 4.91% to 34.00%. Neglecting future urban land use changes would underestimate the inundation extent of storm surges. Moreover, the findings highlight the critical need to account for the long-term impacts of land use changes on urban flood risks in coastal areas. The proposed methodology is applicable to coastal regions worldwide that are susceptible to tropical cyclones.

Advances in artificial intelligence have opened new frontiers in Earth observation, particularly in modeling complex geodynamic phenomena such as Vertical Ground Displacement (VGD). VGD is driven by numerous environmental and hydro-climatic factors, making prediction inherently challenging. This study develops a novel CNN-ConvLSTM hybrid deep-learning architecture that seamlessly integrates static soil characteristics and dynamic spatio-temporal features to predict VGD across the Italian territory. The model achieved an R2 value of 0.59 and a Mean Absolute Error (MAE) of 4.35 mm on the validation dataset, effectively capturing approximately 60% of the VGD variations. Additionally, Explainable AI (XAI) using SHAP (SHapley Additive exPlanations) values was incorporated to interpret the model's predictions. The analysis confirms that while hydro-climatic factors (such as drought and temperature) are the primary drivers of VGD temporal variability, static soil properties (including bulk density and volumetric water content) are the most globally influential predictors, dictating the overall spatial susceptibility of the medium. These findings provide a framework for identifying the key environmental drivers of VGD, which is essential for resource allocation, hazard management, and the development of effective early warning systems in geologically sensitive regions.

This study presents an integrated GIS-based and hydrological modelling assessment of flood vulnerability in the Yola South watershed of Adamawa State, Nigeria—an area experiencing increasingly severe flood events due to rapid urban expansion, land degradation, and intense rainfall. Using high-resolution spatial datasets, the watershed was delineated from a 30-m DEM, and land-use and soil information were utilized to compute Curve Numbers (CN) using the SCS-CN method. A composite CN of 65.33 was derived, indicating moderate infiltration capacity and substantial susceptibility to runoff generation during heavy storms.

The HEC-HMS hydrological model was used to simulate the July 31, 2025, rainfall event across delineated sub-basins. Model outputs revealed peak discharges ranging from 5.3 to 6.0 m³/s, direct runoff volumes of approximately 19 mm, and lag times between 205 and 266 minutes. Sub-basins with increased imperviousness and exposed soils generated faster and higher runoff responses, identifying hydrological hotspots that contribute disproportionately to downstream flooding.

The study demonstrates the utility of combining GIS with HEC-HMS simulation to evaluate watershed behaviour under current land-use conditions. Findings provide actionable guidance for flood risk planning, including targeted drainage improvements, land-use regulation, and nature-based solutions such as vegetation restoration. This research highlights the value of geospatial technologies in supporting climate resilience and aligns with ISPRS priorities on sustainable environmental management and hazard mitigation

This study presents a Transformer-based framework for estimating landslide occurrence dates using fused Sentinel-1 SAR and Sentinel-2 optical time-series data. By leveraging multivariate temporal features and long-range attention, the model substantially improves dating accuracy compared with single-sensor methods, with over 80% of events dated within a 0–15-day offset. The derived occurrence dates enable the creation of landslide activity maps at daily temporal resolution, offering a major advancement over conventional annual assessments. The resulting landslide activity index highlights spatial and temporal variations in slope activity, supporting more precise identification of highly dynamic landslides. The framework offers a valuable tool for monitoring slope hazards and enhancing landslide risk assessment at regional scales.

