Search Results (1,471)

Elevated concentrations of fine particulate matter ($P{M}_{2.5}$) are a critical threat to respiratory health worldwide. Therefore, there is an urgent need for precise urban forecasting systems for public health management. Technological advancements in the domains of continuous environmental monitoring and deep learning have enabled large-scale data acquisition, processing, and modeling. Existing predictive models typically depend on auxiliary meteorological inputs, which are frequently inaccessible within standard ground-level monitoring networks. Furthermore, conventional approaches often fail to adequately capture the complex spatio-temporal interactions of pollutants. To address these limitations, this study presents the Spatio-Temporal Endogenous Attention Network (STEA-Net), a forecasting framework designed to operate exclusively without weather variables. Validated on a comprehensive multi-year historical dataset (Jan 2015–Feb 2020) from diverse monitoring stations in India, STEA-Net employs a hybrid adjacency matrix that integrates physical geographical distances with functional clustering to accurately map pollutant transport pathways. Utilizing this structural map, a Graph Attention Network dynamically evaluates the spatial influence of neighboring nodes, while a Bidirectional LSTM processes the underlying temporal sequences. Experimental results demonstrate that STEA-Net substantially surpasses traditional machine learning algorithms and provides competitive performance against advanced deep learning baselines. The proposed model achieves a peak Coefficient of Determination ($R^2$) of 0.9294 (5-seed average: $0.9273\pm 0.0023$) and a peak RMSE of 14.38 µg/m³ (5-seed average: $14.59\pm 0.23$ µg/m³), effectively adapting to the dynamic volatility of urban pollution levels. The model exhibits architectural stability with a Monte Carlo dropout verified deviation of $\pm 2.22$ µg/m³. This research provides a forecasting architecture that retains competitive predictive performance under the strict operational constraint of meteorology-free deployment in resource-constrained urban monitoring environments. Full article

Machine learning approaches to wildfire spread prediction are constrained by the lack of standardized, multi-source, spatiotemporal datasets that fuse terrain, weather, and fire-state information into a single ML-ready format. We present WildfireCube, a reproducible event-centric pipeline and methodology for constructing dense fourth-order spatiotemporal tensors of shape (T, C, H, W) at 30 m spatial and 3 h temporal resolution. Following the analysis-ready data convention established in the Earth Observation community, the pipeline fuses four open data sources: the Copernicus GLO-30 Digital Elevation Model for static terrain derivatives, ERA5-Land reanalysis for hourly weather forcing, Sentinel-2 Level-2A imagery for spectral vegetation and burn-severity indices, and NASA FIRMS active-fire hotspot detections for fire-state reconstruction via ordinary kriging. The resulting 13-channel normalized tensor separates causal drivers into three physically motivated groups: static landscape controls (elevation, slope, aspect, fuel load), dynamic atmospheric forcings (wind components, temperature, precipitation), and evolving fire state (fire-front mask, burn severity, fractional burn, observation confidence). A physics-informed normalization framework maps all channels to bounded ranges using fixed physical constants rather than sample statistics, ensuring cross-event comparability and exact invertibility. We demonstrate the pipeline on 13 wildfire events across the United States, Canada, and Greece (2017–2023), producing a processed catalog exceeding 300 GB compressed and spanning a 14-fold range in burned area, a 27 °C range in mean temperature, and different fire regimes. Event tensors are stored in chunked Zarr archives with Zstandard compression, achieving a 2.58× compression ratio. As future work, the pipeline will be applied to a 40-event target catalog projected to exceed 2 TB of raw data, providing the multi-regime diversity and scale required for training robust deep learning models for spatiotemporal wildfire prediction. Full article

