Since 2023, although Machine Learning (ML) has been widely applied to vegetation extraction from Remote Sensing Images (RSI), most research has focused on static coverage. Despite recent progress in time-series monitoring and multi-technology integration, persistent issues remain, including low accuracy in quantifying recovery trajectories, inadequate characterization of temporal evolution, and poor adaptability in multi-source data fusion. These limitations hinder precise habitat monitoring along transmission corridors. To address these challenges, this study develops an adaptive recovery trajectory modeling technique for multi-source disturbance fusion. By integrating RSI and LiDAR data, a dynamic fusion method for multi-source remote sensing time-series features is proposed, and a recovery trajectory model that incorporates quantitative construction disturbance factors is constructed. A multi-scale temporal attention mechanism is designed to enhance the detection of key turning points. Additionally, a corridor-specific application system ensures alignment with monitoring scenarios. The core innovations include adaptive stage-wise weight adjustment for multi-source time-series fusion and the introduction of disturbance adaptation factors, which improve trajectory description accuracy. Simulation experiments show that the proposed technique achieves a comprehensive performance score of 0.889–0.905, outperforming YOLOv10 and ConvLSTM by 13.7–23.5 %. Key recovery trajectory turning points are identified with 91.5 % accuracy, the mean absolute error (MAE) is reduced by 32.1–49.4 %, and processing time (RT) is shortened by 26.0 % compared to LiDAR + Optical RSI fusion. Robustness reaches 0.915, and adaptability across three typical monitoring scenes averages 0.889. This technique fills a gap in quantitative habitat recovery modeling for transmission corridors, supports dynamic vegetation monitoring and regulation, and advances intelligent ecological maintenance in power grid operations.
On December 18, 2023, an M S6.2 earthquake stuck Jishishan County, Linxia Prefecture, Gansu Province. This study retrospectively characterized the resistivity anomalies before the M S6.2 Jishishan earthquake by analyzing observational data from seven earth resistivity stations within 300 km of its epicenter. Data was analyzed using original data curves, normalized variation rate method (NVRM) curves, and earth resistivity anisotropy degree (S). NVRM and S are dimensionless methods for geoelectric analysis – NVRM helps avoid subjectivity in anomaly determination and unifies anomaly criteria across stations. S suppresses local disturbances and amplifies electrical anomaly information related to earthquakes. The original data curves show that high-frequency disturbances in the hourly values of the original data occurred more frequently before the earthquake and that the trend of the daily mean values changed. The NVRM curves demonstrate that the normalized variation rates of resistivity at all seven earth resistivity stations exceeded the threshold (±2.4) before the earthquake and that resistivity anomalies were mostly attributed to negative changes, with diverse shapes. The changes in S indicate that the station farthest from the epicenter had the smallest amplitude and earliest occurrence of changes in S and that the station closest to the epicenter had the largest amplitude and latest occurrence of changes in S, revealing clear changes in the degree of earth resistivity anisotropy in the epicentral area and a large range of observation points in its vicinity. These results suggest that the resistivity changes before the Jishishan earthquake likely had a mechanical mechanism closely related to the seismogenic earthquake process.
Accurate prediction of unsaturated soil behaviour depends on the reliable estimation of the soil-water retention curve (SWRC) and the corresponding hydraulic conductivity function (HCF). Although the van Genuchten-Mualem (VG-M) model is widely accepted for this purpose, its parameter estimation typically relies on nonlinear optimisation methods, which may produce values with limited physical meaning. This research presents a graphical method for directly identifying model parameters from essential characteristics of the SWRC, including the saturated and residual water contents, the inflection point, and the residual suction. The proposed approach removes the requirement for repetitive numerical fitting, improves the clarity and physical understanding of the parameters. The method was applied to measured SWRC datasets representing different soil textures. The graphical fits produced R 2 values of 0.9619–0.9926 and RMSE values of 0.0065–0.0172. For comparison, nonlinear least-squares fitting produced R 2 of 0.9909–0.9989 and RMSE values of 0.0020–0.0118. Although optimisation yielded lower errors, the graphical method retained a direct physical interpretation of the parameters. The graphically derived parameters for the VG-M HCF model captured the general hydraulic conductivity trend (R 2 of 0.8540). Deviations at high suction were mainly associated with known limitations of the VG-M model at the dry end, rather than the graphical procedure itself.
