Search Results (991)

Earthquake-induced landslide reactivation poses a sustained hazard for years following major seismic events, yet operational prediction remains constrained by heterogeneous multi-modal data, sparse supervision, and the absence of uncertainty-aware frameworks. This paper presents PIMSEL, a physically guided multi-modal semi-supervised framework for post-seismic landslide reactivation risk assessment. PIMSEL integrates satellite-derived morphological features, precipitation time series, and seismic hazard attributes through four components: entropy-regularized optimal transport for cross-modal semantic alignment without paired supervision; causally constrained hierarchical fusion enforcing domain-consistent modal weighting; scenario-based prototype mutation for semi-supervised learning from sparse expert annotations; and prototype-anchored variational graph clustering that simultaneously stratifies landslides into HIGH, MEDIUM, and LOW risk tiers and produces decomposed aleatoric and epistemic uncertainty estimates for operational triage. The HIGH risk tier operationally corresponds to predicted reactivation, validated against 598 documented reactivation events across 7482 co-seismic landslides from three Sichuan Province earthquake sequences: the 2013 Lushan (Mw 7.0), 2017 Jiuzhaigou (Mw 7.0), and 2022 Luding (Mw 6.8) events. PIMSEL achieves 82.5% reactivation recall and 66.4% precision, outperforming twelve baselines across clustering quality, classification, and uncertainty calibration metrics. Ablation studies confirm that optimal transport alignment contributes the largest individual performance gain. Current limitations include quarterly assessment frequency and dependence on optical imagery under cloud cover, which future integration of real-time meteorological triggers and SAR data should address. Full article

Large-scale smart agriculture requires reliable and energy-efficient wireless connectivity to support distributed environmental sensing across wide rural areas. However, existing low-power wide-area network (LPWAN) technologies often face limitations in scalability, reliability, or infrastructure dependency when deployed in large agricultural fields. This study presents the design and experimental evaluation of a hierarchical sensor network architecture that integrates LoRaMESH for multi-hop sensing communication and Wi-Fi HaLow as a sub-GHz backhaul for data aggregation and cloud connectivity. In the proposed system, LoRaMESH forms intra-cluster sensor networks using a lightweight controlled flooding protocol, while Wi-Fi HaLow provides long-range IP-based connectivity between cluster gateways and a central access point. A real-world deployment covering approximately $2.5\mathrm{km}\times 1\mathrm{km}$ of agricultural area was implemented to evaluate the performance of the proposed architecture. Experimental results show that the LoRaMESH network achieves packet delivery ratios above $90\%$ across one to three hops, with average end-to-end delays between $10.6$ s and $13.3$ s. The Wi-Fi HaLow backhaul demonstrates high reliability within short to medium distances, reaching $99.5\%$ packet delivery ratio at 50 m and $89.68\%$ at 200 m. Energy measurements further indicate that the sensor nodes consume only $21.19\mathsf{\mu }\mathrm{A}$ in sleep mode, enabling long-term battery-powered operation suitable for agricultural monitoring applications. These results indicate that the proposed hierarchical architecture is a feasible connectivity option for the tested large-scale agricultural sensing scenario. Because no side-by-side LoRaWAN or NB-IoT benchmark was conducted on the same testbed, the results should be interpreted as a field validation of the proposed architecture rather than as a direct experimental demonstration of superiority over alternative LPWAN systems. Full article

Despite its continued growth, cloud computing remains susceptible to significant security challenges, as shared virtualised environments pose threats at multiple levels. These vulnerabilities are caused by a lack of security coverage in the responsibility model between the provider and the tenant. In this work, we propose the multi-layered architecture VIRTUOSO (VIRTual Unified Operation Security Optimiser) to cover these security gaps through advanced automation and ML. VIRTUOSO has four layers. The Input Layer extracts key risk components from collected telemetry data. The Deep Automation Security Layer provides automated actions and continuous monitoring of security defences. Its counterpart, the Intelligent Security Layer, predicts threats using anomaly detection. The last layer, the Output Layer, returns an aggregated risk summary. The datasets we used were chosen for their relevance: the UNSW-NB15 dataset, a subset of the web-attack classification from CSE-CIC-IDS2018, and a sample of anonymised log events from AWS CloudTrail. Our ensemble classifiers achieve a best accuracy of 95.08% ± 0.13% on UNSW-NB15 (RF), with statistically significant differences among models confirmed by the Friedman test (p < 0.004) and Nemenyi post hoc analysis, and 99.25% ± 0.52% on web-attack (CatBoost), where ensemble differences are not statistically significant (p = 0.093), consistent with the high separability of this dataset. The training-test gap and DNN curves show no overfitting, whereas our adversarial tests show a maximum accuracy loss of 8.1% at ε = 0.02. With these promising results, we can assert that, pending verification in an actual cloud environment and potential integration with FL, our ensemble classifier model appears to be a good real-world prototype. Full article

