Search Results (484)

Soil erosion and the nutrient loss it causes are core issues threatening sustainable land use in arid and semi-arid regions. In this study, our aim was to reveal the spatiotemporal differentiation characteristics of soil erosion and soil nutrients in an ecologically fragile area of eastern Inner Mongolia—Tuquan County and clarify the relationship between them in order to provide a scientific basis for the precise management of water and soil resources and ecological construction in this region. Based on four sets of remote sensing images and ground observation data from 2012, 2016, 2020, and 2024, the Revised Universal Soil Loss Equation (RUSLE) was used to evaluate the dynamics of soil erosion, statistical methods were employed to analyze the spatial distribution and grade characteristics of soil nutrients (organic carbon, SOC; total nitrogen, TN, total phosphorus, TP) and pH values, and correlation analysis was conducted to explore their association with environmental factors (rainfall erosivity, R; soil erodibility, K; slope length, LS; vegetation cover and management factor, C). Our results demonstrate the following: (1) From 2012 to 2024, the intensity of soil erosion in the study area showed an increasing trend, with the average annual soil erosion modulus increasing from 551.4 t/(km²·a) to 859.6 t/(km²·a), and the high-intensity erosion areas were mainly distributed in the northwest. (2) The soil nutrient content was generally at medium to low levels, with the SOC and TN in the study area mainly categorized as “deficient” and “adequate”. The SOC ranged from 5.8 to 33.8 g·kg⁻¹, with an average content of about 23.5 g·kg⁻¹, while the TN content ranged from 0.45 to 4.63 g·kg⁻¹, with an average content of about 1.50 g·kg⁻¹, and was significantly affected by soil type. (3) There was a significant negative correlation between the soil erosion modulus and the SOC and TN content (p < 0.05), which was a key driving factor for nutrient loss. This conclusion suggests that soil erosion in Tuquan County is intensifying: the risk of nutrient loss is severe, and its spatial pattern is jointly restricted by topography, vegetation cover, and soil background characteristics. Therefore, future ecological engineering should focus on high-intensity erosion areas and combine the prevention of soil and water loss with the conservation of soil fertility in order to achieve sustainable land use in the region. Full article

Agricultural drought threatens rainfed agriculture in northeast Thailand, where variable monsoon rainfall, limited irrigation access, and extensive cropland increase vulnerability. This study developed a Google Earth Engine-based Agricultural Drought Risk Index (ADRI) for the Chi River Basin using six benchmark years (2000, 2005, 2010, 2015, 2020, and 2025). CHIRPS precipitation, MODIS-derived vegetation health, ERA5-Land soil moisture, irrigation accessibility, and agricultural land exposure were normalized and integrated by weighted linear combination. The analysis quantified risk-class areas, irrigated–rainfed contrasts, persistent hotspots, weight sensitivity, and spatial agreement with the official Land Development Department recurring-drought map. Moderate risk dominated most years, but high-risk area expanded to 60.7% in 2015, coincident with severe rainfall deficits during the 2015–2016 El Niño event. Conditions improved in 2020 and 2025 as rainfall, vegetation health, and soil moisture recovered. Rainfed areas consistently had higher ADRI values than irrigated areas, and persistent hotspots were concentrated in southeastern and downstream agricultural zones. The principal spatial and temporal patterns remained stable under ±10% weight perturbations. External validation identified ADRI > 2.90 as the optimal threshold, with raster-level precision, recall, and F1 of 0.779, 0.884, and 0.828, respectively; the 998-point sample produced an F1 of 0.832. ADRI therefore provides a practical basin-scale screening framework for drought monitoring, adaptation prioritization, and agricultural water-management planning. Full article

