Proper use and replenishment of groundwater can help solve various problems, making it crucial to have a scientific understanding of its management. For effective management of groundwater systems and to preserve water quality, it is essential to identify the potential zones for groundwater recharge. Geographic Information System (GIS) methods and remote sensing technologies are utilized to identify areas with potential for groundwater. The integration of remote sensing (RS) and geographic information system (GIS) techniques is employed in this study to develop a standardized methodology for assessing groundwater potential. Groundwater potential map of the Gadilam river basin is generated using GIS technologies. The precise data is generated using satellite imagery and Survey of India (SOI) toposheets at a scale of 1:50,000. This data is utilized to identify parameters relevant to determining groundwater potential zones, including rainfall, geological structure, soil characteristics, slope, drainage density, geomorphic units, lineament density, and land use/land cover. It is then integrated with a weighted overlay in ArcGIS. Weight factors are established for the various geomorphic units based on their ability to store groundwater. This process is repeated for each subsequent level, and the resulting layers are reclassified. In weighted overlay analysis, different thematic layers are assigned a theme weight and class rank. The findings were validated through fieldwork. The resulting map was categorized into three levels: Poor, Moderate, and Excellent. The application of the suggested methodology to the Gadilam river basin is demonstrated. This study will be helpful for useful groundwater management for different area.

Flood hazard assessment—together with vulnerability and risk analysis—is closely linked to flood resilience and has been extensively studied in densely populated areas, where the most catastrophic floods tend to occur. The need for a holistic and transferable methodology is critical considering that, different simulation approaches are often used, while key methodological phases are sometimes omitted. Within the framework of the Programming Agreement of the Prefecture of Attica, the BEYOND Centre (IAASARS/NOA), in cooperation with the NTUA research group have developed the methodology presented in this work. The methodology was implemented at high spatial resolution in five flood-affected river basins in Attica, with the Pikrodafni River basin being presented in detail in this study. Data acquisition constituted a core component of the methodology and involved targeted spatial datasets, Earth-observation imagery, time-series data, historical flood records, and relevant prior studies obtained from the competent authorities. Field visits were conducted to characterize site conditions and verify the collected datasets, identifying high-risk critical points, and measuring the dimensions of hydraulic structures (bridges, culverts) and channel properties. Regarding modeling, design-flood scenarios with typical return periods were analyzed in accordance with the Directive 2007/60/EC. HEC-HMS was used to generate hydrographs for each sub-basin, which were then imported into the quasi-2D LISFLOOD-FP model as a means to prepare and calibrate the HEC-RAS model, where a rain-on-grid methodology integrated the hydrologic and hydraulic flood processes at the area of interest. High spatial resolution was maintained throughout, with particular emphasis on uncertainty analysis and on the detailed representation of infrastructure and urban areas, given their strong influence on flood dynamics. Results indicate that overflow typically occurs in buried streams, along adjacent roads in the downstream reach of the river, at stream confluences, and at the upstream inlet where natural streams enter the drainage pipe network. Up to 200 critical points were identified, of which up to 35% were classified as first-priority sites for intervention.

Indonesia is among the most landslide-prone countries in the world, with thousands of fatalities and widespread infrastructure damage recorded over recent decades. Despite this high hazard level, regional-scale landslide monitoring remains constrained by the limitations of conventional bitemporal satellite imagery, which is susceptible to cloud contamination, dependent on precise acquisition timing, and unable to capture the full temporal dynamics of landslide occurrence and recovery. While the LandTrendr (Landsat-based Detection of Trends in Disturbance and Recovery) algorithm has been widely applied for detecting vegetation disturbances such as forest loss and land-use change, its potential for landslide detection in tropical environments has not been sufficiently explored. This study aims to evaluate the applicability of LandTrendr applied to long-term Landsat time series imagery for automated regional-scale landslide detection and mapping in Indonesia. The method integrates temporal segmentation of the Normalized Difference Vegetation Index (NDVI) derived from Landsat imagery spanning 2000–2022 with slope information from the Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM) to identify the characteristic drop-recovery spectral signature associated with landslide events. The algorithm was applied and evaluated in two geologically distinct study areas: Lombok, West Nusa Tenggara, and Pasaman, West Sumatra. Detection accuracies of 25.9% by location and 20.3% by area were achieved in Lombok and 76.3% by location and 85.3% by area in Pasaman. The lower accuracy in Lombok is primarily attributed to the predominance of small landslides below the sensor’s spatial resolution and rapid vegetation recovery. The proposed approach demonstrates the unique capability of LandTrendr to model the entire life cycle of a mass movement event, from pre-event stability through abrupt disturbance to ecological recovery within a single unified framework, providing a scalable and cost-effective tool for long-term landslide monitoring applicable to other tropical, landslide-prone regions.