(1) Background: Heartbeat detection from electrocardiogram (ECG) and photoplethysmograph (PPG) signals is widely used in wearable devices for health monitoring, fitness tracking, and stress assessment. While numerous methods have been proposed, their practical suitability depends not only on accuracy but also on computational and memory constraints inherent to resource-limited systems. (2) Methods: A scoping review of 52 studies published between 2017 and 2024 was conducted, covering time-domain, frequency-domain, matrix-based, and machine learning approaches. The methods were evaluated according to estimation accuracy, computational complexity, memory footprint, and suitability for on-device implementation. (3) Results: Time-domain peak detection methods consistently provide high accuracy (minimum of $79.25$%, maximum of $99.96$%, and median $\ge 99.69$%) for ECG and reliable heart rate estimation for PPG with linear computational complexity, low memory requirements and low energy consumption. Frequency-domain approaches are suitable for average heart rate estimation from PPG but do not preserve inter-beat intervals (error range of [1.07, 6.4] beats per minute (BPM)). Matrix-based and machine learning methods often entail higher computational cost without proportional performance gains in wearable contexts (error range of [1.07, 6.4] BPM for PPG signals; accuracy in range of $[95.4, 99.96]$% for ECG). (4) Conclusions: Lightweight signal-processing techniques offer the most favorable trade-off between accuracy and efficiency for wearable implementations, whereas computationally intensive approaches are better suited for edge- or cloud-based processing. Full article

Monitoring surface subsidence in mining areas is essential for geological disaster early warning and safe production. Existing geometric difference methods heavily rely on the local consistency of multi-temporal point clouds. When horizontal displacement and vertical subsidence are coupled, horizontal movements often cause local misalignments, leading to spatial deviations and discrete anomalies in vertical estimations. To address this issue, this paper proposes DL-C2C, a deep learning model for subsidence extraction from bi-temporal ground point clouds. Within a unified framework, the model introduces horizontal displacement as an auxiliary constraint into the vertical solving process, effectively improving the stability of vertical subsidence estimation through continuous cross-temporal alignment and correlation updating. For feature extraction, DL-C2C employs a PointConv multi-scale pyramid combined with a proposed scale-adaptive Transformer to enhance cross-scale information interaction under sparse and non-uniform sampling conditions. Furthermore, the network constructs dynamic local associations through iterative alignment within a recursive framework, and introduces diffusion-based residual correction at the fine-scale stage to compensate for detail errors at subsidence basin boundaries and in data-missing regions. Experiments on simulated and real-world datasets—covering aeolian sand and mountainous gully landforms—demonstrate that the method achieves mining 3D error (M3DE) of 0.16 cm and 0.22 cm in simulated scenarios. In real-world mining area validations, compared to existing methods, DL-C2C significantly reduces discrete anomalous points, yields an error distribution closer to zero, and exhibits superior performance in boundary transition continuity and non-subsidence area stability. In conclusion, this model provides reliable technical support for large-scale, high-precision intelligent monitoring of geological disasters in mining areas. Full article