The escalating frequency of extreme weather events poses severe threats to power system security, often resulting in catastrophic economic and societal consequences. As modern information and communication technologies (ICTs) integrate deeply with power grids, post-disaster communication failures and electrical faults become increasingly interdependent, complicating the restoration of distribution cyber–physical systems (CPSs). To bridge the gap where conventional Unmanned Aerial Vehicle (UAV)-enabled emergency communication ignores coordination with power system restoration, this paper proposes a coordinated recovery method featuring a two-stage UAV deployment strategy. First, a coupled cyber–physical model is established to characterize the cross-layer interaction mechanisms. On this basis, a bi-level optimization framework is developed: the upper level formulates a dynamic two-stage UAV deployment strategy to minimize the mobilization of resources, while the lower level executes network topology reconfiguration to maximize weighted load restoration, constrained by the recovered communication coverage. Simulation results on a modified IEEE 33-bus system demonstrate that the proposed method significantly enhances restoration efficiency. Compared with conventional schemes, the cumulative load loss rate is reduced by 15.75% and 2.42% across different scenarios; the two-stage UAV deployment method achieves a time reduction of 67.23%, 21.40% and 71.56%, validating the superior performance of the coordinated recovery strategy in disaster-stricken CPS. Full article

Autonomous driving systems face critical perception failures in dense fog, where conventional RGB cameras suffer from severe degradation due to atmospheric scattering and reduced visibility. This paper presents an adaptive multi-modal fusion framework that synergistically integrates gated imaging with 3D LiDAR point clouds to achieve robust obstacle detection under visibility conditions as low as 50 m. Unlike standard cameras that passively capture scattered ambient light, gated cameras employ time-synchronized active illumination to physically filter backscattered photons, preserving structural features even in low-visibility scenarios. We propose a novel Adaptive Feature-Weighting Network (AFW-Net) that dynamically adjusts sensor modality contributions based on real-time environmental degradation assessment. The framework incorporates three key innovations: (1) a cross-modal feature extraction module that exploits the complementary physical properties of gated imaging and LiDAR, (2) an attention-based adaptive fusion mechanism that quantifies per-modality reliability through uncertainty estimation, and (3) a degradation-aware training strategy using weather-specific augmentation. Extensive experiments on the Princeton Automated Driving Dataset demonstrate that our approach maintains detection average precision (AP) above 82% under dense fog conditions (50 m visibility), representing a 23.7% improvement over state-of-the-art RGB-LiDAR fusion methods that exhibit substantial performance degradation to 58.4% AP. Ablation studies validate the necessity of each component, and cross-dataset evaluation confirms the generalization capability of the proposed framework. The adaptive weighting mechanism proves particularly effective, dynamically rebalancing modality contributions across the gated imaging and LiDAR branches while maintaining LiDAR geometric constraints. This work establishes a robust perception paradigm for safety-critical autonomous systems operating in low-visibility environmental conditions. Full article

Solar filaments (and their limb counterparts, prominences) are critical tracers of the Sun’s magnetic topology and key precursors to coronal mass ejections (CMEs). Precise identification and continuous tracking of these features are essential for understanding solar eruptive mechanisms and improving space weather forecasting. This systematic review evaluates the evolution of automated detection methodologies, addressing the challenge of processing the exponentially growing volume of high-resolution solar observations. We identify deep learning architectures, particularly U-Net variants and Mask R-CNN, as the most promising current paradigms. Compared to traditional image processing, these data-driven models demonstrate superior robustness against noise and variable observing conditions, achieving high-precision segmentation (>90% accuracy) with sub-second inference speeds. This leap in computational efficiency and accuracy directly facilitates real-time operational monitoring and enables large-scale statistical analysis of filament evolution across solar cycles. We conclude that future breakthroughs lie in developing physics-informed AI and standardized benchmarks to bridge the gap between pixel-level segmentation and physical interpretation, ultimately creating detection systems that are both operationally reliable and scientifically meaningful. Full article

