Search Results (474)

Hydrological modeling is critical for assessing water availability and guiding sustainable resource management, particularly in monsoon-dependent, data-scarce basins such as the Chalakudy River Basin (CRB) in Kerala, India. This study integrated the Soil Conservation Service Curve Number (SCS-CN) method within the Google Earth Engine (GEE) platform, making novel use of multi-source, open access datasets (CHIRPS precipitation, MODIS land cover and evapotranspiration, and OpenLand soil data) to estimate spatially distributed long-term runoff (2001–2023). Model calibration against observed runoff showed strong performance (NSE = 0.86, KGE = 0.81, R² = 0.83, RMSE = 29.37 mm and ME = 13.48 mm), validating the approach. Over 75% of annual runoff occurs during the southwest monsoon (June–September), with July alone contributing 220.7 mm. Seasonal assessments highlighted monsoonal excesses and dry-season deficits, while water balance correlated strongly with rainfall (r = 0.93) and runoff (r = 0.94) but negatively with evapotranspiration (r = –0.87). Time-series analysis indicated a slight rise in rainfall, a decline in evapotranspiration, and a marginal improvement in water balance, implying gradual enhancement of regional water availability. Spatial analysis revealed a west–east gradient in precipitation, evapotranspiration, and water balance, producing surpluses in lowlands and deficits in highlands. These findings underscore the potential of cloud-based hydrological modeling to capture spatiotemporal dynamics of hydrological variables and support climate-resilient water management in monsoon-driven and data-scarce river basins. Full article

Spiking neural networks (SNNs) have emerged as a promising model for energy-efficient, event-driven processing of asynchronous event streams from Dynamic Vision Sensors (DVSs), a class of neuromorphic image sensors with microsecond-level latency and high dynamic range. Nevertheless, challenges persist in optimising training and effectively handling spatio-temporal complexity, which limits their potential for real-time applications on embedded sensing systems such as object tracking and recognition. Targeting this neuromorphic sensing pipeline, this paper proposes the Dynamic Tracking with Event Attention Spiking Network (DTEASN), a novel framework designed to address these challenges by employing a pure SNN architecture, bypassing conventional convolutional neural network (CNN) operations, and reducing GPU resource dependency, while tailoring the processing to DVS signal characteristics (asynchrony, sparsity, and polarity). The model incorporates two innovative, self-developed components: an event-driven multi-scale attention mechanism and a spatio-temporal event convolver, both of which significantly enhance spatio-temporal feature extraction from raw DVS events. An Event-Weighted Spiking Loss (EW-SLoss) is introduced to optimise the learning process by prioritising informative events and improving robustness to sensor noise. Additionally, a lightweight event tracking mechanism and a custom synaptic connection rule are proposed to further improve model efficiency for low-power, edge deployment. The efficacy of DTEASN is demonstrated through empirical results on event-based (DVS) object recognition and tracking benchmarks, where it outperforms conventional methods in accuracy, latency, event throughput (events/s) and spike rate (spikes/s), memory footprint, spike-efficiency (energy proxy), and overall computational efficiency under typical DVS settings. By virtue of its event-aligned, sparse computation, the framework is amenable to highly parallel neuromorphic hardware, supporting on- or near-sensor inference for embedded applications. Full article