Quarry waste piles are heterogeneous engineered embankments that are susceptible to slope instability, yet early detection of pre-failure surface changes remains challenging due to complex surface conditions and measurement uncertainty. This study presents an integrated remote sensing-based framework for monitoring quarry waste pile instability by combining multi-temporal change detection with scale-dependent surface roughness analysis. The original contribution of the proposed framework lies in linking displacement-based change detection with multi-scale characterization of surface roughness, enabling both observed surface movement and topographical conditions associated with developing instability to be evaluated within a unified monitoring approach. Multi-epoch Unmanned Aerial Vehicle (UAV)-mounted Light Detection and Ranging (LiDAR) and photogrammetric point clouds were acquired before and after documented failure events at an active quarry site at active quarry sites located northeast of Montreal, Quebec, Canada. The regional climatic conditions, characterized by seasonal freeze–thaw cycles, rapid snowmelt, and periods of heavy rainfall, can promote water infiltration and elevated pore-water pressures, thereby increasing the susceptibility of these heterogeneous waste piles to slope instability. A standardized workflow was implemented, including precision alignment using a Recursive Iterative Closest Point (R-ICP) registration strategy, vegetation filtering with a multiscale CANUPO classifier, and uncertainty quantification through a Level of Detection (LoD) analysis. The resulting LoD thresholds were 10–15 cm for LiDAR-to-LiDAR comparisons and 34–36 cm for mixed-sensor datasets. Multi-scale roughness analysis revealed that zones which later experienced instability exhibited consistently higher and more heterogeneous roughness than adjacent stable areas within a well-defined linear scale range. A roughness-based A/D indicator enabled objective delineation of hazardous zones prior to failure. Post-failure monitoring showed surface smoothing following major displacement, followed by renewed roughness increases associated with secondary movements. These results demonstrate that scale-dependent roughness provides complementary information to displacement-based change detection, enabling potentially unstable areas to be identified and prioritized before substantial displacement becomes evident. The integrated framework can assist quarry managers in targeting field inspections and monitoring efforts toward higher-risk areas and support earlier preventive actions to reduce slope-failure risk. Full article

Landslide susceptibility mapping (LSM) is fundamental to disaster prevention and spatial risk management in mountainous regions. However, conventional LSM approaches that rely mainly on static landslide influencing factors and empirical classification thresholds may have limited temporal relevance and interpretability. In response to these limitations, this study proposed an LSM framework constrained by interferometric synthetic aperture radar (InSAR)-derived deformation information. Wangmo County, Guizhou Province, China, was selected as the study area. Multi-source data, including small baseline subset InSAR (SBAS-InSAR) deformation results, optical remote sensing imagery, geo-environmental factors, and field investigation data, were used to construct and validate four machine learning models: logistic regression (LR), random forest (RF), support vector machine (SVM), and back-propagation neural network (BPNN). The validation results showed that the RF and BPNN models performed better than the LR and SVM models in terms of AUC, accuracy, precision, recall, and F1-score. Accordingly, an RF–BPNN combined model was constructed using an equal-weight averaging strategy. Shapley value analysis indicated that terrain- and rainfall-related factors made dominant contributions to landslide susceptibility prediction, a finding consistent with the landslide development characteristics in the study area. InSAR-derived deformation information was extracted from 31 Sentinel-1A images using SBAS-InSAR. A classification adjustment strategy based on kernel density estimation (KDE) and the Pearson correlation coefficient (PCC) was then used to identify the susceptibility classification scheme with relatively high spatial consistency with deformation activity during the observation period. The optimized classification scheme achieved a PCC value of 0.65, compared with 0.61 for the natural breaks classification, indicating a modest improvement in the spatial consistency between susceptibility zoning and deformation activity. The Xiangle and Namu landslides were used as representative cases to illustrate the adjustment effects of the deformation-constrained classification scheme. The proposed framework provides a practical approach for incorporating observation-period InSAR-derived deformation information into regional LSM and can support landslide monitoring and decision-making in complex terrains. Full article

