Search Results (1,094)

Accurate cooling load forecasting in chiller units is critical for building energy optimization, yet remains challenging due to non-stationary nonlinear dynamics driven by coupled external weather variability (solar radiation, ambient temperature) and internal thermal loads. Conventional models fail to capture the spatiotemporal coupling inherent in load time series, violating their stationarity assumptions. To address this, this research proposes OptiNet, a spatiotemporal forecasting framework integrating patch-specific dynamic filtering with graph neural networks. OptiNet partitions multi-sensor data into non-overlapping time patches to develop a dynamic spatiotemporal graph. A learnable routing mechanism then performs adaptive dependency filtering to capture time-varying temporal–spatial correlations, followed by graph convolution for load prediction. Validated on long-term industrial logs (52,075 multi-sensor samples at 20 min; district cooling plant in Zhangjiang, Shanghai, with multiple chillers, towers, pumps, building meters, and a weather station), OptiNet achieves consistently lower MAE and MSE than Graph WaveNet across 6–144-step horizons and sampling frequencies of 20–60 min; among 30 set-tings it leads in 26, with MSE reductions up to 27.8% (60 min, 72-step) and typical long-horizon (72–144 steps) gains of ≈2–18% MSE and ≈1–15% MAE. Crucially, the model provides interpretable spatial-temporal dependencies (e.g., “Zone B solar radiation influences Unit 2 load with 4-h lag”), enabling data-driven chiller sequencing strategies that reduce electricity consumption by 12.7% in real-world deployments—directly advancing energy-efficient building operations. Full article

The accurate diagnosis of Hip Osteoarthritis (HOA) and the prediction of Total Hip Arthroplasty (THA) outcomes are crucial for reliable decision-making on treatment and rehabilitation strategies. Gait analysis (GA) is commonly employed for gait disorder examination in clinical settings, but it is still limited due to the massive data size and accuracy problems. A Machine Learning (ML) methodology has seen rapid growth in the past decade, but its development in the context of HOA and THA GA has not been previously examined. Thus, the novel contribution of this review is the evaluation of the current state of ML frameworks for the analysis of HOA and post-THA gaits. Five databases, namely PubMed, Embase, IEEE Xplore, ACM Digital Library, and Scopus, were searched in accordance with the PRISMA framework. Relevant publications published until May 2025 were retrieved, and information on reliability, applicability, and interpretability were extracted for quality assessment. Out of the 759 publications initially considered, 19 studies were selected, with 14 articles focused on classification and 5 articles on outcome prediction. Eight classification studies utilized kinematic features, while four outcome prediction articles utilized spatiotemporal parameters and mostly focused on post-THA gaits. The reported accuracy ranges between 70 and 100%, with the support vector machine (SVM) as the most frequently utilized ML algorithm. Scarce datasets, small sample sizes, and limited design description were the main hindrances revealed in our quality assessment. Nevertheless, this review demonstrated the recent developments in the utilization of ML techniques and evidently improved applicability through a consensus on the important gait features for HOA and post-THA gait analysis. Reliability and interpretability are still major concerns before ML models become widely accepted by medical practitioners. Future research should consider dataset quality, transparent validation protocol, model interpretability, and results’ explainability. Full article

Satellite-retrieved solar-induced chlorophyll fluorescence (SIF) provides an advanced proxy for global vegetation productivity. Recently, new high-quality remote sensing SIF datasets and reanalysis products have significantly advanced the application of SIF. However, the lack of long-term, daily resolution datasets continues to limit the precise exploration of vegetation dynamics, primarily due to challenges in daily modeling accuracy, substantial data volume, and computational demands. In this study, supported by the Google Earth Engine (GEE) platform, we developed a data-driven approach based on the Moving Spatial–Temporal Window Sampling (MSTWS) strategy for reconstructing long-term daily SIF. By learning the relationship between high-spatial-resolution Orbiting Carbon Observatory (OCO)-3 SIF and MODIS surface reflectance, we established a spatially and temporally specific daily prediction model for each day of the year (DOY), reconstructing the long-term daily OCO-3 SIF (DOSIF) from 2001 to the present with a global contiguous distribution. The prediction framework demonstrated robust performance with an R² of 0.92 on the training set and 0.81 on the validation set, indicating strong predictive ability and resistance to overfitting. Systematic evaluation of the dataset showed that DOSIF accurately captures the expected spatiotemporal distribution patterns. Cross-sensor validation with independent airborne SIF measurements further enhanced the reliability of the DOSIF dataset. Full article