We analyze the evolution of atmospheric and surface physical properties over the region of the Earth observed by the Meteosat Second Generation (MSG) satellites during the period 2005–2024. Long-term changes are detected in the observed radiances, with a decrease in the solar domain (−1.3%) and an increase in the thermal infrared domain (+0.4%), consistent with trends reported by independent broadband radiometers such as CERES. The outgoing solar radiance (OSR) exhibits a marked decline, which we associate with a reduction in low-level cloud cover within the nominal Meteosat field of view (MFoV) centered at 0° longitude. Changes in atmospheric CO₂ concentration also contribute to the observed radiative imbalance at the top of the atmosphere (TOA). Instrument calibration stability and inter-satellite homogenization across the MSG series are explicitly addressed, enabling the detection of robust interdecadal signals. By subdividing the MFoV into 60 regional sectors, we characterize spatial variations in cloud amount at low and high atmospheric levels and relate these changes to regional TOA radiative imbalances and concurrent variations in Atlantic sea surface temperature (SSTs). The spectral information provided by SEVIRI allows a more detailed attribution of radiative changes than broadband observations alone from other instruments. In particular, radiances measured in the atmospheric split-window region near 11 µm are shown to be sensitive to variations in low-tropospheric humidity, which exhibits a widespread decadal-scale increase. The results indicate a close coupling between cloud-cover changes, radiative fluxes, and SST evolution on the recent interdecadal time scale. The observed decrease in low-level total cloud cover is independently in line with ECMWF ERA5 reanalysis data. These findings highlight the value of long, stable geostationary observations for investigating atmosphere–ocean interactions and their role in regional climate variability. Full article

Lake ice monitoring is critical for assessing climate change, but in-situ observations are often limited. Sentinel-1 Synthetic Aperture Radar (SAR) data is a strong method for ice detection because it is not restricted by cloud cover and it is readily available. However, SAR-based classification can be affected by atmospheric and surface-related noise. This study examines the impact of noise on machine learning-based lake ice detection over Lake Śniardwy, Poland, using Sentinel-1 Vertical-Vertical (VV) and Vertical-Horizontal (VH) backscatter data. Binary logistic regression models were trained on scenes with strong class separability between ice and water and then validated on separate low- and high-noise datasets. The models achieved high accuracy under low-noise scenes, reaching up to 96.9%, but performed poorly on high-noise scenes. The results show that wind-related surface roughness and associated atmospheric conditions can significantly reduce classification reliability. Comparison with backscatter from a nearby coniferous forest confirmed that the main disturbances were concentrated over the lake surface. The study highlights the importance of careful scene selection and noise assessment in SAR-based lake ice classification. Full article

High-precision 3D building data are pivotal for smart city development, urban planning, and disaster management. However, large-scale building extraction from airborne LiDAR point clouds remains challenging due to semantic ambiguity, uneven point density, and complex architectural structures. To address these limitations, we propose a novel framework integrating geometric topology perception with cross-dimensional attention mechanisms within a Sparse Voxel Convolutional Neural Network (SPVCNN). The key contributions include: (1) an enhanced LaserMix++ multi-scale hybrid augmentation strategy featuring cross-scene block replacement, ground normal–constrained rotation, and non-uniform scaling; (2) a dual-branch SPVCNN architecture embedding a collaborative module of Geometric Self-Attention (GSA) and Cross-Space Residual Attention (CSRA) to preserve topological consistency and enable cross-dimensional feature interaction; and (3) a Boundary Enhancement Module (BEM) specifically designed to resolve boundary ambiguity and overlapping predictions. Evaluated on a 177 km² dataset covering Washington, D.C., our method significantly outperforms the baseline SPVCNN, improving accuracy by 12.04 percentage points (0.8212 to 0.9416) and Intersection over Union (IoU) by 9.96 percentage points (0.866 to 0.9656). Furthermore, it surpasses mainstream networks such as Cylinder3D and MinkResNet by over 50% in absolute accuracy gain. These results demonstrate the effectiveness of synergistically combining geometric perception with adaptive attention for robust building extraction from large-scale LiDAR data. Full article

Remote sensing has become a cornerstone of modern agricultural monitoring, addressing the dual challenges of increasing production while ensuring environmental sustainability. Based on a conceptual framework developed over the past decade, key application areas include yield estimation, phenology, stress assessment (e.g., drought), crop mapping, and land-use change detection. In Central Europe, regionally specific conditions such as fragmented land ownership, small and irregular plots, and high climate variability shape these applications. Annual field crops, such as cereals, oilseeds, maize, and forage crops dominate production and represent the primary focus of monitoring efforts. Optical data from Sentinel-2 are effective for mapping crop types and analyzing phenology, especially when dense time series are available. However, persistent cloud cover during critical growth phases limits the effectiveness of optical approaches, prompting the integration of radar data from Sentinel-1. Multi-sensor strategies increase the robustness of classification and temporal continuity, supporting monitoring under adverse conditions. Reliable reference data from systems such as the Land Parcel Identification System enables parcel-level validation and facilitates object-oriented analyses in line with management needs. Future developments will increasingly rely on advanced time-series analysis, machine learning, and the integration of agrometeorological and crop model data. As climate change intensifies drought frequency and yield variability, remote sensing will play a pivotal role in enabling near-real-time monitoring and decision support within the evolving landscape of digital agriculture ecosystems. The aim of this review article is to provide an overview of crop monitoring in the Central European region over approximately the past fifteen years, emphasizing trends in subsequent technological and procedural developments. Full article