Floods remain one of the most frequent and disruptive natural hazards in Sri Lanka, particularly within low-lying monsoon-driven basins where cloud-covered conditions hinder optical monitoring. This study develops a multi-temporal flood-mapping framework that integrates Sentinel-1 C-band SAR data with DEM-derived terrain information to assess flood dynamics and identify recurrent inundation zones in the Attanagalu Oya Basin. Three major flood events in 2016, 2017, and 2018 were analysed using pre- and peak-flood SAR acquisitions processed through a standard workflow of orbit correction, radiometric calibration, speckle filtering, and terrain correction. Threshold-based segmentation of VV backscatter (−15 to −13 dB) was applied to delineate inundation, followed by hydrologically guided refinement using slope (<1–3°) and elevation constraints to reduce false positives associated with built-up areas, vegetation, and radar shadow. The results illustrate distinct spatial variations across the three years, with 2016 showing the most extensive inundation and 2018 presenting spatially concentrated flooding. DEM integration significantly improved classification accuracy by eliminating physically implausible detections. Validation against the Survey Department–Sri Lanka Navy inundation map for 2016 produced a spatial agreement of 72.18%, demonstrating the reliability of the SAR–DEM fusion approach. Multi-year overlay of flood layers revealed a persistent high-risk corridor stretching from Gampaha to Katunayake, reflecting entrenched drainage limitations and ongoing floodplain encroachment. The proposed framework provides an operational, scalable method for flood monitoring in cloud-prone environments and offers essential insights for risk-sensitive land-use planning, hazard zoning, and infrastructure design. The approach also forms a basis for future automated, machine-learning–enhanced flood early-warning systems.

Climate change has intensified the variability and frequency of extreme events, particularly affecting vulnerable coastal urban

areas. In Caraguatatuba, located on the north coast of S˜ao Paulo, the impacts are exacerbated by social inequalities and inadequate

infrastructure. However, detailed analyses of climate indices in the region are still scarce. This study analyzes precipitation patterns

in Caraguatatuba, calculating climate indices to identify trends and support risk mitigation. CHIRPS daily precipitation data at

0.05° spatial resolution were processed using Google Earth Engine and R to compute the ETCCDI indices PRCPTOT, CDD, CWD,

R95p, R99p, Rx1day, and Rx5day. Statistical tests were applied to detect significant trends. The results indicate an increase in

consecutive dry periods (CDD, p=0.0065) and at the intensity of daily extreme rainfall (Rx1day, p=0.0044) in the southwestern

area, while other indices did not show significant trends. These findings highlight the city’s increasing climate vulnerability and the

urgent need for adaptation strategies. By offering a replicable framework based on open-access remote sensing and cloud platforms,

this study supports policy development to enhance urban resilience and monitoring climate-related disasters in coastal cities.

Agricultural drought is a recurring and complex hazard that significantly impacts food security, water resources, and rural livelihoods, particularly in vulnerable regions such as the southern agro-climatic region of Tamil Nadu, India. Over the past few decades, the region has experienced numerous drought-related phenomena. In this context, the primary objective of this study is to monitor the dynamics of agricultural drought across the region between 2015 and 2023, with the aim of promoting sustainable agricultural management and climate-resilient practices.

MODIS NDVI and LST products were utilized to derive drought-related indices. Rainfall data from CHIRPS and temperature data from TerraClimate (1982–2023) were used to calculate the SPEI, enabling the identification of representative dry and wet years. The analysis uses the Normalized Vegetation Supply Water Index (NVSWI), derived from remote-sensing time-series datasets. Finally, a correlation analysis was conducted between monthly NVSWI, one-month SPEI, and VHI during the primary growing season of Rabi crops (October to December).

Results reveal that NVSWI identified spatial and temporal patterns of acute water stress in vegetation. Ramanathapuram, Sivaganga, Virudhunagar, Thoothukudi, and Pudukkottai were consistently drought-prone, experiencing moderate to severe drought intensity in multiple years. In 2021, drought conditions were minimal. A strong positive correlation was observed between monthly NVSWI and both SPEI and VHI, confirming its suitability for drought monitoring.

The findings highlight the effectiveness of NVSWI and multi-source satellite data for drought detection, supporting the development of early warning systems and climate-resilient agricultural planning in drought-prone regions.

The deglaciation due to global warming and changing climate has given rise to the formation and expansion of numerous glacial lakes, particularly in the High Mountain Asia region. Many of these glacial lakes are susceptible to experiencing Glacial Lake Outburst Floods (GLOFs) events which can release millions of cubic meters of water and debris, leading to widespread impacts on lives, property, infrastructure, agriculture and livelihoods amongst remote downstream communities. The research investigates the potential of multi-source data, focusing on District Chitral in Northern Pakistan, with elevated potential implications for GLOF and associated risk. A total of 12 vulnerable sites are identified, out of which 5 are highly susceptible to GLOF. A spatio-temporal analysis of the vulnerable sites have been carried out in Google Earth Engine (GEE). The maps were generated in the GIS environment of ArcMap considering key contributing factors with high impact potential including, lake area change, elevation, slope, aspect, temperature and precipitation, LULC, change in snow and glacier cover area, distance from fault line, and proximity to impact area, among others. A pronounced decline in the snow and glacial cover, and an increase in land surface temperature (LST) retrieved from satellite data could be responsible snow/glacial melting resulting to higher frequency of GLOFs and flash floods. The potential implications on population, infrastructure, schools, forest and agriculture, and water quality of Chitral have been estimated. The findings are of great significance for policymakers and disaster management authorities, providing valuable insights to formulate efficient and effective measures for mitigating the risks.