The early developmental transition from endogenous to exogenous feeding is a critical period in carnivorous fish larvae, often associated with high mortality rates in aquaculture. Although trypsin, a key protease in protein digestion, is hypothesized to play a pivotal role in initiating exogenous feeding, the expression dynamics and functional contributions of trypsin and isoforms during early development remain poorly characterized in carnivorous species. This study explores the critical role of trypsin in the early feeding process of carnivorous fish, using mandarin fish (Siniperca chuatsi) as a model, which is a commercially valuable species that faces significant challenges during this phase due to its strict dependence on live prey and underdeveloped digestive system. Phylogenetic analysis indicates that, compared to herbivorous and omnivorous fish, carnivorous fish have evolved a greater number of trypsins, with a distinct branch specifically dedicated to try. RNA-seq data revealed the expression profiles of 13 trypsins during the early developmental stages of the mandarin fish. Most trypsins began to be expressed in large quantities with the appearance of the pancreas, reaching a peak prior to feeding. In situ hybridization revealed the spatiotemporal expression pattern of trypsins, starting from the pancreas in early development and later extending to the intestines. Furthermore, inhibition of trypsins activity successfully suppressed early oral feeding in mandarin fish, which was achieved by increasing the expression of cholecystokinin 2 (CCK2) and proopiomelanocortin (POMC) to suppress appetite. These findings enhance our understanding of the adaptive relationship between the ontogeny of the digestive enzyme system and feeding behavior in carnivorous fish. This research may help alleviate bottleneck issues in aquaculture production by improving the survival rate and growth performance of carnivorous fish during critical early life stages. Full article

The increasing adoption of electric delivery fleets introduces significant challenges related to uneven energy utilization and suboptimal scheduling efficiency. Vehicle-to-Vehicle (V2V) energy sharing presents a promising solution, but its effectiveness critically depends on precise matching and co-optimization within dynamic urban traffic environments. This paper proposes a hierarchical optimization framework to minimize total fleet operational costs, incorporating a comprehensive analysis that includes battery degradation. The core innovation of the framework lies in coupling high-level path planning with low-level real-time speed control. First, a high-fidelity energy consumption surrogate model is constructed through model predictive control simulations, incorporating vehicle dynamics and signal phase and timing information. Second, the spatiotemporal longest common subsequence algorithm is employed to match the spatio-temporal trajectories of energy-provider and energy-consumer vehicles. A battery aging model is integrated to quantify the long-term costs associated with different operational strategies. Finally, a multi-objective particle swarm optimization algorithm, integrated with MPC, co-optimizes the rendezvous paths and speed profiles. In a case study based on a logistics network, simulation results demonstrate that, compared to the conventional station-based charging mode, the proposed V2V framework reduces total fleet operational costs by a net 12.5% and total energy consumption by 17.4% while increasing the energy utilization efficiency of EV-Ps by 21.4%. This net saving is achieved even though the V2V strategy incurs a marginal increase in battery aging costs, which is overwhelmingly offset by substantial savings in logistical efficiency. This study provides an efficient and economical solution for the dynamic energy management of electric fleets under realistic traffic conditions, contributing to a more sustainable and resilient urban logistics ecosystem. Full article

Modern fire alarm systems are essential for public safety, yet they often fail to exploit the wealth of historical alarm data and the complex spatiotemporal dependencies inherent in urban environments. Graph Neural Networks (GNNs) are currently among the most popular methods for handling complex spatiotemporal dependencies. While a range of dynamic GNN approaches have been proposed, many existing GNN-based predictors still rely on a static topology, which limits their ability to fully capture the evolving nature of risk propagation. Furthermore, even among dynamic graph methods, most focus on temporal link prediction or social interaction modeling, with limited exploration in safety-critical applications such as fire alarm prediction. DeaGAT dynamically updates inter-building edge weights through an attention mechanism, enabling the graph structure to evolve in response to shifting risk patterns. A margin-based contrastive learning objective further enhances the quality of node embeddings by distinguishing subtle differences in risk states. In addition, DeaGAT jointly models static building attributes and dynamic alarm sequences, effectively integrating long-term semantic context with short-term temporal dynamics. Extensive experiments on real-world datasets, including comparisons with state-of-the-art baselines and comprehensive ablation studies, demonstrate that DeaGAT achieves superior accuracy and F1-score, validating the effectiveness of dynamic graph updating and contrastive learning in enhancing proactive fire early-warning capabilities. Full article