The construction site of a silo, which was built as two sites, consists of very soft alluvial soils. Since the superstructure loads exceeded the bearing capacity of the shallow foundation, the structure was built on a deep foundation with bored piles. Due to the very soft soils, some difficulties were encountered during the bored pile construction at site-1. As a result of the difficulties encountered, Jetgrout columns were constructed around the bored piles in site-2 and then the bored piles were constructed. Loading tests were performed on the bored piles at both sites and their bearing capacity performance was evaluated. The study comparatively describes the bored pile construction before and after soil improvement in very soft soils, the difficulties encountered during the construction phase, the solutions to the difficulties, and the bearing capacity performance of the piles. At the same time, in this article, it is explained in detail with field observations that bored piles installed in partial composite soil improved with jetgrout columns are faster and more economical. This study is a field study that aims to contribute to the literature by evaluating the geological, geophysical and geotechnical evaluation of such industrial structures constructed with large-scale budgets from an academic point of view.
In a previous study, the Normalized Difference Gamma-Ray Index (NDGRI) was introduced as a spectral approach to highlight surface conditions associated with elevated gamma radiation. When validated against car-borne and airborne spectrometry surveys, NDGRI demonstrated potential for delineating radiation anomalies in semi-arid terrains. Detecting such anomalies from satellite data is of practical importance, as it offers a cost-effective way to narrow down prospective zones before deploying expensive car-borne or airborne surveys. Building on this foundation, this study assesses whether deep learning can extend this approach beyond index-based methods by classifying elevated radiation from Sentinel-2 spectral data. Multilayer perceptron (MLP) networks were implemented for per-pixel spectral classification, using fully connected dense layers with dropout regularization. Hyperparameters, including number of layers, neurons per layer, dropout rates, optimizers, and learning rates, were tuned via Keras Tuner and AutoKeras in Python to balance model complexity and generalization. Training and validation samples were extracted from car-borne and air-borne gamma-ray spectrometry heatmaps, providing pixel-level labels for supervised learning. Three input configurations were tested: NDGRI provided as a single-band raster, the full set of 11 Sentinel-2 bands, and selected band subsets that showed diagnostic potential in our previous NDGRI analysis. The results showed that the NDGRI raster performed strongly as a compact predictor of elevated radiation, whereas multi-band inputs provided better overall performance, with certain subsets consistently outperforming NDGRI alone. These findings indicate that although NDGRI efficiently captures key spectral contrasts, the inclusion of targeted spectral combinations enables the network to exploit complementary features and improve classification robustness. The findings suggest that deep learning can serve as a powerful extension of NDGRI-based workflows, combining the interpretability of targeted indices with the flexibility of data-driven classification. This approach could offer a scalable and computationally efficient means of mapping elevated gamma-radiation anomalies from multispectral satellite imagery, with implications for cost-effective uranium exploration in semi-arid terrains.
In engineering survey practice, the inversion analysis of Rayleigh Wave Dispersion Curves (RWDCs) has become a key technical means for obtaining the physical and mechanical parameters of subsurface. However, traditional global optimization algorithms generally face bottlenecks such as low convergence efficiency, limited solution accuracy, and susceptibility to premature convergence in this inversion process. To address this, this paper proposes an improved RWDCs inversion model based on the Improved Salp Swarm Algorithm (ISSA): the Tent chaotic mapping is used to initialize the population configuration, enabling precise control over the initial spatial distribution of the population; combined with an adaptive update mechanism, this dynamically balances the algorithm’s global exploration and local optimization performance. To validate the applicability and reliability of the ISSA in obtaining subsurface shear wave velocity structures through RWDCs inversion, a practical experimental design was developed: systematic inversion tests were conducted using three simulated scenarios based on theoretical geological models and two actual engineering survey data scenarios. The experimental results show that ISSA can be stably applied to the RWDCs inversion process and demonstrates excellent inversion performance. Compared with the classical Particle Swarm Optimization (PSO), ISSA exhibits significant advantages in terms of inversion efficiency (convergence speed), solution accuracy (parameter agreement), and numerical stability. Joint inversion results using active source data from Iceland and active-passive source data from Italy demonstrate the high reliability of ISSA inversions, indicating that this method can provide efficient and precise algorithmic support for the quantitative interpretation of RWDCs.