Remote sensing and GIS-based geomorphometric mapping are powerful tools for analyzing neotectonic activity. This study focuses on the Nanga Parbat Syntax (NPS) and its adjoining regions, among the fastest uplifting zones of the Himalayas, rising at 8–10 mm/year. Using SRTM DEM, Trend Analysis of Polynomial Surfaces (TAPS), Local Base Level (LBL), and Vertical Dissection (VD) maps were produced to interpret surface dynamics. The study area, located along the Main Mantle Thrust (MMT) and below the Main Karakoram Thrust (MKT) in Gilgit-Baltistan, encompasses five key geomorphometric zones—two defined by drainage dissection, two by relative relief, and the expansive, relatively flat Deosai plateau in the southeast with prominent VD signatures. A residual elevation map was derived by subtracting a 12th-order polynomial trend surface from the DEM, highlighting spatial elevation anomalies. This trend surface effectively captures the NE–SW and NW–SE uplift patterns across the Nanga Parbat Haramosh Massif Zone (NPHMZ). Elevated anomalies align with the Sassi Raikot Fault Zone (SRFZ), NPHMZ, MKT, and northwest of Jaglot toward the Hindukush. In contrast, areas such as Deosai Plateau, Skardu, Kachura, Gorikot (Astore Valley), Jaglot, and Gunar exhibit negative elevation anomalies. LBL maps generated from 2nd- and 3rd-order Strahler streams yielded insightful correlations with tectonic structures. Both isobase and VD analyses indicate significant dissection and elevation in regions near the NPHMZ, MKT, and upper Astore Valley. The spatial alignment of high residuals with these structural features underscores active tectonic uplift, reaffirming ongoing neotectonic processes across the NPHMZ and its surroundings.

The frequency of wildland fires are increasing due to warmer and drier conditions resulting from climate change. Identifying fire prone areas is essential for planning and mitigating potential impacts. This study aims to create a wildland fire danger map using Random Forest (RF) and competing classification methods for Ontario’s Managed Forest (MF). The critical role of the classification method in wildland fire danger mapping motivated us to evaluate and compare the effectiveness of three classification methods, including Natural Break, Geometric Interval, and Standard Deviation Interval. A total of 42 key static and dynamic variables were analyzed, covering the period from 2020 to 2022. The static variables: distance from roads, railways, settlements, rivers, and water bodies and topographic features like elevation and various indices derived from the NASA Digital Elevation Model (NASADEM). To capture these dynamic environmental conditions, several environmental variables and indices as well as key meteorological parameters were incorporated into the modelling from MODIS and ERA5 land. We used Recursive Feature Elimination and Cross-Validation (RFECV) to select optimize features for the model. To address the opacity inherent in machine learning models, SHapley Additive exPlanations (SHAP) were utilized to quantify the marginal contribution of each variable to the predicted distance from fire. Our results showed that the GI classifier provided the most consistent and well-balanced performance and reliable predictions across all evaluation metrics. The resulting fire danger map highlights high-risk areas, supporting targeted management, prevention, and resource allocation to reduce future wildfire impacts.