Spaceborne Synthetic Aperture Radar (SAR) can be applied for monitoring tropical cyclones (TCs), but co-polarized C-band SAR suffers from signal saturation such that it is improper for high wind-speed conditions. In contrast, cross-polarized SAR data does not suffer from this issue, but the retrieval algorithm needs more deliberation. Previously, many geophysical model functions (GMFs) have been developed using cross-polarized data, which obtain wind speeds using the complex relationships described by radar backscatter, incidence angle, wind direction, and radar look direction. In this regard, the rapid development of artificial intelligence technology has provided versatile machine learning methods for such a nonlinear inversion problem. In this study, we comprehensively compare the wind-speed retrieval performance of several models including Back Propagation Neural Network (BPNN), Support Vector Machine (SVM), Random Forest (RF), and Deep Neural Network (DNN), which were developed based on spatio-temporal matching and correlation analysis of stepped frequency microwave radiometer (SFMR) and dual-polarized Sentinel-1 SAR data after noise removal. A data set with ~2800 samples is generated during TCs for training and validating the inversion model. The generalization ability of different models is tested by the reserved independent data. When using similar parameters with GMFs, RF inversion has the highest accuracy with a Root Mean Square Error (RMSE) of 3.40 m/s and correlation coefficient of 0.94. Furthermore, considering that the sea surface temperature is a crucial factor for generating TCs and influencing ocean backscattering, its effects on the proposed RF model are also explored, the results of which show improved wind-speed retrieval performances. Full article

To elaborate on the effects of hydraulic projects and physicochemical factors on the spatiotemporal distribution of phytoplankton communities, we monitored the phytoplankton communities and related water parameters in the Ganjiang River’s main channel over a five-year period. The survey revealed 65 species across six phyla, with Chlorophyta, Cyanophyta and Bacillariophyta as the most diverse groups. Phytoplankton abundance and biomass exhibited significant seasonal variations (p < 0.001), peaking in summer and autumn and reaching their lowest values in winter and spring. Spatially, phytoplankton abundance and biomass were not significantly different (p > 0.05), the abundance and biomass of Cyanophyta were higher in the two reservoir areas compared to the upstream sampling points. This suggests that the hydraulic projects altered the river’s flow and velocity, which led to a succession in phytoplankton community composition. Correlation analysis showed a strong positive association between the abundance and biomass of both Cyanophyta and Chlorophyta and water temperature (p < 0.001), but showed a significant negative relationship with nitrogen (p < 0.05). In contrast, Bacillariophyta abundance and biomass were positively and significantly correlated with ammonium nitrogen (p < 0.05). Redundancy analysis confirmed that water temperature and nitrogen are the primary environmental variables influencing the phytoplankton community’s succession. The direct alteration of river hydrodynamic characteristics by hydraulic projects, coupled with the reservoir-induced water stratification and its influence on vertical water temperature distribution, ultimately results in the profound reshaping of the phytoplankton community structure through coupled effects with nitrogen cycling. The findings from this study can scientifically inform the ecological scheduling, water quality management and water supply security of the Ganjiang River basin’s cascade reservoirs. Full article

Lipreading, or visual speech recognition, is the task of interpreting utterances solely from visual cues of lip movements. While early approaches relied on Hidden Markov Models (HMMs) and handcrafted spatiotemporal descriptors, recent advances in deep learning have enabled end-to-end recognition using large-scale datasets. However, such methods often require millions of labeled or pretraining samples and struggle to generalize under low-resource or speaker-independent conditions. In this work, we revisit lipreading from a multi-view learning perspective. We introduce MVIB-Lip, a framework that integrates two complementary representations of lip movements: (i) raw landmark trajectories modeled as multivariate time series, and (ii) recurrence plot (RP) images that encode structural dynamics in a texture form. A Transformer encoder processes the temporal sequences, while a ResNet-18 extracts features from RPs; the two views are fused via a product-of-experts posterior regularized by the multi-view information bottleneck. Experiments on the OuluVS and a self-collected dataset demonstrate that MVIB-Lip consistently outperforms handcrafted baselines and improves generalization to speaker-independent recognition. Our results suggest that recurrence plots, when coupled with deep multi-view learning, offer a principled and data-efficient path forward for robust visual speech recognition. Full article