This study examines the ancient Roman settlement of Municipium S, active from the 1st to the 4th century AD. The sites are located near Pljevlja (Montenegro). The research aimed to assess its scientific, natural, conservation, and tourist value using both quantitative and qualitative methods. It focused in particular on cultural policy and sustainable development. A major challenge is the lack of systematic tools for evaluating archaeological and geological heritage. This makes it harder to include the sites in planning and tourism. The study used the geoheritage assessment model (GAM) and urban geoheritage assessment model (UGAM) models, as well as SWOT, GAP, and PESTEL analyses. Data were collected from fieldwork, surveys, focus groups, and expert interviews. The urban geosite value of 166.88 and the tourist value of 49.70 place Municipium S in field U31 of the UGAM matrix. This is typical for research-oriented sites. The scientific value (100.47) and natural value (35.88) indicate strong archaeological significance and a unique landscape. The lower conservation value of 29.73 and the tourism value of 49.70 indicate limited infrastructure, insufficient conservation and insufficient coordination. This site is an important scientific site in Montenegro. A special contribution of this article is that the importance factor from the UGAM model was calculated for the Montenegrin tourism market. This research provides a framework for the assessment of archaeological and geological heritage. It helps decision-makers guide long-term management and promotion strategies.
Map generation from remote sensing imagery is a critical task in urban planning and Earth observation. Recently, deep learning-based style transfer methods have emerged as a transformative approach due to the ability to handle complex image features and significantly enhance production efficiency. Unlike existing surveys that broadly cover general image-to-image translation, this paper provides a uniquely targeted review focusing explicitly on the intersection of style transfer algorithms and geographical map synthesis. We systematically trace the evolutionary path from traditional map-making methods and conventional image style transfer techniques to advanced deep learning models, explicitly linking how general feature extraction architectures have been adapted for the strict spatial constraints of remote sensing data. Furthermore, we critically analyze the evolution of generative adversarial networks (GANs)-based style transfer methods and the application in generating map images, categorizing them into text-free and text-annotated generation tasks – a crucial distinction representing the latest innovative frontier in this field. In this paper, the inherent disadvantages of current methods for generating map images from remote sensing imagery, such as incomplete retention of topological map information, blurred road edges, and unstable GAN training, are comprehensively evaluated. Finally, possible solutions and future research directions are proposed to offer new perspectives for advancing highly accurate and practically applicable map image generation through remote sensing style transfer.
Deran Lake, located within Hutovo Blato Nature Park in Bosnia and Herzegovina, is a shallow wetland ecosystem shaped by complex hydrological connectivity, meteorological variability, and anthropogenic pressures. This study applies and adapts a spatial multicriteria decision-making framework, originally developed for water quality modelling in a different lake system, to assess eutrophication susceptibility in Deran Lake. The main objective was to evaluate the flexibility and robustness of the framework under different hydrodynamic and ecological conditions. Monitoring was conducted from June to October 2025 and included 76 in situ measurements of water temperature (WT), dissolved oxygen (DO), electrical conductivity (EC), and turbidity. These data were integrated with meteorological variables, hydrological characteristics, land use patterns, and potential pollution sources. All criteria were standardized using fuzzy membership functions, weighted through a hierarchical procedure, and aggregated into a spatial susceptibility index. The results revealed a clear spatial gradient, with higher susceptibility in the southern and southwestern parts of the lake and lower susceptibility in the northern sector. These patterns reflect the combined effects of nutrient transport pathways, wind exposure, and anthropogenic pressures. Spatially explicit sensitivity analysis showed that the model is robust, with DO, WT, turbidity, and EC exerting the strongest influence. The proposed framework provides a transparent, transferable, and decision-oriented tool for wetland management and for prioritising future monitoring efforts.