Olive cultivation is one of the most economically important agricultural activities in the Mediterranean region, yet its productivity is significantly threatened by olive peacock spot, caused by the fungus Cycloconium oleaginum. This disease reduces photosynthetic activity, induces premature defoliation, deteriorates fruit quality, and causes considerable yield losses. Although deep learning has significantly improved automated plant disease diagnosis, most existing approaches rely solely on leaf images and overlook environmental factors that influence disease development and progression. This study proposes an environmentally enriched multimodal deep-learning framework that integrates RGB images of olive leaves with heterogeneous environmental descriptors, including meteorological conditions, soil characteristics, rainfall-derived moisture indicators, vegetation indices, and environmental stress variables obtained from authoritative public data sources. Visual features are extracted using a fine-tuned ResNet50 convolutional neural network, while environmental descriptors are modeled using a multilayer perceptron (MLP). The extracted features are fused at the feature level to simultaneously perform seven-stage disease classification and continuous estimation of lesion coverage and leaf yellowing within a unified multi-task learning framework. Unlike synchronized field-sensor datasets, the proposed dataset combines publicly available olive leaf images with representative environmental observations, providing a reproducible proof-of-concept for multimodal disease diagnosis. Results demonstrate that incorporating environmental information substantially improves disease-stage recognition, achieving an accuracy of 97.77%, a macro F1-score of 0.9809, and a weighted F1-score of 0.9776. The proposed framework also achieved accurate severity estimation, with a mean MAE of 1.29%, RMSE values of 4.86% and 4.75%, and $R^2$ values of 0.957 and 0.969 for lesion coverage and leaf yellowing, respectively. These findings demonstrate the potential of multimodal deep learning to support precision agriculture and intelligent disease monitoring systems. Full article

Fruit preharvest–postharvest quality assessment is essential for precision harvesting, intelligent grading, storage management, and supply-chain loss reduction. However, conventional approaches mainly rely on single-point postharvest inspection and cannot adequately capture the long-term effects of preharvest fruit phenotypes and environmental dynamics on quality formation. To address limited prediction accuracy under few-label conditions, insufficient multimodal fusion, and weak cross-orchard generalization, this study proposes FruitSSL-QNet, a self-supervised multimodal learning framework for jointly modeling preharvest fruit images, environmental sensor time series, and postharvest quality indicators. The framework employs visual masked reconstruction to learn fine-grained phenotype features, including color, texture, lenticel distribution, disease spots, and maturity patterns. Environmental temporal masked modeling is used to capture the cumulative effects of temperature, humidity, light intensity, soil moisture, and rainfall. Bidirectional cross-attention, gated fusion, and contrastive alignment are further integrated to learn complementary and semantically consistent image–environment representations. Experimental results demonstrate that FruitSSL-QNet outperforms SVM, Random Forest, XGBoost, LSTM, GRU, TCN, Transformer, and MM-Transformer across multiple quality assessment tasks. The proposed model achieves a maturity recognition accuracy of $89.6\%$, exceeding MM-Transformer by $4.4$ percentage points. Compared with the corresponding baseline results, the prediction errors for sugar content, firmness, and shelf life are reduced by $21.1\%$, $22.2\%$, and $22.5\%$, respectively. The decay-risk AUC reaches $0.921$, and the cross-site F1-score reaches $0.867$, indicating strong risk discrimination and stable generalization across orchard environments. Ablation experiments further confirm the contributions of visual self-supervision, environmental temporal self-supervision, and cross-modal alignment. Full article

The Mao open-pit coal mine waste dump in Hequ, Shanxi, is a loose, anthropogenic mass accumulated over the original topography. Following a recent sliding and significant settlement event, this dump became the subject of intense stability concerns. Due to the high moisture sensitivity of its interlayered soil and coal gangue structure, rainfall infiltration can reduce internal effective stress, triggering slope instability. Although conventional geological surveys have mapped surface fractures, implementing precise, targeted drainage control requires characterizing the internal geometric structure and preferred seepage directions. To address this, this study integrates electrical resistivity tomography (ERT), surface nuclear magnetic resonance (SNMR), and spontaneous potential (SP) methods. Multiple ERT profiles (270–600 m long) were deployed across several benches at varying elevations, supplemented by fixed-point SNMR sounding over typical low-resistivity anomalies and dense SP grid scanning. The integrated results successfully delineate the internal architecture and seepage characteristics of the dump. Specifically, ERT imaging resolves the primary geoelectrical interface (tentatively inferred as the potential sliding surface) separating the overlying loose mass from the stable underlying strata while mapping the spatial extent of the inferred water accumulation zone (IWAZ). SNMR sounding quantitatively reveals a two-layer water-bearing structure at the specific sounding site, with a deep primary water-bearing zone at 45–80 m depth. Furthermore, SP inversions illuminate the seepage process, demonstrating that meteoric water deflects along the geoelectrical interface to converge laterally toward the central axis at approximately 42°, before transitioning into a high-angle vertical deep infiltration zone (61.7°) within the axial region. These findings suggest a potential engineering direction for remediating surficial fractures and designing subsurface drainage along this 1040 m bench axis, which would mitigate future landslide risks by reducing internal pore water pressure. Full article