Adobe construction, as part of earthen architecture, is a traditional building technique that is widely used but particularly vulnerable to the effects of water and other climatic factors. This article analyses the physical and mechanical behaviour of three different grain sizes of adobe specimens, classified according to the predominant presence of coarse aggregates (CA), fine aggregates (FA), and fine aggregates with plant fibres (AF). In order to assess their response to climatic scenarios, these specimens are subjected to wetting–drying cycles (3, 5, and 7 cycles) and accelerated weathering tests (E) under controlled laboratory conditions. The main objective is to determine the influence of particle size distribution and the incorporation of plant fibres on the strength, stiffness, durability, and hydraulic behaviour of the material. For this purpose, an experimental programme was developed based on compression, modulus of elasticity, ultrasonic, abrasion, hydraulic erosion, and capillary absorption tests, and carried out at different stages of deterioration. Thus, six specimens were analysed for each of the five time points studied (0, 3, 5, 7, E) and for each proposed particle size distributions, giving a total of 450 samples analysed. The results show that the coarse mix exhibits greater overall mechanical stability, whereas the fine mix is more sensitive to the action of water. Although the addition of fibres improves ductility and resistance to surface erosion, it alters the porous structure of the material. Overall, the results confirm that particle size distribution and fibre reinforcement decisively influence the durability of adobe. Full article

The risk of wildfires is increasing due to high temperatures and dry weather conditions caused by climate change. Outbreaks and spread of wildfires are usually conditioned by weather, topography, and fuel characteristics. In the Republic of Korea (hereafter, the ROK), most wildfires are caused by anthropogenic factors rather than natural ones. However, the current forest fire forecasting system being operated in the ROK does not account for anthropogenic factors. To analyze the impact of human and physical factors on wildfire occurrence, a binary logistic regression model was constructed using data from the Gangwon and Gyeongbuk provinces from January 2022 to August 2025. The dependent variable was defined as the occurrence of a wildfire, while the independent variables comprised meteorological, seasonal, stand, and anthropogenic factors. To address multicollinearity, variables with high correlation coefficients were excluded from the independent variables, which were selected by three estimating approaches, including logistic regression and two machine learning techniques (namely, Random Forest and XGBoost). With machine learning, the variables with high feature importance were identified. The explanatory power of the logistic regression analysis with independent variables selected by the machine learning models was about 1.3 times higher than that of the model using variables adjusted solely for multicollinearity. The results of logistic regression analysis revealed that weather and coniferous forests are the most important factors fostering wildfires, while the mean stand age was the most significant factor in hindering wildfires. Among the anthropogenic factors, forest road density acted as a suppressor of wildfire spread rather than a promoter of occurrence. Conversely, trail density tends to increase the risk of wildfire occurrence. Among forest management activities, plantation forests may increase the risk of forest fires, although this remains uncertain. These findings suggest that preventing wildfires requires a paradigm shift in forest resource management policies, including extending forest rotation ages and converting coniferous forests to broadleaf forests. Meanwhile, it also indicates the need to restrict the expansion of hiking trails and improve regulations regarding hiker access and behavior to prevent wildfires. Full article