Cellular automata (CA) are common instruments for simulating spatial dynamics in geography, ecology, and urban science. Most implementations, however, rely on fixed transition rules or static suitability surfaces that remain invariant across simulation time. This paper argues that such assumptions are inadequate for systems in which local decision metrics are themselves products of spatial interaction, competitive selection, and historical accumulation. I propose the Adaptive Multi-Criteria Ecotope Cellular Automaton (AME-CA), in which each cell on a spatial lattice evaluates its viability through a weighted multi-criteria fitness function whose criterion weights evolve endogenously. Weight adaptation is governed by local outcome history, neighbourhood diffusion of behavioural tendencies, and competitive pressure from adjacent cells. The framework introduces the multi-criteria ecotope: a spatially localized behavioural niche defined not only by environmental conditions but by historically stabilized adaptive decision rules. I implement the model on a synthetic 100 × 100 lattice and compare three variants: (1) a classic CA with fixed rules, (2) a CA with static multi-criteria evaluation, and (3) the full AME-CA, across metrics of spatial pattern formation, path dependence, competitive dynamics, and equilibrium convergence. Results show the adaptive model generates richer heterogeneity, distinct ecotopes, stable local regimes, and emergent clustering absent in fixed-rule alternatives.
On 27 September 2025, an M S5.6 earthquake occurred in Longxi, Gansu, situated on the Zhangxian segment of the West Qinling Fault. Based on GNSS observations, we systematically analyzed pre-seismic horizontal movement, strain rate evolution, fault behavior, cross-fault baseline variations, and the spatio-temporal distribution of the regional strain field. Results showed that the study area exhibited a deformation pattern of NE–SW shortening and NW–SE extension, which aligned with the NE-oriented principal compressive stress field of the northeastern Tibetan Plateau. The epicenter was located at the margins of high negative values of dilatation and E–W strain rates, characterized by continuous contractional strain accumulation. Approximately three years prior to the event, the time series of multi-parameter deformation and strain deviated from linear trends and exhibited a notable deceleration and leveling-off, suggesting a possible transition from steady-state tectonic loading toward enhanced fault locking. These findings record the seismogenic process from regional loading to localized strain accumulation, suggesting that the Zhangxian segment was in a state of high seismic hazard before the rupture.
Tourmaline, a chemically complex borosilicate mineral, serves as a sensitive recorder of geological processes across a wide range of environments. However, the rapid expansion and disciplinary diversification of tourmaline research over the past two decades have made it difficult to evaluate its global development objectively. This study aims to systematically map and quantify the intellectual structure, thematic evolution, and collaboration networks of tourmaline research from 2000 to 2025. A total of 2,797 publications were retrieved from the Web of Science Core Collection and analyzed using CiteSpace and VOSviewer to analyze co-citation patterns, keyword clustering, collaboration networks, and dual-map overlays. The results reveal a steady increase in publication output and a transition from traditional mineralogical and petrological studies to interdisciplinary applications spanning geochemistry, materials science, and environmental engineering. China has emerged as the most prolific contributor, while Europe and North America maintain high network centrality and influence through methodological innovation and collaboration. Core journals such as Ore Geology Reviews, American Mineralogist, and Canadian Mineralogist serve as primary hubs for knowledge dissemination. Emerging hotspots include boron isotope geochemistry, U–Pb geochronology, and tourmaline-based environmental applications. This study provides a comprehensive, data-driven overview of global tourmaline research by identifying its intellectual structure, thematic evolution, and collaboration patterns. The findings clarify major publication trends, core journals, influential research topics, and emerging directions, thereby offering a bibliometric reference for future studies on tourmaline in geoscience, resource exploration, and environmental applications.
Moisture infiltration and subsequent flowstone deposition present significant challenges for railway tunnels constructed in karst limestone environments. This study evaluates the applicability of terrestrial laser scanning (TLS) intensity for mapping moisture distribution on tunnel walls and identifying areas with increased risk of future flowstone growth. A 100 m section of the Jurgovski railway tunnel (Slovenia) was scanned during two epochs in 2013 using a Leica ScanStation C10. Scans were acquired during slightly humid weather (E00) and following rainfall (E01). Point clouds were georeferenced, intensity-normalised to reduce geometric effects, and downsampled to a 4 × 4 cm grid. A moisture classification threshold of 0.19 was applied, based on prior experimental studies using the same scanner model. Moist areas increased from 2.9 % (17.91 m2) of the analysed surface in E00 to 31.8 % (197.05 m2) in E01, reflecting rapid infiltration typical of karst limestone. Identified wet zones correspond to locations most susceptible to calcite precipitation due to repeated wetting–drying cycles. The results demonstrate that TLS intensity can serve as a rapid, non-destructive screening method for locating moisture-prone segments and supporting maintenance decision-making in operational railway tunnels. Limitations include the absence of in-situ moisture measurements and laboratory calibration of tunnel material, which should be addressed in future work.