The monthly CHIRPS ver. 2 gridded rainfall dataset from 1981 to 2016 was employed to analyze distinct precipitation variability patterns and regimes in semi-arid Botswana. An S-mode eigen analysis was performed on the correlation matrix of the rainfall data to extract principal components. The principal component scores (pc-scores) were further rotated using the Varimax eigen analysis method to yield unique precipitation patterns. The rotated pc-scores indicated three separate sub-regions displaying varying precipitation patterns over time. The application of non-hierarchal clustering (K-means) on the pc-scores identified four distinct zones characterized by unique rainfall patterns. A regional frequency study of rainfall in the sub-regions was performed using L-moments. Probabilistic analysis was utilized to model annual rainfall using six common regional frequency analysis probability distribution functions (pdfs): Pearson Type 3 (PE III); three-parameter Weibull; generalized; extreme value (GEV), normal (GNO), logistic (GLO), and Pareto (GPA). The pdfs that demonstrated the optimal correspondence were determined by the goodness-of-fit test, utilizing the Z-statistic. Each cluster displayed unique pdfs and goodness-of-fit pdfs, with the GLO, GEV, GNO, and Weibull offering the most precise representations. Full article

Purpose: To address the measurement inaccuracies of weighing-type rain gauges caused by environmental disturbances such as vibration, temperature drift, and creep, this study aims to develop a robust error modeling and compensation framework adaptable to complex conditions. Method: A nonlinear error model was constructed by analyzing multi-source disturbance factors and incorporating both linear and nonlinear temperature terms. A BP neural network was employed to compensate for complex error patterns, and several intelligent optimization algorithms (a genetic Algorithm (GA), a particle swarm algorithm (PSO), and a GOOSE algorithm (GOOSE)) were used to enhance training performance. An improved adaptive GOOSE algorithm (ADGOOSE) was further proposed to optimize the BP network by integrating dynamic control coefficients and perturbation-based restart strategies. Results: Experiments under various rainfall intensities and temperatures demonstrated that the ADGOOSE-BP model outperformed traditional filtering and other optimization methods, achieving the lowest RMSE of 0.0494 and the highest R2 of 0.9835. Conclusion: The proposed method effectively models and compensates for environmentally induced errors in weighing rain gauges, demonstrating strong potential as a high-precision, adaptive compensation framework that provides a solid foundation for future field-deployable hydrological monitoring systems. Full article

Landslide susceptibility mapping (LSM) is an essential tool for hazard assessment and land-use planning in landslide-prone areas. This study compares three statistical models—Frequency Ratio (FR), Weight of Evidence (WoE), and Logistic Regression (LR)—with six machine-learning algorithms: Support Vector Machine (SVM), Random Forest (RF), Extreme Gradient Boosting (XGBoost), Artificial Neural Network (ANN), k-Nearest Neighbor (KNN), and Decision Tree (DT), for regional landslide susceptibility assessment in Hiroshima Prefecture, Japan. A balanced dataset comprising 1936 landslide and 1936 non-landslide samples was developed from the 2018 rainfall-induced landslide inventory, utilizing seven conditioning factors: slope angle, profile curvature, aspect, elevation, lithology, soil water index, and 24 h cumulative rainfall. Model performance was assessed using the area under the receiver operating characteristic curve (AUC), accuracy, precision, recall, and F1-score. Among the statistical models, WoE exhibited the highest performance, while SVM provided the most balanced results among the machine-learning models. Both modeling approaches consistently identified lithology and slope angle as the primary controls on landslide occurrence. Independent validation demonstrated comparable predictive performance for both models; however, spatial validation showed that WoE assigned 96.72% of observed landslides to the High and Very High susceptibility classes, compared to 72.54% for SVM. These findings underscore the importance of integrating conventional classification metrics with spatial validation to enhance the evaluation and interpretation of landslide susceptibility models for regional hazard assessment. Full article