Accurate and long-horizon forecasting is essential for reliable planning of solar and wind power generation. Most existing models rely on high-resolution meteorological data, which are often unavailable in practical microgrid environments. This study proposes SNORF, a solar–wind neural ordinary differential equation model that uses only basic meteorological variables such as solar radiation, cloud cover, and wind speed together with lagged generation values. However, reliance on lagged generation inherently limits the effective forecasting horizon of conventional models. To address this limitation, SNORF extends the forecasting horizon while maintaining accuracy and stability through a rolling forecasting framework with a bounded input-dependent drift function and repeated Euler integration that promotes smooth hidden-state dynamics. SNORF supports both deterministic point forecasting and probabilistic forecasting through quantile-based loss functions. Experiments on solar and wind power datasets show that SNORF consistently outperforms representative time-series forecasting models. For solar forecasting, SNORF reduces RMSE and MAE by 15.35% and 16.93% on average compared with baseline models. For wind forecasting, the improvements reach 44.77% in RMSE and 52.85% in MAE on average. Furthermore, when evaluated under the official protocol of the 2024 IEEE Hybrid Energy Forecasting and Trading Competition, SNORF achieves a top 4% ranking using only the officially provided basic weather dataset, demonstrating its practical applicability to renewable energy forecasting in microgrids and virtual power plants. Full article

This study experimentally evaluates the thermal performance of a compact parabolic trough solar collector (PTSC) operating under actual solar conditions in Chachapoyas, a high-Andean city in northern Peru characterized by frequent cloud cover and variable irradiance. Despite the growing interest in solar thermal systems, few studies have assessed PTC behavior under high-altitude, diffuse radiation conditions typical of Andean regions. The PTSC, aligned along the north–south axis and equipped with a manual solar tracking system, was monitored for 30 consecutive days. Solar irradiance, ambient temperature, and water inlet/outlet temperatures were recorded at 30 min intervals using a DAVIS Vantage Pro Plus weather station and infrared thermometers (±0.5 °C accuracy). Thermal efficiency was determined from the ratio of useful heat gain to incident solar energy, based on instantaneous irradiance data. Results showed peak irradiance values of 1000 W m⁻² and maximum outlet water temperatures of 85 °C, achieving an average efficiency of 68 ± 2.5%. The collector maintained stable operation even under fluctuating radiation, confirming its suitability for domestic hot-water and low-temperature industrial applications. These findings provide the first experimental evidence of efficient solar-thermal conversion in cloudy highland environments of Peru, supporting the deployment of decentralized renewable energy systems in the Andean region. Full article

Satellite observations of methane are frequently compromised by extensive data gaps caused by cloud cover and aerosol contamination, limiting their utility for continuous regional monitoring. To reconstruct these spatiotemporal discontinuities, this study developed the Stacked Ensemble Reconstruction Framework for Methane (SERF-XCH₄). By integrating Sentinel-5P TROPOMI retrievals with 25 multi-source environmental covariates, we generated a spatiotemporally continuous, high-resolution (0.1°) monthly dataset (SERF-XCH₄-IM) for Inner Mongolia spanning 2019 to 2023. Comprehensive validation demonstrates that the framework achieves exceptional predictive fidelity with a Coefficient of Determination ($R^2$) of 0.93 and a Root Mean Square Error (RMSE) of 7.89 ppb, significantly surpassing the performance of individual base learners and traditional interpolation methods. Furthermore, spatial block cross-validation confirmed robust generalization capabilities ($R^2=0.90$) in data-void regions. To unravel the “black box” of the model, SHapley Additive exPlanations (SHAP) analysis was employed, revealing that temporal factors (contributing 63.9%), air temperature, and elevation are the dominant drivers governing XCH₄ variability. Spatiotemporal analysis further identified the Hulunbuir region as a significant growth “hotspot” with an annual increase rate exceeding 18.5 ppb/yr, a trend primarily driven by intensified emissions during the autumn and winter seasons. Consequently, this framework establishes a high-precision, interpretable paradigm for regional methane monitoring and geo-information reconstruction. Full article