With the rapid development of intelligent connected vehicle technology, false data injection (FDI) attacks have become a major challenge in the Internet of Vehicles (IoV). While deep learning methods can effectively identify such attacks, the dynamic, distributed architecture of the IoV and limited computing resources hinder both privacy protection and lightweight computation. To address this, we propose FedIFD, a federated learning (FL)-based detection method for false data injection attacks. The lightweight threat detection model utilizes basic safety messages (BSM) for local incremental training, and the Q-FedCG algorithm compresses gradients for global aggregation. Original features are reshaped using a time window. To ensure temporal and spatial consistency, a sliding average strategy aligns samples before spatial feature extraction. A dual-branch architecture enables parallel extraction of spatiotemporal features: a three-layer stacked Bidirectional Long Short-Term Memory (BiLSTM) captures temporal dependencies, and a lightweight Transformer models spatial relationships. A dynamic feature fusion weight matrix calculates attention scores for adaptive feature weighting. Finally, a differentiated pooling strategy is applied to emphasize critical features. Experiments on the VeReMi dataset show that the accuracy reaches 97.8%. Full article

Both water and sediment resource allocation are critical for achieving sustainable development in sediment-laden river basins. However, current understanding lacks a holistic perspective and fails to capture the inseparability of water and sediment. The Yellow River Basin (YRB) is the world’s most sediment-laden river, characterized by pronounced ecological fragility and uneven socio-economic development. This study introduces integrated water-sediment allocation frameworks for the YRB based on the perspective of the water-sediment nexus, aiming to regulate their impacts on socio-economic and ecological systems. The frameworks were established for both artificial units (e.g., irrigation zones and reservoirs) and geological units (e.g., the Jiziwan region, lower channels, and estuarine deltas) within the YRB. The common feature of the joint allocation of water and sediment across the five units lies in shaping a coordinated water–sediment relationship, though their focuses differ, including in-stream water-sediment processes and combinations, the utilization of water and sediment resources, and the constraints imposed by socio-economic and ecological systems on water-sediment distribution. In irrigation zones, the primary challenge lies in engineering-based control of inflow magnitude and spatiotemporal distribution for both water and sediment. In reservoir systems, effective management requires dynamic regulation through density current flushing and coordinated operations to achieve water-sediment balance. In the Jiziwan region, reconciling socio-economic development with ecological integrity requires establishing science-based thresholds for water and sediment use while ensuring a balance between utilization and protection. Along the lower channel, sustainable management depends on delineating zones for human activities and ecological preservation within floodplains. For deltaic systems, key strategies involve adjusting upstream sediment and refining depositional processes. Full article

Anomaly detection in multivariate time series (MTS) remains challenging due to the presence of complex and dynamic spatiotemporal dependencies. To address this, we propose the Dynamic Spatiotemporal Graph Mixed Network (DTS-MixNet), which takes a sliding window data as input to predict the next time series data and determine its state. The model comprises five blocks. The Temporal Graph Structure Learner (TGSL) generates the attention-weighted graphs via two types of neighbor relationships and the multi-head-attention-based neighbor degrees. Then, the Cross-Temporal Dynamic Encoder (CTDE) aggregates the cross-temporal dependencies from attention-weighted graphs, and encodes them into a proxy multivariate sequence (PMS), which is fed into the proposed Cross-Variable Dynamic Encoder (CVDE). Subsequently, the CVDE captures the sensors-among spatial relationship through multiple local spatial graphs and a global spatial graph, and produces a spatial graph sequence (SGS). Finally, the Spatiotemporal Mixer (TSM) mixes PMS and SGS to build a spatiotemporal mixed sequence (TSMS) for downstream tasks, e.g., classification or prediction. We evaluate on two industrial control datasets and discuss applicability to non-industrial multivariate time series. The experimental results on benchmark datasets show that the proposed DTS-MixNet is encouraging. Full article