Dharali, a Himalayan village in Uttarkashi, Uttarakhand, lies along the narrow Kheer Ganga valley, a terrain marked by steep slopes, high relief, and complex topography- conditions that render it highly susceptible to geomorphological and hydrological hazards. On August 5, 2025, a catastrophic flash flood and debris flow, triggered by a sudden cloudburst or upstream slope failure, caused extensive destruction across 0.54 km² of the settlement. Despite the increasing frequency of such high-magnitude events in the Himalaya, quantitative assessments of recent localized geomorphic and hydrological impacts remain limited, particularly in small, high-altitude villages like Dharali. This study addresses that gap by employing high-resolution PlanetScope imagery (3 m) from pre-event (July 19, 2025) and post-event (August 22, 2025) periods to detect and quantify surface alterations using spectral thresholding and spatial change detection methods. The analysis revealed pronounced spectral shifts, a 39% increase in surface water extent, and topography-driven hydrological redistribution. The statistical association (𝜑 = 0.35; ϗ = 0.34) indicates moderate spatial agreement between the temporal datasets. The findings demonstrate the utility of fine-resolution satellite data in capturing rapid, small-scale landscape transformations and emphasize the urgent need for systematic, event-based monitoring frameworks to improve disaster preparedness and resilience in fragile Himalayan environments.

This paper presents PICANTEO, a modular and multi-modal change detection framework designed for remote sensing applications in natural disaster response. The framework aims to support damage assessment during both the rapid mapping phase, which occurs in the immediate aftermath of a disaster, and the longer recovery phase. PICANTEO provides automated, reliable disaster-related change detection maps and associated impacted areas to support a wide range of disaster monitoring activities. The integration of uncertainty and ambiguity concepts ensures reliable and qualified results.

PICANTEO handles multi-modal remote sensing data, including very high-resolution optical imagery, Digital Surface Models, and Synthetic Aperture Radar (SAR) data. Its modular architecture enables users to apply ready-to-use pipelines or implement their own workflows. The provided scalable components can be combined or extended by custom methods to define new applied pipelines. Several real-world case studies demonstrate PICANTEO’s ability to address various disaster scenarios across diverse geographic contexts. Source code is available at: https://github.com/CNES/picanteo.

Coastal regions are the foci of intense economic activity, but, these dynamic and ecologically sensitive low-lying lands are increasingly threatened due to climate change induced eustatic sea level rise, and anthropogenic activities at regional and local scale leading to relative sea-level rise, thereby necessitating to understand the vulnerability of a coast for their protection and sustainable development. In this background, the present study is focused along the densely populated ~590 km long coastal stretch of Kerala state, India to build an Integrated Coastal Vulnerability Index (ICVI) by using the i) physical vulnerability index (PVI) variables such as a) Geomorphology, b) Coastal Slope, c) Bathymetry, d) Shoreline change history, e) Spring tide range, and f) Significant wave height, and ii) socio-economic vulnerability index (SVI) variables like a) population density, b) land use/land cover, c) number of household, d) fisher-folk population density, e) literacy rate, f) occupation, g) road density, h) railway network and i) tourist spots integrated in Geographic Information System (GIS) environment, and through Analytic Hierarchy Process (APH) technique. Kerala state is located on the south-western margin of the Indian Peninsula, with nine administrative districts i.e., Kasargod, Kannur, Kozhikode, Malappuram, Thrissur, Ernakulam, Allapuzha, Kollam and Thiruvananthapuram districts (from North to South). The Integrated Coastal Vulnerability Index (ICVI) along the Kerala coast revealed that 96.82 km (16.4%) under very low vulnerability, 105 km (18%) under low vulnerability, 145.75 km (24.6%) under moderately vulnerable, 123 km (20.9%) under highly vulnerable and the remaining 119.27 km (20.2) under very highly vulnerable.

This study examines how wildfire disturbance changes urban runoff behaviour in a small Mediterranean basin. The research uses high-resolution UAV images together with GIS processing to create detailed surface models and land-cover maps after the fire event. These spatial datasets were then integrated with SWMM and HEC-RAS 2D to simulate different rainfall scenarios and to understand how the basin responds during storms.

The results show that burned areas have faster runoff, higher peak flow, and stronger surface connectivity even under moderate rainfall. The modelling framework was able to map sensitive zones where water tends to accumulate and where the risk of local flooding increases. The study also demonstrates that UAV-based photogrammetry can improve hydrological and hydraulic simulations by providing more accurate information on terrain shape and surface conditions.

Overall, this contribution presents a practical workflow that combines remote sensing and process-based modelling to support flood-risk assessment in fire-affected urban environments. The approach is suitable for other regions with similar challenges and contributes to ISPRS goals of using geospatial technologies for climate-resilient and sustainable water-management planning.