This study develops a machine learning-based predictive model for identifying high-level clouds (HLCs). The model uses meteorological parameters as input features and is trained against human-recorded meteorological observations. A statistical analysis of the relationship between two independent methods of registering HLCs—lidar and meteorological observations—has been performed. Optimal thresholds for the total amount of cloud cover, at which meteorological data are consistent with lidar data, have been determined. The results demonstrate the promising performance of ML models in identifying the links between weather conditions and the probability of HLC detection, which is confirmed by ROC AUC (Area Under the Curve of the Receiver Operating Characteristic) values in the range of 0.87–0.88 for the presence and 0.77–0.78 for the absence of clouds, as well as balanced metrics Precision, Recall, and F1. The XGBoost (eXtreme Gradient Boosting) model proved to be the most robust, demonstrating the ability to effectively integrate data of various types for reliable prediction in various conditions. Full article

Retrieving farmland soil water content with both high accuracy and physical interpretability remains a significant challenge, particularly for winter wheat. To bridge the gap between purely empirical data-driven approaches and mechanistic scattering models, this study proposed a novel hybrid framework that integrates an improved water cloud model (IWCM) with machine learning algorithms. Multi-modal unmanned aerial vehicle (UAV) experiments were conducted during the heading stage of winter wheat over two consecutive years (2024–2025) using a synchronized system equipped with a miniature synthetic aperture radar (MiniSAR) and a multi-spectral sensor. The core innovation of the proposed framework lies in the IWCM, which explicitly decouples vegetation and soil scattering contributions by incorporating fractional vegetation cover, thereby deriving physically meaningful soil backscatter coefficients from complex microwave signals. Unlike traditional methods that treat remote sensing variables as black box inputs, our approach employed these physics-derived features to guide data-driven modeling. Four feature input schemes including spectral reflectance, vegetation indices, MiniSAR polarimetric parameters, and their multi-source fusion were systematically evaluated using back propagation neural network (BPNN) and random forest (RF) regressors. The results demonstrated that the proposed framework significantly enhances retrieval performance. Notably, the RF model driven by spectral band reflectance within this physically constrained architecture achieved optimal accuracy, with a coefficient of determination (R²) of 0.865, a mean absolute error (MAE) of 0.0152, and a root mean square error (RMSE) of 0.0197. Compared to purely empirical approaches, the IWCM significantly improved the physical interpretability of microwave polarimetric characteristics, enabling the multi-source data fusion to better represent the interactions among vegetation, soil, and microwave scattering. This study demonstrated that integrating mechanistic models with multi-source UAV remote sensing data not only improves soil water content retrieval accuracy in winter wheat fields but also provides a valuable reference for developing operationally applicable and physically interpretable farmland soil water content monitoring systems. Full article

This study examines the dynamics of aridity in Mexico in relation to El Niño–Southern Oscillation (ENSO) phases (El Niño, La Niña and neutral conditions) between 1999 and 2024. The aim is to identify ecosystems that are exposed to emerging aridification. Aridity was estimated using the Lang index at a resolution of 1 km across nearly two million grid cells. Aridity intensity and long-term trends were calculated and analysed by ENSO phase to identify areas of double exposure. Over 60% of Mexico is classified as arid or semi-arid. During El Niño, up to 100% of the central and southern regions exhibit increased aridity, affecting an area of 290,852 km² (14.7%), where both the intensity and the trend are high. Although La Niña typically brings wetter conditions, 150,022 km² (7.6%) still exhibit increasing aridity. Areas exposed to aridity under both ENSO phases cover 16,224 km² (0.8%), particularly affecting cloud forests, secondary vegetation and agricultural landscapes. This suggests a process of persistent aridification. The average arid area was 64% ± 7.51% during El Niño, 67% ± 1.44% during La Niña and 64% ± 8.14% during neutral years, indicating substantial variability beyond phase dependence. These findings reveal a complex, non-linear ENSO influence and suggest chronic hydroclimatic stress in some regions. Understanding which ecosystems experience recurrent aridity is crucial for effective water management, biodiversity conservation, and climate adaptation planning. Full article