Clustering of aircraft trajectories in terminal airspace is essential for procedure evaluation, flow monitoring, and anomaly detection, yet it is challenged by dense traffic, irregular sampling, and diverse maneuvering behaviors. This study proposes a unified framework that integrates dynamics-aware segmentation, Transformer–Variational Autoencoder (Transformer–VAE)-based representation learning, and density-aware clustering with joint optimization. A dynamic-feature Minimum Description Length (DFE-MDL) algorithm is introduced to preserve maneuver boundaries and reduce reconstruction errors, while the Transformer–VAE encoder captures nonlinear spatiotemporal dependencies and generates compact latent embeddings. Clusters are initialized using Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN) and further refined through Kullback–Leibler (KL) divergence minimization to improve consistency and separability. Experiments on large-scale ADS-B data from Guangzhou Baiyun International Airport, comprising over 27,000 trajectories, demonstrate that the framework outperforms conventional geometric and deep learning baselines. Results show higher reconstruction fidelity, clearer cluster separation, and reduced computation time, enabling interpretable flow structures that reflect operational practices. Overall, the framework provides a data-driven and scalable approach for terminal-area trajectory analysis, offering practical value for STAR/SID compliance monitoring, anomaly detection, and airspace management. Full article

This study provides the first integrated spatio-temporal assessment of water quality in Lakes Tonga and Oubeira, two adjacent Ramsar-designated wetlands within El Kala National Park (Algeria). The objective was to identify major pollution sources and inform targeted conservation strategies. Physico-chemical, microbiological, and heavy metal analyses were performed on water samples collected monthly over one year (September 2022–August 2023) from two sites per lake. Applying robust statistical analyses (ANOVA, Kruskal–Wallis, PCA, boxplots) and high-resolution spatial mapping, we revealed significant spatio-temporal heterogeneity and distinct pollution profiles between the two lakes. Specifically, Lake Tonga exhibited higher concentrations of organic and bacterial pollutants, likely linked to agricultural runoff and domestic discharge, while Lake Oubeira was characterized by elevated heavy metal concentrations and higher mineralization. The calculated Water Quality Index (WQI) classified the water quality of both lakes predominantly as “Moderate”, with punctual “Poor” quality episodes. Numerous parameters consistently exceeded water quality standards, indicating substantial ecological and health risks. Spatial distribution maps clearly pinpointed pollution hotspots, guiding lake-specific management measures. These findings underscore the urgent need for differentiated, targeted management interventions and an integrated, multidisciplinary approach for the effective conservation of these valuable wetland ecosystems. Full article

Despite growing evidence of the widespread impacts of human activities on carnivores and their prey, it remains unclear how different types and intensities of human disturbance reshape predator–prey interactions. In this study, we conducted a systematic camera-trapping survey on a threatened carnivore, the small Indian civet (Viverricula indica). This species forages on prey with contrasting diel patterns, including nocturnal rats and diurnal species such as Pallas’s squirrel (Callosciurus erythraeus) and Chinese bamboo partridge (Bambusicola thoracica) in the southern Wuyi Mountains of southeastern China. Based on data from an extensive sampling effort (60,901 trap days at 180 camera stations), we used kernel density estimation and Pianka’s index to examine whether and how different types and intensities of human activity (human presence, roads, and settlements), as well as diverse altitudes and different seasons, affect the spatiotemporal interactions between small Indian civets and their potential prey. We found that all studied species adjusted their activity patterns, either advancing or delaying their peaks, to achieve temporal segregation under varying types and intensities of human disturbance and different altitudes and seasons. At the temporal scale, interactions between small Indian civets and their potential prey supported the human shield hypothesis, suggesting that increased human disturbance provides diurnal prey with refuge from predation pressure. Conversely, at both spatial and spatiotemporal scales, higher levels of human disturbance increased the overlap between small Indian civets and their prey species. These findings highlight that human impacts on wildlife interactions are scale-dependent: temporal refuge for prey does not necessarily reduce spatial or spatiotemporal overlap, which may still increase encounter rates and predation risk. Because our sampling relied on ground-level cameras, our inferences are limited to terrestrial interactions; arboreal interactions remain unquantified and require combined ground–canopy sampling in future work. Effective conservation management thus requires considering these scale-dependent effects of human activities on wildlife interactions. Full article