Digital elevation models are the primary data used in remote sensing and geographic information systems (GIS) for terrain analysis and three-dimensional spatial data processing [1]. In surveying and mapping investigations, SRTM1 (Shuttle Radar Topography Mission) DEM, ASTER GDEMV3 (Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Digital Elevation Model), and AW3D30 (ALOS World 3D-30m) DEM have become key data sources. This study compares the elevation accuracy of three open-source DEM datasets to high-precision ICESat-2/ATLAS altimetry data. GIS statistical analysis, error correlation analysis, and mathematical statistical approaches are used to compare elevation accuracy in DEM.

Four error assessment metrics are utilized: Mean Error (ME), Standard Deviation of Error (SDE), Root Mean Square Error (RMSE), and Correlation Coefficient (CORR). Furthermore, error correlation analysis is conducted to visually characterize the spatial distribution patterns and error features of the three open-source DEMs in relation to ICESat-2/ATL08 observations.

The AW3D30 DEM has the highest accuracy in plains, with the SRTM1 DEM coming in second. In mountainous terrain, SRTM1 DEM was the most accurate, followed by AW3D30 DEM. Although ASTER GDEMV3 fared less well in the two study locations listed above than previous studies in plateau regions, its accuracy in mountainous and plateau areas is comparable to that of SRTM1 DEM and AW3D30 DEM.The RMSE for all three DEM datasets is roughly 15 m in wooded mountainous regions, around 5 m in artificial surfaces and barren areas, and exhibits the greatest inaccuracy in forested and grassland regions on plains, with the least error occurring in wetlands.

The radiometric adjustment of aerial imagery is a process of very high importance considering the influences this step can generate not only on the look of the image data (white balance), but even more importantly on derived information like indices (NDVI). In comparison to the Aerial Triangulation where it is relatively straight forward to set up thresholds that need to be met to achieve a high-quality result, the world of radiometric adjustment is dramatically different. There is no single standard or guideline that dictates what a high-quality radiometric result will look like. Apart from these challenges there is also a rather big gap between the rich and longstanding academic work done in the field of radiometry and actual application in real-life projects. The by far biggest discrepancy is the usage of single images, especially when dealing with absolute radiometry approaches versus multiple thousand color balanced images in a single block in an actual production environment. In this paper we present the advantages of utilizing reflectance measurements as a method to stabilize radiometric adjustments, as well as utilizing them as anchor to create indices like the NDVI that actually correspond to the value range given by literature.

Urban Spatial Monitoring (USM) data is a new form of fundamental geographic information product. As an important component of the natural resources survey and monitoring system, USM provides strong support for the construction of the China Spatial Planning Observation Network (CSPON). The quality of USM data results is related to the accuracy of statistical analysis outcomes, the scientific decision-making for national economy, people's livelihood and social development, as well as the reliability of natural resources management applications. Based on the analysis of the characteristics of USM results, this paper proposes an inspection process including overall general inspection, detailed inspection, and out-of-sample general inspection; analyzes inspection parameters and items, and determines the inspection methods for different items; further identifies quality elements that can be automatically inspected by programs, and realizes batch automatic inspection of some items by establishing a rule base; then, studies the implementation methods of other quality inspection items, clarifies common problems, and improves the human-computer interactive inspection for quality items. Finally, the quality inspection results of urban space monitoring data products covering 170 prefecture-level city survey areas spanning approximately 6 million square kilometers demonstrate that the technical route proposed in this paper is feasible and the quality evaluation results are objective.

Seagrasses are ecologically-important yet highly threatened blue carbon ecosystems that play a critical role in environmental protection, biodiversity conservation, and carbon sequestration. However, their spatial heterogeneity and dynamic temporal behavior pose challenges to accurate mapping and long-term monitoring. The availability of publicly accessible satellite images with high spatial and temporal resolution, and advances in machine learning, have gradually expanded seagrass geospatial research and led to more accurate and robust image classifications. This study evaluated the performance of traditional and machine learning methods for seagrass and benthic habitat mapping using clear and sun-glinted 3-meter resolution PlanetScope imagery. Classification accuracy metrics were compared across multiple algorithms and varying image quality, using two different reference datasets. Results indicate that the Maximum Likelihood Classification and Support Vector Machine Classification achieved the highest overall accuracy and kappa statistics for the clearest image used, the 8-band PlanetScope image acquired on February 16. As expected, the application of the sun glint correction procedure improved classification accuracies for lower-quality images, particularly for the Random Forest Classification which showed consistent and pronounced gains after deglinting. These findings demonstrate the potential of PlanetScope images for seagrass and benthic mapping, keeping in mind that careful image selection remains essential due to the imagery’s inherent sensitivity to sun glint and other radiometric inconsistencies affecting classification performance. In the absence of optimal or clear images, scenes with lower image quality may still be effectively utilized with the application of radiometric correction procedures such as sun glint removal.