This article investigates the 2022 Yahidne war crime, during which Russian forces confined approximately 368 civilians, including 69 children, to the basement of the local school. Drawing on Jacques Derrida’s concept of hauntology, the authors explore satellite images of unfolding events and the enduring material traces of the occupation—drawings, abandoned toys, military rations, propaganda newspapers, spent military equipment and damaged infrastructure. They consider how these traces contribute to processes of collective memory and to the transformation of the site’s significance through public memorialisation, reflecting on the role of contemporary archaeology in documenting and interpreting material legacies of recent conflict.

ABSTRACT Phenological metrics are a set of measurements obtained from Earth observation (EO) satellite image time series that allow the estimation of phenological stages. These include indicators like the start of the greening season, the onset of senescence, and the growing season length. They are useful for crop monitoring. Today, large volumes of images are produced and made available by different EO satellites. These large EO data sets pose a challenge for storage and processing systems, exceeding the capacity of personal computers to handle them. This paper presents a free and open-source tool for phenological metrics analysis from large EO image collections that runs on server-side infrastructure and does not require local data downloads. The Web Crop Phenology Metrics Service (WCPMS) is the core of this tool, designed to estimate phenological metrics as a web service. The tool extracts phenological metrics associated with spatial locations, based on the Brazil Data Cube (BDC) platform. It calculates phenological metrics from data cubes of distinct remote sensing image collections. The potential of the tool is shown through an experiment estimating soybean sowing dates using phenological metrics compared with field data obtained in the Central-South region of Brazil.

ABSTRACT The amount of Earth observation data is getting larger day by day. This rapid evolution of the field and the exponential growth of data generation mediums urges the need for modern ways of fast analysis and processing of satellite images. Earth Observation Data Cubes enhance the way Earth data are distributed, handled, stored, and analyzed by providers and users. In this study the development of the National Earth Observation Data Cube of Cyprus is described and the importance of such software-based infrastructure is demonstrated through a water resource monitoring use case in the context of climate crisis in the Mediterranean region. Moreover, this study exhibits the capabilities of semantic classification in Earth observation, introducing a spatially generalized approach for easier environmental monitoring. Such developments can enhance the preservation of the environment and its protection from future natural and human disasters.

Rapid assessment of hazardous aerosol dispersion is critical for emergency response, yet operational dispersion workflows can exhibit end-to-end latency that is incompatible with the first minutes of decision-making. This study develops and validates a deep learning approach for near-real-time nowcasting of volcanic ash dispersion from geostationary observations. The model was trained on an archive of volcanic ash satellite imagery from EUMETSAT's SEVIRI instrument (Ash RGB composite) and achieved a structural similarity index of 0.88 for 15-minute next-frame forecasts. The complete edge workflow, including data download and inference, runs in under five seconds on an NVIDIA Jetson AGX Orin. To illustrate how the same nowcasting pipeline can be used for hypothetical scenario exploration across particulate sources, a pixel-based event-injection algorithm is introduced to overlay synthetic plumes of varying sizes into real-time satellite frames before inference. Scenario demonstrations parameterized by nuclear-yield-inspired sizes (10kt to 100Mt) are presented at urban (Paris, London, Berlin), national (Iberian Peninsula), and continental (Europe-wide) scales. These scenario outputs are intended as illustrative, low-latency visualizations of kinematic transport patterns in the SEVIRI observation space, as validated predictions of nuclear plume morphology. The primary contribution is a fast, low-cost volcanic ash nowcasting system, complemented by a generalizable injection framework for rapid scenario visualization on edge computing.