Dynamic path generation in complex transportation networks is essential for intelligent transportation systems. Traditional methods, such as shortest path algorithms or heuristic-based models, often fail to capture real-world travel behaviors due to their reliance on simplified assumptions and limited ability to handle long-range dependencies or non-linear patterns. To address these limitations, we propose PathGen-LLM, a large language model (LLM) designed to learn spatial–temporal patterns from historical paths without requiring handcrafted features or graph-specific architectures. Exploiting the structural similarity between path sequences and natural language, PathGen-LLM converts spatiotemporal trajectories into text-formatted token sequences by encoding node IDs and timestamps. This enables the model to learn global dependencies and semantic relationships through self-supervised pretraining. The model integrates a hierarchical Transformer architecture with dynamic constraint decoding, which synchronizes spatial node transitions with temporal timestamps to ensure physically valid paths in large-scale road networks. Experimental results on real-world urban datasets demonstrate that PathGen-LLM outperforms baseline methods, particularly in long-distance path generation. By bridging sequence modeling and complex network analysis, PathGen-LLM offers a novel framework for intelligent transportation systems, highlighting the potential of LLMs to address challenges in large-scale, real-time network tasks. Full article

A quantitative study on the spatial structure and spatiotemporal variation characteristics of net carbon sinks in regional farmland ecosystems is of significant importance for uncovering the multifunctional roles of farmland ecosystems and formulating region-specific agricultural policies and management strategies. Based on the measurement of net carbon sinks in county-level farmland ecosystems across Hunan Province from 2005 to 2020, this research employs methodologies, including the standard deviational ellipse (SDE), spatial autocorrelation, and exploratory spatiotemporal data analysis (ESTDA) to investigate the spatiotemporal evolution characteristics of net carbon sinks in Hunan’s county-level farmland ecosystems. The results show that the net carbon sinks of county-level farmland ecosystems in Hunan Province exhibits a “northeast–southwest” spatial distribution pattern, with a trend toward spatial agglomeration during contraction, and the center of gravity of net carbon sinks has generally shifted northwestward over time. A significant positive spatial correlation exists globally in the net carbon sinks of Hunan’s county-level farmland ecosystems, and the degree of spatial agglomeration has gradually intensified amid fluctuations. The dynamic evolution of local spatial patterns of net carbon sinks in county-level farmland ecosystems in Hunan Province varied significantly, showing strong stability in both local spatial structure and spatial dependence direction. In contrast, eastern and central Hunan exhibited more dynamic local spatial structures compared to southern and northern regions. The local spatial association patterns of the net carbon sinks in county-level farmland ecosystems remained relatively stable, with weak spatial synergy and a pronounced path-dependent locking effect in spatial agglomeration. Full article

Background/Objectives: 1,8-Cineole, an epoxy monoterpene, is a key volatile component of Sugmel-3, a traditional Mongolian medicine used for treating insomnia. Although previous studies suggest that 1,8-Cineole can cross the blood–brain barrier (BBB), its precise spatiotemporal distribution in the brain and its in situ association with alterations in neurotransmitter (NT) levels remain unclear. This study utilized ambient mass spectrometry imaging (AFADESI-MSI) to investigate the dynamic brain distribution of 1,8-Cineole and its major metabolite, as well as their correlation with NT levels. Methods: Sprague Dawley rats (n = 3 per time point) received oral administration of 1,8-Cineole (65 mg/kg). Brain tissues were harvested 5 min, 30 min, 3 h, and 6 h post dose and analyzed using AFADESI-MSI. The spatial and temporal distributions of 1,8-Cineole, its metabolite 2-hydroxy-1,8-Cineole, key neurotransmitters (e.g., 5-HT, GABA, glutamine, melatonin), and related endogenous metabolites were mapped across 13 functionally distinct brain microregions. Results: AFADESI-MSI demonstrated rapid brain entry of 1,8-Cineole and its metabolite, with distinct spatiotemporal pharmacokinetics. The metabolite exhibited higher brain exposure, with 1,8-Cineole predominant in the cortex (CTX) and hippocampus (HP), while its metabolite showed pronounced accumulation in the pineal gland (PG), alongside CTX/HP. Region-dependent alterations in neurotransmitter levels (notably in PG, HP) correlated with drug concentrations, with observed increases in key molecules of the serotonergic and GABAergic pathways. Conclusions: Using AFADESI-MSI, this study provides the first spatiotemporal map of 1,8-Cineole and its metabolite in the brain. The correlation between their region-specific distribution and local neurotransmitter alterations suggests a direct mechanistic link to Sugmel-3′s sedative–hypnotic efficacy, guiding future target identification. Full article