The National Ecological Observatory Network (NEON) operates the Airborne Observation Platform (AOP) which collects airborne lidar, imaging spectroscopy and RGB camera information to support the characterization and forecasting of environmental and environmental processes. NEON operates at a continental scale, including observations at sites across the continental Unites States, Alaska, Hawaii and Puerto Rico, and will operate for 30 years. The AOP has been collecting data at sites with NEON for over 10 years, representing a highly valuable resource for conducting ecological change analysis. NEON AOP data highly standardized, and undergoes rigorous quality control and quality assurance processes. Data collected by the AOP is processed into a series of data products that are made freely available for educational and scientific endeavors. This presentation details the current status and future plans for NEON's AOP.

Peatlands are a major component of Canada’s boreal region and play critical roles in carbon storage, biodiversity, hydrology, and climate regulation. Different peatland types respond differently to climate-driven changes in temperature, precipitation, and wildfire risk. Accurate maps of these sub-classes are essential for conservation planning, carbon accounting, and wildfire management. Although national-scale wetland maps for Canada have advanced in recent years, many lack detailed peatland sub-class information and often omit swamps.

This research builds on recent efforts by expanding the spatial extent of peatland sub-class mapping across Canada (excluding the Northern Arctic and Arctic Cordillera) and explicitly incorporating swamps as a separate class. A three-stage hierarchical framework was developed using a combination of optical, radar, and terrain-derived variables. Predictor datasets included Landsat spectral mosaics, NDVI harmonic coefficients, canopy height and closure, ALOS-2/PALSAR-2 L-band backscatter, seasonal Sentinel-1 coherence, and hydrological and geomorphometric derivatives from FABDEM. Reference data were compiled from multiple validated wetland inventories.

A Random Forest classifier was trained and validated at each stage: (1) wetlands vs. uplands and water, (2) peatlands vs. mineral wetlands, and (3) peatland sub-classes and swamps. Accuracy exceeded prior national efforts, with 87% accuracy at Stage 1, 94% at Stage 2, and 72% at Stage 3. Shapley Additive Explanations showed that the SAGA Wetness Index was consistently among the most important predictors, highlighting the central role of topography and moisture distribution. These results demonstrate the value of integrating multi-sensor remote sensing with terrain metrics to improve national-scale wetland classification.

Mangrove forests along Hong Kong’s densely populated subtropical coastline fulfil significant blue carbon storage, shoreline protection and habitat functions but are vulnerable to hydroclimatic and anthropogenic pressures. This study integrates long-term satellite spectral analysis with field-based laser scanning to examine mangrove canopy dynamics and above-ground biomass. Seasonal composites of Landsat-8 imagery (2013 to 2025) were processed in Google Earth Engine, where NDVI, EVI, ATVI and GEMI were calculated. GreenValley DGC-50 SLAM-based backpack laser scanning system to collect plot-scale structural data. We registered, denoised, normalized and segment point clouds to retrieve tree height, crown diameter and diameter at breast height for being further used in species specific allometric equations to estimate biomass. The spectral time series indicates a persistent greening pattern with recurrent seasonal cycles and stronger canopy development after 2018. Comparison with field observations showed that laser-derived tree height was more consistent than DBH and crown diameter, indicating variable parameter accuracy in dense mangrove stands. The LiDAR survey provides valuable detailed structural information and supports biomass estimation for inaccessible areas. The LiDAR survey provides detailed structural information and supports the estimation of biomass, where traditional measurements are hard to obtain. The field survey does not validate long-term spectral trends but rather serves a contemporaneous structural reference frame for interpreting seasonal and interannual spectral variability. The combined framework enables enhanced estimation of mangrove biomass, blue carbon stock monitoring and coastal ecosystem management in Hong Kong.