The conservation reserve program (CRP) plays a vital role in preserving ecologically sensitive lands across the United States by encouraging the voluntary conservation of marginal farmland. Despite its significance, limited tools exist for nationwide continuous monitoring and assessment of hydrological conditions within CRP lands. This study conceptualizes surface water inundation as an indicator of hydrologic connectivity and ecological function, reflecting how water dynamics influence CRP sites resilience. Our study presents a novel, scalable approach to assess inundation dynamics across 1.3million CRP sites from 2018 to 2024 using a synergistic framework that integrates Sentinel-2 satellite imagery, Dynamic World land cover data, and machine learning classifiers (Support Vector Machine, Random Forest, CART) within Google Earth Engine. Our framework integrates spectral water indices (NDWI, MNDWI and NDMI) with dynamic world land-cover classifications to generate quarterly inundation maps at 10-m spatial resolution, enabling consistent identification of surface-water extent across time. The SVM model achieved best performance in surface water detection. Our analysis shows that the core areas of wetland-related CRP lands consistently hold water seasonally acting natural sponges for agricultural water quality improvement, groundwater recharging and flood mitigation. While the CRP program effectively maintains the inundation performance of wetland-related lands, it also increases inundated areas on non-wetland-related lands, highlighting the valuable contributions of CRP lands to the agricultural landscape nationwide. CRP sites with longer enrollment periods exhibited higher and more stable inundation, indicating improved wetland functionality. Overlays with the national wetlands inventory (NWI) and hydric soil data confirmed strong links between site characteristics and surface water, particularly in the Midwest and Lower Mississippi River Basin. Spring and summer emerged as key periods of persistent or episodic inundation, providing critical habitats for migratory birds and supporting biodiversity. This research provides a long-term hydrological monitoring approach for CRP lands and supports targeted conservation decision-making to identify and implement best practices. By identifying high-priority CRP sites for restoration and demonstrating the utility of real-time, high-resolution remote sensing data, this study provides a valuable foundation for adaptive management and policy strategies that enhance the resilience and functionality of CRP lands.

This study presents the development of a Python-based flood-susceptibility risk-mapping tool, implemented in Jupyter Notebook, applied to Rwanda. A Flood Susceptibility Index (FSI) was developed by integrating 20 causal factors associated with flood occurrences, including topographic, hydrological, geological, and anthropogenic variables. Logistic regression, and Variance Inflation Factor were implemented in Python using libraries such as Numpy, Arcpy, traceback, scipy, Pandas, Seaborn, and statsmodel to assign weights to each factor, and to address multicollinearity. The model was validated against flood extent data derived from Sentinel-1 satellite imagery for the major historical flood event that occurred from 2014 to 2024, ensuring spatial consistency and predictive reliability. To project future flood susceptibility for 2030, precipitation data from the Institut Pierre Simon Laplace Coupled Model, version 5A, Medium Resolution (IPSL-CM5A-MR) climate model under the Representative Concentration Pathway 8.5 (RCP 8.5) scenario were utilized. The resulting FSI was classified into five susceptibility levels, from very low to very high, and visualized using Python’s geospatial and plotting tools within Jupyter Notebook in ArcGIS Pro 3.5. It indicates that areas with high amounts of rainfall, and proximity to wetlands and rivers reveal the highest flood risk. The automated and reproducible approach offered by Python enhances transparency and scalability, providing a decision-support tool for disaster risk reduction and climate adaptation planning in Rwanda.