The Qinling Mountains ecosystem serves as a vital ecological barrier and geographic demarcation line in China. Monitoring long-term land cover changes in the Qinling Mountains is essential for ecosystem pattern evaluation, environmental protection, and sustainable development. Focusing on the Qinling Mountains in Shaanxi Province, this study aimed to quantify the land cover changes from 1986 to 2020 using remote sensing and GIS technologies. An optimized Support Vector Machine (SVM) classification method was developed using Landsat satellite images and historical field samples. The method was employed to conduct land cover classification across eight discrete time periods: 1986, 1990, 1995, 2000, 2005, 2010, 2015, and 2020. The average overall accuracy (OA) of the classification results for the eight time periods was 96.42%, with a Kappa coefficient (K) of 0.9230, thus confirming the reliability of the mapping results. We subsequently developed a spatiotemporal Geo-information Tupu that facilitated a detailed analysis of land cover changes in the study area across different periods. The results show the following: (1) Forest was the dominant land cover type, followed by cropland. From 1986 to 2020, the forest, impervious surface, and water body areas showed overall increasing trends, although fluctuations were observed over time, and the increase was estimated at 6677.30 km², 557.57 km², and 135.71 km², respectively. In contrast, the areas of cropland, grassland, and bare soil showed a fluctuating decreasing trend, with a decrease in areal coverage of 2790.57 km², 1528.76 km², and 3042.66 km², respectively. During the study period, the forest area experienced the greatest increase but maintained the lowest dynamic degree. In contrast, bare soil showed the largest decrease and the highest dynamic degree. (2) A total of 30.74% of the area underwent dynamic changes during the study period, with the most active transformation occurring after 2010; these changes were mainly manifested in the outflow of cropland (4997.27 km²), the transfer of forest (8557.43 km²), and the expansion of impervious surfaces (771.33 km²). In conclusion, the overall ecological environment is improving. The results demonstrate a land cover reconstruction process that enables the management department to rationally utilize natural resources in the Qinling Mountains. Full article

To overcome the limited accuracy of existing street vitality assessments under dense occlusion and their lack of dynamic, multi-source data fusion, this study proposes an integrated dynamic model that couples an enhanced YOLOv11 with heterogeneous spatiotemporal datasets. The network introduces a two-backbone architecture for stronger multi-scale fusion, Spatial Pyramid Depth Convolution (SPDConv) for richer urban scene features, and Dynamic Sparse Sampling (DySample) for robust occlusion handling. Validated in Yibin, the model achieves 90.4% precision, 67.3% recall, and 77.2% mAP@50 gains of 6.5%, 5.3%, and 5.1% over the baseline. By fusing Baidu heatmaps, street-view imagery, road networks, and POI data, a spatial coupling framework quantifies the interplay between commercial facilities and street vitality, enabling dynamic assessment of urban dynamics based on multi-source data fusion, offering insights for targeted retail regulation and adaptive traffic management. By enabling continuous monitoring of urban space use, the model enhances the allocation of public resources and cuts energy waste from idle traffic, thereby advancing urban sustainability via improved commercial planning and responsive traffic control. The work provides a methodological foundation for shifting urban resource allocation from static planning to dynamic, responsive systems. Full article

Accurate wildfire prediction and identification of key environmental drivers are critical for effective wildfire management. We propose a spatiotemporally adaptive framework integrating ST-DBSCAN clustering with GTWR-RFR. In this hybrid model, Random Forest captures local nonlinear relationships, while GTWR assigns adaptive spatiotemporal weights to refine predictions. Using historical wildfire records from Hunan Province, China, we first derived wildfire occurrence probabilities via ST-DBSCAN, avoiding the need for artificial non-fire samples. We then benchmarked GTWR-RFR against seven models, finding that our approach achieved the highest accuracy (R² = 0.969; RMSE = 0.1743). The framework effectively captures spatiotemporal heterogeneity and quantifies dynamic impacts of environmental drivers. Key contributing drivers include DEM, GDP, population density, and distance to roads and water bodies. Risk maps reveal that central and southern Hunan are at high risk during winter and early spring. Our approach enhances both predictive performance and interpretability, offering a replicable methodology for data-driven wildfire risk assessment. Full article