Spectrum prediction is essential for cognitive radio, enabling dynamic management and enhanced utilization, particularly in multi-band environments. Yet, its complex spatiotemporal nature and non-stationarity pose significant challenges for achieving high accuracy. Motivated by this, we propose a multi-scale Mamba-based multi-band spectrum prediction method. The core Mamba module combines Bidirectional Selective State Space Models (SSMs) for long-range dependencies and dynamic convolution for local features, efficiently extracting spatiotemporal characteristics. A multi-scale pyramid and adaptive prediction head select appropriate feature levels per prediction step, avoiding full-sequence processing to ensure accuracy while reducing computational cost. Experiments on real-world datasets across multiple frequency bands demonstrate effective handling of spectrum non-stationarity. Compared to baseline models, the method reduces root mean square error (RMSE) by 14.9% (indoor) and 7.9% (outdoor) while cutting GPU memory by 17%. Full article

As a low-carbon and clean energy source, Coalbed methane (CBM) is of great significance in reducing greenhouse gas emissions, optimizing the energy structure, safeguarding mine safety, and promoting the transformation to a green economy to achieve sustainable development. Coalbed methane (CBM) in Xinjiang’s steeply dipping coal seams is abundant but difficult to predict due to complex geology and distinct gas flow behaviors, making traditional methods ineffective. This study proposes GCN-BiGRU, a parallel dual-module model integrating seepage mechanics, reservoir engineering, geological structures, and production history. The GCN module models wells as nodes, using geological attributes and spatial distances to capture inter-well interference; the BiGRU module extracts temporal dependencies from production sequences. An adaptive fusion mechanism dynamically combines spatiotemporal features for robust prediction. Validated on Baiyanghe block data, the model achieved MAE 59.04, RMSE 94.25, and improved accuracy from 64.47% to 92.8% as training wells increased from 20 to 84. It also showed strong transferability to independent sub-regions, enabling real-time prediction and scenario analysis for CBM development and reservoir management. Full article

This study, grounded in traveling wave theory, develops a cross-scale reaction-diffusion model to describe nutrient-dependent bacterial growth on agar surfaces and applies it to in silico investigations of microbial population dynamics. The approach is based on the coupling of a modified Allen–Cahn equation with the formulation of quorum sensing signal dynamics, incorporating a nutrient-dependent regulatory threshold and stochastic diffusion. A closed-loop model of bacterial growth regulated by quorum sensing is developed through theoretical analysis, numerical simulations, and computational experiments.The model is implemented using Yanenko’s computational scheme, which incorporates corrective refinement via Heun’s method to account for nonlinear components. Numerical simulations are carried out in MATLAB, allowing for accurate computation of spatio-temporal patterns and facilitating the identification of key mechanisms governing the collective behavior of bacterial communities. Full article

Modern logistics operations require real-time adaptive solutions for three-dimensional bin packing that maintain spatial symmetry and load balance. This paper introduces a time-series-based online 3D packing problem with dual unknown sequences, where containers and items arrive dynamically. The challenge lies in achieving symmetric distribution for stability and optimal space utilization. We propose the Second-Order Dual Pointer Adversarial Network (So-DPAN), a deep reinforcement learning architecture that leverages symmetry principles to decompose spatiotemporal optimization into sequence matching and spatial arrangement sub-problems. The dual pointer mechanism enables efficient item-container pairing, while the second-order structure captures temporal dependencies by maintaining symmetric packing patterns. Our approach considers geometric symmetry for spatial arrangement and temporal symmetry for sequence matching. The Actor-Critic framework uses symmetry-based reward functions to guide learning toward balanced configurations. Experiments demonstrate that So-DPAN outperforms DQN, DDPG, and traditional heuristics in solution quality and efficiency while maintaining superior symmetry metrics in center-of-gravity positioning and load distribution. The algorithm exploits inherent symmetries in packing structure, advancing theoretical understanding through symmetry-aware optimization while providing a deployable framework for Industry 4.0 smart logistics. Full article