Flash floods are among the most destructive natural hazards, often causing substantial loss of life and severe damage to infrastructure and property. Predicting flood-prone areas remains challenging because flood generation is controlled by complex interactions among topographic, hydrological, climatic, and environmental factors. In this study, six machine learning algorithms—Random Forest (RF), Logistic Regression (LR), Support Vector Machine (SVM), Decision Tree Classifier (DTC), AdaBoost, and Artificial Neural Network (ANN)—were developed to predict flash-flood inundation locations using satellite-derived flood inventories from two major rainfall events in Wadi El-Darb and Wadi El-Allaqi, Egypt. Model performance was evaluated using accuracy, precision, recall, and F1-score. During model development, Random Forest and Decision Tree Classifier achieved the highest prediction accuracy (94%), followed by AdaBoost and ANN (92%), while Logistic Regression (89%) and SVM (88%) also produced satisfactory results. To evaluate model generalization, the trained models were independently validated using a rainfall event in Wadi Hodein (Egypt) and a major flash-flood event that occurred in Oman during April 2024. The external validation showed that AdaBoost achieved the highest predictive performance in both validation basins, with accuracies of 87% for Wadi Hodein and 83% for Oman, providing encouraging initial evidence of applicability across hydrologically similar arid watersheds, While AdaBoost and Logistic Regression maintained satisfactory performance during external validation, other algorithms exhibited noticeable reductions in recall and F1-score, particularly in the Oman case study, indicating variability in model generalization across independent watersheds These findings suggest that the proposed framework may support flood susceptibility assessment in ungauged arid environments with comparable hydrological characteristics, although further validation across a wider range of climatic and geological settings is needed. Overall, the results highlight the value of integrating satellite remote sensing with machine learning to support flood hazard assessment, disaster preparedness, early warning systems, and flood risk management in data-scarce regions. Full article

The medium- and low-voltage distribution network is a critical hub connecting the transmission grid and end-users, and its power supply reliability directly determines livelihood security and socio-economic operational efficiency. Typhoon-induced strong winds and rainfall often trigger large-scale power outage risks, severely threatening power grid security and resilience. To achieve the rapid and accurate perception of outage risk areas based on building damage after typhoons, this study proposes the ResiDS-Net method, which infers distribution network outage risk levels by identifying building damage levels. An improved Dempster–Shafer evidence theory is here adopted to fuse the outputs of CM-ResNet50, Inception-V3 and DenseNet121, enhancing perception accuracy. A two-stage “coarse screening–fine judgment” framework using dual datasets is established to quickly identify large-scale suspected power outage areas from building group damage data. To address the issue that equating building damage with power outages reduces judgment accuracy, this study further develops a hierarchical building damage dataset, classifying individual buildings by damage level to achieve precise outage risk identification. Our experiments show that ResiDS-Net achieves 96.66% and 93.00% accuracy on the two datasets, 2.13% and 2.50% higher than nine comparative networks including Inception-V3. The proposed method effectively improves outage risk perception precision and provides a scientific basis for power emergency repair. It should be noted that the proposed method provides building-damage-based outage risk inference for emergency decision support, rather than the direct detection of verified actual outage status. Full article

The cascading damage of expressways induced by extreme heavy rainfall presents a persistent threat to transportation safety and regional sustainable development. To investigate the chain-like evolution characteristics of expressway damage caused by heavy rainfall and to identify precise strategies for mitigating disaster risks by breaking the chain. Firstly, directed causal event pairs were extracted, and clustering generalization was performed on disaster events.; the asymmetric Jaccard index was used to calculate edge weights, thereby establishing a directed causal knowledge graph of disaster chain evolution; Secondly, based on systematic risk assessment and chain-breaking priority indicators, we achieved the precise identification and quantification of critical vulnerable links; finally, we selected three typical damage cases—the ‘5·1’ case on the Meida Expressway in Guangdong, the ‘7·19’ case on the Danning Expressway in Shaanxi, and the ‘8·3’ case on the Yakang Expressway in Sichuan—for case validation, and proposed chain-breaking strategies. The research findings indicate that: (1) under specific hazard-forming environment, secondary disasters can supplant the primary causative factors to become the dominant driving nodes in chain evolution; (2) edge vulnerability and source-path diversity loss indicators respectively point to two distinct categories of high-risk edges; the comprehensive chain-breaking index compensates for the assessment blind spots of single indicators through two-dimensional weighting; (3) core vulnerabilities in disaster chains vary significantly across different regions: the Meida Expressway, the Danning Expressway, and the Yakang Expressway correspond to terminal response, pavement control node, and dual vulnerabilities at the source and structural levels, respectively, necessitating tailored chain-breaking strategies adapted to local conditions. These research findings offer a quantitative tool for infrastructure risk governance, contributing to the safety and sustainability of expressway transportation. Full article