The ocean–atmosphere turbulent heat exchange plays a critical role in the energy and moisture budgets of the Tropical Atlantic Ocean (TAO) and in weather and climate forecasts. However, its estimation strongly depends on the choice of bulk parameterization, as direct in situ measurements are sparse. This study evaluates sensible (Hs) and latent (Hl) heat fluxes derived from three bulk parameterization schemes used operationally in models at the Brazilian Center for Weather Forecast and Climate Studies (CPTEC) of the National Institute for Space Research (INPE), Brazil: the Brazilian Atmospheric Model (BAM), the Modular Ocean Model version 6 (MOM6), and the Weather Research and Forecasting (WRF) model. Using daily in situ observations from seven Prediction and Research Moored Array in the Tropical Atlantic (PIRATA) buoys across the TAO during 1997–2023, we computed monthly mean fluxes and compared them against the Coupled Ocean–atmosphere Response Experiment (COARE) algorithm version 3.0b (COARE 3.0b) reference. COARE version 3.6 (COARE 3.6) and European Centre for Medium-Range Weather Forecast (ECMWF) Reanalysis 5th generation (ERA5) data were included as additional benchmarks. All offline schemes were forced with identical buoy data, isolating differences in internal physical assumptions. Hl is approximately one order of magnitude larger than Hs across all sites, and inter-scheme differences are substantially larger for Hl (±50 W∙m⁻²) than for Hs (±5 W∙m⁻²). All schemes reproduce the seasonal cycle linked to the Intertropical Convergence Zone (ITCZ) migration and trade-wind variability, with correlations generally exceeding 0.8 (p < 0.001) for most buoys. However, systematic magnitude biases remain. The Coordinated Ocean Research Experiments (CORE) bulk formulation implemented in MOM6 (MOM6-CORE) shows high temporal correlation (often r ≈ 1.0) but a persistent negative bias for both Hs and Hl (e.g., B1 Hl bias = −24.0 W∙m⁻²), indicating weaker turbulent exchange relative to COARE 3.0b. BAM overestimates Hs (by 1–3 W∙m⁻²) and underestimates Hl at most northern and southern sites, while the parametrization of the Yonsei University (YSU) implemented in the WRF model (WRF-YSU) amplifies Hs variability intermittently, particularly at the equator (B4). As expected, COARE 3.6 remains the closest to the reference (differences < 1 W∙m⁻² for Hs and <7 W∙m⁻² for Hl; r ≈ 0.99). ERA5 captures temporal variability well (r ≈ 0.7–0.9) but systematically overestimates Hl (positive bias up to +47.6 W∙m⁻² at B7), implying stronger evaporative cooling. Buoy-specific regimes modulate skill. The choice of bulk formulation thus remains a first-order source of uncertainty in turbulent heat flux estimates over the TAO, with direct implications for mixed-layer heat budgets, SST evolution, and coupled ocean–atmosphere variability. MOM6-CORE provides the most consistent performance relative to the COARE reference and emerges as the most robust option for operational applications at CPTEC/INPE. The findings also provide guidance for improving the representation of ocean–atmosphere turbulent exchanges in MONAN (Model for Ocean-Land-Atmosphere Prediction), the new Brazilian Earth System Model under development for weather and climate prediction. Full article

Transboundary fine particulate matter (PM2.5) in Southeast Asia is commonly assessed using static source–receptor frameworks or descriptive associations that may not resolve how directional dependence changes through time under shifting meteorological conditions. This study examines regional PM2.5 as a time-varying, meteorology-adjusted directional coupling system using monthly data for 2003–2024 from the Copernicus Atmosphere Monitoring Service (CAMS) reanalysis, European Centre for Medium-Range Weather Forecasts Reanalysis v5 (ERA5) meteorological covariates, climate controls, and administrative aggregation. Using a rolling-window directed network framework based on Peter and Clark Momentary Conditional Independence (PCMCI) causal discovery, we inferred lagged conditional-dependence networks from covariate-adjusted PM2.5 anomalies and summarized their structure at national and first-order administrative levels. The inferred network structure varies over time but retains measurable continuity across rolling windows. At the country level, cross-border links consistently account for a large share of the directed structure, indicating that PM2.5 variability within the study domain is strongly shaped by transboundary coupling rather than by country-contained dynamics alone. A recurrent backbone of country-level directional coupling corridors emerges, including persistent links among China, Indonesia, Myanmar, and Thailand. At the first administrative level, stable gateways and receptor basins become more evident, especially the bidirectional coupling corridor between Yunnan Province, China, and Shan State, Myanmar, which appears throughout the full window sequence. These results show that subnational structure can reveal transport-relevant coupling patterns that national summaries may conceal. The framework provides an interpretable basis for corridor-oriented monitoring and regime-aware early warning, while the inferred links should be interpreted as directional statistical dependence rather than direct emissions attribution or resolved physical transport pathways. Full article