Soil erosion represents a critical threat to natural resource sustainability in arid environments, yet validation of empirical models in hyper-arid watersheds remains limited. This study employs the Revised Universal Soil Loss Equation (RUSLE) integrated with Geographic Information Systems (GIS) and remote sensing to quantify rainfall erosivity and erosion risk dynamics in Wadi Allith basin, Saudi Arabia, across 2016–2018. RUSLE factors, rainfall erosivity (R), soil erodibility (K), slope length-steepness (LS), cover management (C), and support practices (P) were derived from WorldClim precipitation data, FAO-UN Soil Map, SRTM digital elevation model, and Landsat-derived NDVI. Results reveal a pronounced orographic gradient in rainfall erosivity, increasing 29 % from 2016 (120 mm yr−1) to 2018 (155 mm yr−1), with high-erosivity zones expanding from 15 % to 35 % of basin area. Despite this intensification, the basin maintains predominantly “slight” erosion risk (ERC 1–2, <5 t ha−1·yr−1) across 95–97 % of its area, attributed to low mean LS-factor (≈2.3). However, a statistically significant 6 % relative increase in soil loss (0.3 t ha−1·yr−1) occurred, concentrated in 3.2 % of highland tributaries where LS > 10. This demonstrates that while low-gradient topography provides a geomorphic buffer, climate-driven erosivity increases threaten to breach critical erosion thresholds. The successful application of open-source GIS tools confirms cost-effective scalability for long-term monitoring in data-scarce arid regions. These findings inform targeted conservation strategies and highlight the urgency of climate-adaptive soil management in the Arabian Peninsula’s vulnerable dryland watersheds.
Cardiovascular diseases represent one of the most dominant categories of non-communicable diseases, with a high prevalence both in Serbia and globally. Understanding their spatial distribution provides valuable insights for public health planning and disease prevention. This study applies geospatial analysis to examine the patterns of CVD among patients treated at the Institute for Cardiovascular Diseases of Vojvodina in Serbia, over the period 1995–2023. Beyond identifying spatial clusters, the analysis reveals long-term intensification of CVD burden in the South Bačka urban-suburban zone, indicating a sustained spatial polarization of cardiovascular morbidity over nearly three decades. Mann-Kendall trends per settlement were calculated. The spatial dimension of the research was addressed using the Emerging Hot Spot Analysis to detect statistically significant clusters of high and low values. To further refine the spatial representation of CVD clusters, the Kriging interpolation method was applied. The findings underscore the significance of geospatial methods in pinpointing high-risk areas in cardiovascular diseases, providing an evidence-based foundation for targeted healthcare interventions and resource allocation. Moreover, this study demonstrates the potential of integrating spatial geostatistics with health data to support the development of preventive strategies and to guide future epidemiological research on non-communicable diseases in Serbia and beyond.
To investigate the negative skin friction of pile foundations induced by strong earthquake-triggered loess seismic subsidence, a load transfer function for a single pile embedded in a homogeneous loess foundation was established under the combined action of pile head load and seismic loading. Based on the depth-dependent characteristics of seismic subsidence, an exponential function was proposed to fit the subsidence–depth relationship. The fitted function was then incorporated into the load transfer model to derive closed-form analytical solutions for the axial force and shaft friction distribution along the pile. An engineering case study was conducted to validate the proposed method. The theoretical results were compared with numerical simulations performed using PLAXIS 3D finite element software. The comparison shows good agreement between the two approaches, demonstrating the validity and feasibility of the proposed analytical method. The findings of this study provide useful guidance and reference for the practical design of pile foundations in loess regions subjected to strong seismic events.