Building image segmentation is a critical task in urban planning, disaster management, and environmental monitoring. Traditional segmentation methods using either satellite or drone imagery face challenges such as resolution limitations and vegetation occlusion. In this study, a novel approach integrating multi-source remote sensing data-combining drone imagery captured at multiple heights (150m, 200m, 250m, 300m) with multi-view perspectives combined with satellite imagery to enhance building segmentation accuracy. A key challenge in segmentation is the interference of vegetation and shadows, which can obscure building boundaries. To address this, advanced vegetation removal techniques have been incorporated to refine the extracted structures. By leveraging the complementary advantages of drone and satellite imagery, this approach improves segmentation robustness and reliability. Our method achieves a segmentation accuracy of 96%, significantly outperforming conventional techniques. This approach has been evaluated against existing segmentation methods, demonstrating its effectiveness in extracting high-precision building footprints, even in complex environments. The findings highlight the potential of integrating multi-source data and vegetation suppression strategies to enhance automated urban feature mapping.

Digital Twins for cultural heritage buildings commonly depend on high-fidelity 3D scanning, detailed onsite surveys, and unrestricted data acquisition. In many countries, however, legal, regulatory, and conservation constraints render such methods inaccessible or explicitly prohibited, significantly limiting the deployment of digital-heritage technologies in real settings. This paper introduces HeritageTwin Lite, a regulation-compliant workflow for constructing low-detail yet operational Digital Twins of protected cultural heritage buildings using only publicly permissible data sources. The proposed approach relies on a GIS-based 2D application through which users select a site and manually delineate building footprints and structural outlines. These 2D sketches are combined with satellite imagery, publicly available photographs, archival records, and open datasets to generate a massing-level 3D model. Building height and volumetric characteristics are estimated using contextual cues such as surrounding structures, known architectural typologies, and scale references derived from people or urban elements. The resulting Digital Twin prioritizes geometric plausibility over fine architectural detail, enabling simulation, analysis, and decision-support tasks, such as environmental modeling, airflow and CFD approximation, and high-level Heritage BIM integration, while fully respecting cultural heritage restrictions. Three case studies illustrate the proposed workflow and systematically identify which components of conventional smart-building and Digital Twin pipelines remain feasible and which become infeasible under heritage regulations. The results demonstrate a practical and scalable path toward compliant Digital Twins for protected buildings, positioning low-detail modeling not as a limitation but as a regulatory necessity.

The author proposes a method for combining orthophotoplan fragments with an ESRI map’s satellite image raster by geographic coordinates in GeoTiff format. The features of creating a raster map in the Global Mapper Pro 24.1 application and fragments of an orthophotoplan depiction in Agisoft Metashape Professional are considered. A description of the combination algorithm and interface of the developed program in Python is given using the example of processing the Amur coastal zone’s image. The features of the algorithm include the following. At the first stage, rectangular frames are formed on the satellite image raster, the geographic coordinates of which correspond to fragments of the orthophotomap obtained using the Geoscan 401 UAV. Recommendations are given for choosing the size of fragments, taking into account the permissible maximum resolution and convenient viewing in zoom mode on a monitor screen with a resolution of at least 2K. Next, when the user selects the frame of interest using a simple mouse click, a scaled fragment of the orthophotoplan is automatically inserted into the raster image of the map. Finally, an analysis of the experimental results and conclusions of minor errors in the algorithm that arise due to the relative shift of geographic coordinates are presented