Traditional wireless communication signals are often susceptible to physical obstructions and background noise in complex geographical environments or adverse weather conditions, hindering stable and reliable data transmission. Ultraviolet communication (UVC) offers a compelling solution; its unique scattering mechanism and low background noise characteristics facilitate robust communication under non-line-of-sight (NLOS) conditions. At present, there remains a relative lack of comprehensive reviews spanning UVC, including fundamental theory, physical devices, channel models and networking technologies. This review synthesizes the current state of global research, providing a systematic overview of the background, advantages and application scenarios of UVC. It examines the hardware characteristics of light sources and detectors, evaluates NLOS scattering channel models, analyzes key signal processing techniques, including modulation/demodulation, coding/decoding and multiple-input multiple-output technology. Furthermore, this review conducts an in-depth analysis of multi-user networking protocols and three-dimensional topology control mechanisms. Finally, it identifies the prevailing technical challenges and outlines promising directions for future development. Full article

Accurate forecasting of urban electric vehicle (EV) charging demand is important for power system operation, sustainable transport electrification, and charging infrastructure planning. However, this task remains challenging because EV charging demand is shaped by temporal usage patterns as well as changing relationships among weather conditions, operational factors, and historical charging behavior. Many existing forecasting models treat these explanatory variables mainly as parallel inputs, while their mutual relationships are often predefined, simplified, or left implicit in the temporal learning process. To support AI-driven charging demand management, this study proposes an adaptive graph-based recurrent network (A-GRN) for city-level aggregated EV charging-load forecasting. In the proposed framework, key explanatory variables are represented as feature nodes, and their connections are learned through an adaptive adjacency matrix rather than a fixed spatial topology. The adaptive graph neural network (AGN) module captures feature-level interactions, while a dual-path gated recurrent unit module (DG-GRU) extracts temporal representations from the charging-load sequence. Experiments on a city-level EV charging dataset show that A-GRN outperforms several baseline models, including naive persistence forecasting, GRU, LSTM, BiGRU, TCN, and GCN. Compared with the BiGRU baseline, A-GRN reduces MAE, MSE, and RMSE by 31.36%, 34.65%, and 20.48%, respectively. In the original physical unit, the MAE is reduced from 187.43 kWh to 128.64 kWh, and the RMSE is reduced from 222.69 kWh to 177.08 kWh. The results indicate that feature-level graph learning can improve short-term EV charging-load forecasting, especially when the target is an aggregated urban load rather than the load of a single charging station. The proposed model provides a data-driven forecasting tool for sustainable urban charging demand management, low-carbon transport operation, and charging infrastructure planning. Full article

Accurate simulation of landfalling typhoons is essential for urban resilience in the densely populated Pearl River Delta. Using Super Typhoon Mangkhut (2018) as a case study, this paper evaluates the Weather Research and Forecasting (WRF) model through a contribution analysis designed to disentangle the roles of surface layer, planetary boundary layer (PBL), urban canopy model (UCM), and eddy-coefficient/diffusion closure parameterizations in wind-hazard prediction. Model results are validated against observations at the Hong Kong Observatory headquarters (HKO) and King’s Park (KP) stations, demonstrating that the hierarchy of physical controls is strongly metric-dependent. Substantial and structured spread is found among the tested configurations. Controlled comparisons show that PBL selection is the primary driver of variability in peak timing and high-wind persistence, whereas surface-layer formulation and diffusion closure exert secondary but systematic influences by shifting distributional centers and reshaping variability and upper tails. Urban canopy effects are comparatively weaker in aggregate but become more apparent during the impact and recovery phases. Overall, the results confirm that no single parameterization is consistently optimal across all metrics and motivate a multi-objective physics-selection strategy, in which multi-physics ensembles are used to better represent uncertainty in wind-event duration and associated loading risks in complex urban environments. Full article