Rolling bearings constitute critical rotating components within rolling mill equipment. Production efficiency and the operational safety of the whole mechanical system are directly governed by their operational health state. To address the dual challenges of the over-reliance of conventional diagnostic methods on expert experience and the scarcity of fault samples in industrial scenarios, we propose a virtual–physical data fusion-optimized intelligent fault diagnosis framework. Initially, a dynamics-based digital twin model for rolling bearings is developed by leveraging their geometric symmetry. It is capable of generating comprehensive fault datasets through parametric adjustments of bearing dimensions and operational environments in virtual space. Subsequently, a symmetry-informed architecture is constructed, which integrates multi-scale convolutional neural networks with attention mechanisms and bidirectional gated recurrent units (MCNN-AT-BiGRU). This architecture enables spatiotemporal feature extraction and enhances critical fault characteristics. The experimental results demonstrate 99.5% fault identification accuracy under single operating conditions. It maintains stable performance under low SNR conditions. Furthermore, the framework exhibits superior generalization capability and transferability across the different bearing types. Full article

Soil organic carbon (SOC) plays a critical role in the global carbon cycle and serves as a sensitive indicator of climate change impacts, with its dynamics significantly influencing regional ecological security and sustainable development. This study focuses on the Songnen Plain in Northeast China—a key black soil agricultural region increasingly affected by water erosion, primarily through surface runoff and rill formation on gently sloping cultivated land. We aim to investigate the spatiotemporal dynamics of SOC stocks across different cultivated land erosion zones under projected future climate change scenarios. To quantify current and future SOC stocks, we applied a boosted regression tree (BRT) model based on 130 topsoil samples (0–30 cm) and eight environmental variables representing topographic and climatic conditions. The model demonstrated strong predictive performance through 10-fold cross-validation, yielding high R² and Lin’s concordance correlation coefficient (LCCC) values, as well as low mean absolute error (MAE) and root mean square error (RMSE). Key drivers of SOC stock spatial variation were identified as mean annual temperature, elevation, and slope aspect. Using a space-for-time substitution approach, we projected SOC stocks under the SSP245 and SSP585 climate scenarios for the 2050s and 2090s. Results indicate a decline of 177.66 Tg C (SSP245) and 186.44 Tg C (SSP585) by the 2050s relative to 2023 levels. By the 2090s, SOC losses under SSP245 and SSP585 are projected to reach 2.84% and 1.41%, respectively, highlighting divergent carbon dynamics under varying emission pathways. Spatially, SOC stocks were predominantly located in areas of slight (67%) and light (22%) water erosion, underscoring the linkage between erosion intensity and carbon distribution. This study underscores the importance of incorporating both climate and anthropogenic influences in SOC assessments. The resulting high-resolution SOC distribution map provides a scientific basis for targeted ecological restoration, black soil conservation, and sustainable land management in the Songnen Plain, thereby supporting regional climate resilience and China’s “dual carbon” goals. These insights also contribute to global efforts in enhancing soil carbon sequestration and achieving carbon neutrality goals. Full article

Benthic macroinvertebrates are widely used for river ecosystem health monitoring, yet challenges remain in non-wadable rivers, particularly regarding sampling effort. We evaluated hand-net sampling efficiency at three sites along the Danjiang River (a Yangtze River tributary) by analyzing taxa richness across taxonomic levels under varying replicate numbers. In total, 61 taxa (41 families) of benthic macroinvertebrates were identified. Non-metric multidimensional scaling analysis indicated no significant spatiotemporal variation in community composition. However, sampling effort increased, and the benthic macroinvertebrate taxa richness at both genus/species and family levels also increased. At eight sample replicates, the taxa accumulation curve at the genus/species level did not show an asymptote, with the observed richness reaching 67–80% of the predicted values calculated by Jackknife 1. In contrast, the family-level curve exhibited a clear asymptotic trend, with the observed richness reaching 82–100% of the predicted values. As sampling effort increased, bias decreased and accuracy improved, particularly for family-level taxa. Additionally, the BMWP scores also increased with the sampling effort. When the replicate number was no less than six, the BMWP reached stable assessment grades for all cases. From the perspective of bioassessment in non-wadable rivers, the hand net is suitable for collecting benthic macroinvertebrates. However, there is a risk of underestimating taxa richness due to insufficient sampling effort. Using family-level taxa can partially mitigate the impacts caused by insufficient sampling efforts to a certain extent, but further validation is needed for other non-wadable rivers (e.g., those with soft substrates). In conclusion, our research results indicate that six replicate hand-net samplings in non-wadable hard-bottom rivers can be regarded as a cost-effective and reliable sampling method for benthic macroinvertebrate BMWP assessment. This strategy provides a relatively practical reference for the monitoring of benthic macroinvertebrate in the same type of rivers in China. Full article