Flash floods represent one of the most recurrent and devastating geohazards affecting Egypt’s northern area of the Red Sea coast, particularly in hyper-arid catchments such as Wadi Araba. This study develops a geospatial decision-support framework to identify flash-flood-prone areas and determine optimal dam locations for flood mitigation and groundwater recharge. The analysis incorporates eight thematic layers: slope, LULC, rainfall, DEM, drainage density, lineament density, distance to main streams and roads. The data used for thematic layers was derived from satellite imagery and ancillary datasets, processed using remote sensing and GIS tools. A multi-criteria decision analysis (AHP) generated a weighted overlay model of flood susceptibility. The resulting map classified Wadi Araba into high-risk (0.53 km²), moderate-risk (2354.8 km²), and low-risk (1671.34 km²) zones, with the highest risk concentrated in the southern Galala Plateau, moderate-risk zones occurring in lowland and downstream basins, and low-risk zones mainly located in the northern Galala Plateau and western Wadi Araba. The RS–GIS/AHP framework identified top-ranked dam sites with favourable storage geometry; validation returned medium AUC for susceptibility of flash flooding (≈ 0.6–0.7) and good AUC for dam suitability (≈ 0.7–0.8).

Abstract The extraction and utilization of fossil fuels from mining areas worldwide have led to significant CO2 emissions. Post-mining landscapes also present challenging environmental conditions that can hinder effective land use. In recent years, the installation of photovoltaic (PV) and floating photovoltaic (FPV) systems in abandoned mining areas has emerged as a promising solution. However, large-scale solar energy potential assessment methods in these areas are still lacking. To address these challenges, SolarMiner is presented, combining the approaches of both computer vision model and satellite imagery. By segmenting and identifying different types of mining areas and calculating their respective surface areas, the installation potential of both traditional PV and FPV systems is calculated. The model performance is validated using data from a province of China, Shanxi. Results reveal that there is a substantial solar energy potential in the mining areas, exceeding 1446 TWh annually, which was 6.52 times the total electricity consumption of Shanxi in 2023. The levelized costs of electricity range from 0.023 to 0.042 USD/kWh.

The detection of urban and environmental change using satellite imagery is crucial for urban planning, sustainable development, and environmental monitoring. Remote sensing and artificial intelligence breakthroughs have enabled the analysis of large-scale, high-resolution data to reveal long-term alterations. This study introduces Urban-E, a satellite-based system for monitoring urban growth and environmental changes in Egypt’s New Administrative Capital from 2016 to 2024. A dataset of 1,605 multi-temporal satellite images was methodically collected and preprocessed to improve their clarity utilizing techniques like denoising, histogram equalization, and edge detection. Deep learning is employed to classify satellite images into predefined geographic regions, enabling automated spatial localization of urban zones. Change analysis is then performed using image-based comparison techniques, including pixel-level differencing, texture analysis, segmentation, and NDVI-based vegetation assessment, to quantify urban expansion and environmental variation over time. In addition, geoJSON datasets from OpenStreetMap are integrated to analyze the spatial distribution of urban services such as schools, hospitals, and commercial facilities. The CNN-based regional classification achieved an accuracy of 93.3%, facilitating reliable region-specific change analysis. The results reveal significant urban expansion, infrastructure growth, and increased green coverage across multiple zones. The proposed framework demonstrates how deep learning–assisted regional classification combined with satellite-based analytical techniques can support evidence-based urban and environmental monitoring for sustainable city planning.

Development of Spatio-Temporal Cross-Attention-Based Adaptive Contrastive Learning with Residual LSTM for Global Conservation Efforts Using Satellite Images

ABSTRACT Deep learning has become central to recent advances in segmenting remote sensing imagery due to its powerful feature extraction capabilities. Most existing deep learning-based approaches concentrate on spatial or channel-wise feature extraction in isolation. However, overlooking the interplay between these two aspects will significantly impair the performance of semantic segmentation. Motivated by this observation, this paper proposes a Spatial-Channel Cross-Decoding Transformer model (SCCFormer). To effectively extract spatial and channel features from remote sensing images, SCCFormer introduces a dual-stream Transformer decoder network based on a deformable attention mechanism that captures the contextual dependencies of spatial and channel features. Regarding the correlation between spatial and channel features, SCCFormer proposes a multi-stage, spatial-channel features cross-decoding method, which cross-decodes and fuses spatial and channel features at various feature scale levels, thereby capturing the relevant semantic information of both spatial and channel features. The proposed SCCFormer is evaluated based on the Potsdam, Vaihingen, LoveDA and Multi-Source Satellite Imagery for Segmentation remote sensing datasets which demonstrated the effectiveness of the model in the task of remote sensing image semantic segmentation.