Accurate short-term estimation of clear-sky Global Horizontal Irradiance (GHI) is vital for solar resource assessment and grid operations, yet existing methods rely on sparse radiometers and coarse global weather reanalysis (e.g., MERRA-2 at 50–70 km spatial resolution with 1 month latency). To achieve scalability in high-precision estimation, we propose a framework that removes dependence on ground measurements by combining multi-satellite observations with reanalysis variables in a physics-supervised machine-learning paradigm. We developed a multi-source-fused high-resolution environmental dataset with 5 min granularity and 1 km spatial precision, incorporating Geostationary Operational Environmental Satellite (GOES-16) observations, polar-orbiting satellite (AURA) data, and MERRA-2 reanalysis. As supervisory physics, we harmonized two complementary parameterized radiative transfer models (MAC2 and REST2V5). The harmonized GHI estimates are used as training labels for a Multilayer Perceptron (MLP) and a Residual Long Short-Term Memory (LSTM) network model. The trained MLP model achieved a root mean square error (RMSE) of 66.67 $\mathrm{W}/{\mathrm{m}}^2$, representing a 7.50% reduction over the conventional MERRA-2-driven baseline. For 30-min-ahead forecasting, the LSTM model reduced RMSE by 29.37% over the persistence baseline. Evaluated at four climatically diverse U.S. sites, the system achieves ground-sensor-like accuracy and is deployable anywhere within GOES-16 coverage. Full article

Wood strength, renewability and appearance make it one of the most preferred and widely used natural materials in structural and cultural applications. The gradual degradation of wood from abiotic and biotic factors has an adverse impact on its structural durability and service life. This study investigates the effect of surface treatment of wood of the invasive tree species of tree-of-heaven, through short-term immersion in an acrylic polymer (Paraloid-B72) containing silica dioxide (SiO₂) nanoparticles at low concentrations (0–4% w/v) to impart hydrophobic behavior and weathering resistance. FTIR analysis confirmed the successful incorporation of the acrylic polymer and silica nanoparticles within the wood structure without altering the chemical integrity of the substrate. For both treated and untreated wood specimens, the physical properties (density, equilibrium moisture content, surface roughness, color-parameters), hygroscopic properties (swelling/absorption, contact angle) and weathering resistance tests were conducted using xenon-arc combined with wetting–drying cycles. The findings revealed that treated wood has significantly improved hydrophobic performance and dimensional stability, reducing moisture uptake. Treatment significantly increased the samples’ resistance to artificial weathering, with the effectiveness dependent on nanoparticle concentration. Although moderate surface color changes were observed in treated samples (compared to untreated ones), during their exposure to weathering, reduced lightness and slight increases in red and yellow chromatic coordinates were observed, with treated specimens exhibiting higher color stability during aging. Nevertheless, surface roughness increased significantly by the treatment, slightly restricting the method when a highly smooth surface touch is required. The proposed modification method appears promising to prolong the wooden structures’ service-life, meanwhile inspiring modern strategies for conserving historical timber structures that cannot be moved and should be protected by applying less invasive protective methods. Full article

Cloud properties such as cloud optical thickness (COT) and cloud effective radius (CER) are essential for weather forecasting, climate monitoring, and Earth’s energy budget estimation. Traditional physics-based retrievals using independent pixel approximation (IPA) often incur biases due to three-dimensional radiative effects. While existing deep learning approaches reduce these biases, they demand large annotated datasets and high computational cost. This study frames cloud property retrieval as an information-limited learning problem (limited spectral information and limited training samples) and incorporates CloudUNet with Attention Module (CAM), a compact deep learning model with channel attention for joint estimation of COT, CER, and cloud mask from bi-spectral radiance observations. Using synthetic datasets from large-eddy simulation (LES) cloud fields, CAM outperforms state-of-the-art models in both direct radiance-based retrieval and IPA correction, achieving 38% better performance in terms of mean absolute errors (MAE) and higher correlation with true properties. Ablation studies demonstrate that CAM-based IPA correction achieves 73% and 80% MAE reduction relative to the IPA baseline when using no radiance input and single-band radiance, respectively. Including cloud mask information as input improves COT retrieval across deep learning models (except CAM) but degrades CER retrieval for all models except CAM, which shows a slight 3% MAE improvement. These findings highlight the advantage of joint retrievals of multiple cloud properties and IPA correction models under limited labeled data constraints. Full article