Flood-season lake spatiotemporal dynamics are vital for ecological security and socioeconomic development, requiring consistent high-resolution monitoring. However, precipitation fluctuations and sediment turbidity significantly alter water quality, while blurred boundaries between water and floodplain wetlands challenge precise monitoring. To address these issues, this study proposes a water body extraction method leveraging polarimetric Synthetic Aperture Radar data. utilizes the maximum between-class variance algorithm for initial segmentation, optimizes the threshold via a genetic algorithm, and employs dynamic morphological operations to refine boundary details. The method was validated using 2015–2025 Sentinel-1 flood-season time series of Dongting Lake on Google Earth Engine. The results demonstrate that the proposed method achieves stable and accurate water extraction across various years and seasons, with an overall accuracy surpassing 0.93, confirming its robustness and broad applicability. Furthermore, the spatiotemporal hydrodynamics and driving mechanisms of Dongting Lake were analyzed by integrating the extracted water areas with multi-source data, including water level, precipitation, discharge, temperature, and sunshine duration. Findings indicate that the flood-season water area exhibited a fluctuating trend, initially increasing and subsequently decreasing, peaking at 2202.26 km² in 2020 and dropping to 614.04 km² in 2025, a pattern primarily driven by extreme meteorological events such as heavy rainfall and prolonged droughts. Spatially, inundation patterns were characterized by deeper water in the north and shallower depths in the south, separated by a topographically higher central region. Regression analysis revealed a robust correlation between water area and water level with an R² of 0.931, providing a quantitative reference for water level estimation in ungauged regions. Additionally, discharge and precipitation were positively correlated with water area, whereas temperature and sunshine duration exerted a negligible influence. This study supports flood regulation in the Dongting Lake basin and provides a robust framework for analyzing lake dynamics using long-term SAR data. Full article

Accurate short-term flood forecasting is vital for effective risk management and early warning systems. However, many data-driven models struggle to generalise with limited historical data and fail to consistently capture complex temporal dependencies across varying forecasting horizons. To address these challenges, this study proposes GHDFloodNet (Generalised Hybrid Data-limited Flood Prediction Network), a hybrid deep learning framework designed for robust multi-step-ahead forecasting. GHDFloodNet integrates First-Order Model-Agnostic Meta-Learning (FOMAML) with a Temporal Fusion Transformer (TFT) to enable rapid task adaptation and effectively capture long-range temporal dependencies and variable interactions. To further enhance predictive consistency, the framework incorporates a bidirectional Long Short-Term Memory (BiLSTM) network augmented with an additive attention mechanism and static feature fusion as a core learner within a meta-ensemble architecture. Bayesian hyperparameter optimisation within an AutoML framework identifies optimal model configurations, while a dedicated data handling layer with real-time augmentation improves stability under non-stationary conditions. The framework was evaluated for multi-horizon water level forecasting across four lead time ranges (1–6 h, 6–12 h, 12–24 h, and 24–48 h) using rainfall and lagged water level observations as primary inputs. Experimental results demonstrate that GHDFloodNet achieves robust, nearly invariant error distributions across the full 1–48 h forecast window, reporting an MSE of 0.53–0.55, RMSE of 0.72–0.74, and MAE of 0.35–0.36. Furthermore, the model exhibits stable goodness-of-fit, with R² and NSE values consistently ranging from 0.44 to 0.47 across all lead times, significantly outperforming conventional baselines, which typically exhibit pronounced error escalation at longer horizons. Overall, GHDFloodNet demonstrates that horizon-independent forecast reliability can be architecturally engineered, offering critical value for operational flood forecasting where consistent performance across all lead times outweighs peak short-range precision. Full article