Land Use and Land Cover (LULC) classification plays a crucial role in remote sensing applications such as urban planning, environmental monitoring, agricultural analysis, and climate studies. Recent advances in deep learning, particularly convolutional neural networks (CNNs), have significantly improved classification accuracy for satellite imagery. This thesis presents a comparative study of two deep learning approaches for LULC classification using the EuroSAT dataset: a convolutional neural network trained from scratch and a transfer learning model based on a pre-trained VGG-19 architecture. The EuroSAT dataset consists of Sentinel-2 satellite images categorized into ten land cover classes. Experimental results demonstrate that transfer learning achieves superior classification performance compared to training a CNN from scratch, highlighting the effectiveness of pre-trained models for remote sensing image analysis.

Abstract British colonial urban planners in Kampala, Uganda, designed segregated neighborhoods for Europeans, Asians, and Africans, under the colonial public health guidance. No studies have investigated how these historical urban design decisions relate to modern air pollution exposure disparities in African cities. We sought to determine whether colonial‐era racial segregation of settlements in Kampala, Uganda, influences modern intra‐urban spatial distribution of ambient fine particulate matter (PM 2.5 ) concentrations in the city. Using an urban network of 52 low‐cost PM sensors, we estimated monthly PM 2.5 concentrations across Kampala from 2021 to 2022. We manually digitized and correlated 2020 satellite imagery and street maps of the three historical boundaries of European/Asian and African settlements from a 1951 map. Extreme gradient boosting trees, random forest models, and empirical Bayes kriging were used to predict gridded monthly PM 2.5 averages for the study area. We used non‐parametric statistical tests and multivariable linear regression to compare monthly PM 2.5 levels between European/Asian and African settlements. We observed significantly higher median monthly PM 2.5 concentration in African settlements (median = 33.0 μg/m 3 [interquartile range = 28.2, 39.3]) compared to European/Asian areas (median = 28.8 μg/m 3 [IQR = 25.2, 33.5]). In crude linear regression models, African settlements had a higher PM 2.5 concentration by 9.2% (95% CI: 6.4, 18.5). African settlements showed greater seasonal PM 2.5 variations and higher current social vulnerability. This study highlights the enduring influence of colonial urban planning on today's environmental health inequalities.

River erosion is one of the most dangerous and unpredictable natural hazards to riparian communities worldwide. The Bhagirathi-Hugli River, which is an affluent of the Ganga, has undergone massive geomorphological modifications over the last few decades, especially after the Farakka Barrage in 1975. The proposed research explores the spatiotemporal change of the bank erosion and channel shifting of a 45-kilometre section of the Bhagirathi-Hugli River between the Katwa and Purbasthali blocks in West Bengal, India, over 30 years (1990-2020). The patterns of erosion and accretion, the rate of meandering, and the process of formation of the oxbow lakes were measured, which have undergone significant geomorphological changes over the last few decades, especially after the construction of temporal satellite images combined with Geographic Information System (GIS) and verified with the help of large field surveys. The findings indicate that the channel is largely unstable, with the maximum left bank deposition at Agradwip and the right bank deposition in the same location. During the study period, three oxbow lakes formed, and two additional meander cutoffs are predicted to occur in the next 10-15 years. The study also investigates the trickle-down effects of bank erosion on agricultural livelihood, settlement patterns, and socio-economic susceptibility of the affected communities. Results show that the hydrological changes achieved by the Farakka Barrage have significantly altered the river's natural balance, affecting discharge, competence, and sediment load, leading to faster meander migration and bank erosion. The research has been significant to the literature on fluvial geomorphology by offering quantitative data on channel changes after the barrage and also gives practical advice on integrated river basin management and livelihood adaptation measures in the erosive-prone riparian areas.