With the increasing prevalence of indoor activities and the growing demand for refined spatial information, 3D indoor scene reconstruction has become a core supporting technology for geospatial services, digital twins, and smart cities. This paper presents a comprehensive survey of data-driven approaches for automated 3D indoor modeling. We first systematically summarize the major data modalities, including floor plans, point clouds, RGB-D images, images, and textual descriptions, along with representative open datasets and key spatial data standards. We then provide a structured review of the full technical pipeline, covering permanent building component identification, scene reconstruction, spatial topology construction, and model texture optimization, as well as state-of-the-art learning-based and generative AI methods. Finally, we conduct in-depth comparisons of accuracy, efficiency, and automation across different sensing and modeling paradigms, and discuss the extended values of these techniques for geoscience applications such as indoor navigation, emergency response, building monitoring, and digital twin management. This review aims to clarify the technical evolution, current challenges, and future trends of data-driven indoor 3D reconstruction, providing a systematic reference for researchers in geoinformatics, computer vision, and digital construction.
During various operations, whether civilian or military, helicopters are frequently used in inaccessible areas. This research aims to identify and classify suitable areas, i.e., safe helicopter landing sites, within the City of Kraljevo’s territory. By applying multi-criteria spatial modelling, slope calculation, and Land Use/Land Cover (LULC) enhancement through multispectral analysis of high-resolution satellite imagery within a Geographic Information Systems (GIS) environment and Python programming language, safe helicopter landing sites were proposed. The results indicate that terrain morphometric characteristics represent a key limiting factor in the selection of landing sites; however, they also show that the application of spatial filtering and geoprocessing enables the identification of areas that meet strict technical and safety requirements. The development and application of such methodologies contribute to more reliable identification of suitable landing areas, thereby improving safety, efficiency, and timeliness in emergency aviation operations.
With the aim of improving the effective utilization rate of solid wastes such as loess and coal gangue, and mitigating the disadvantages of high alkalinity and environmental vulnerability associated with traditional vegetation concrete, this study used loess instead of cement as a cementitious material and coal gangue instead of stone as a coarse aggregate to prepare a loess-based composite coal gangue porous vegetation concrete (LCCG-PVC). An orthogonal experimental approach was adopted to investigate how variations in water–cement ratio, cement dosage, target porosity, and water-reducing agent content affect the hydraulic permeability, compressive performance, and effective void ratio of vegetation concrete, then the optimal mix proportion was comprehensively evaluated and determined. At the same time, by setting up multiple sets of planting experiments under different conditions, the dynamic changes of plant height and root length were monitored to verify the planting performance of the material. The results show that under the premise of meeting the sufficient porosity and low pH value required for plant growth, the optimal mix proportion parameters of LCCG-PVC are as follows: target porosity 25 %, water-cement ratio 0.42, cement content 15 % (mass fraction), the proportion of water-reducing agent is 2.0 % (mass fraction). In the planting experiment, experimental group with an average plant height of 18.9 cm had the best growth condition. Based on this, the suitable planting method and the optimum ratio of planting concrete grouting liquid were further determined.
With the frequent occurrence of geological disasters such as uneven settlement and surface subsidence in mining areas, traditional monitoring techniques face shortcomings such as atmospheric errors, limitations in data collection, and difficulties in selecting ground control points, which seriously affect the accuracy and efficiency of monitoring. Therefore, a small baseline subset interferometric synthetic aperture radar method that integrates Beidou satellite data is proposed. By improving the data fusion method and using Beidou satellite base stations as the selection benchmark for ground control points, a high spatio-temporal resolution monitoring strategy is established. Moreover, its effectiveness is verified through experiments conducted in a mining area in Dazhou City, Sichuan Province, China. The experimental results indicate that the proposed method shows high consistency with Beidou satellite observations and ground truth measurements. Compared with the non-fusion scheme and other existing methods, the proposed approach achieves significantly improved accuracy and reliability in deformation monitoring. The findings illustrate that the fusion of Beidou satellite data and synthetic aperture radar data can not only improve the accuracy and reliability of mining area monitoring, but also has important significance for the timely detection and early warning of potential geological disasters. The proposed method for monitoring surface subsidence in mining areas provides strong technical support for safety production and environmental protection in mining areas.
