Search Results (920)

Efficient management of soil–water resources is critical for global food security under intensifying climatic and demographic pressures. This review provides a comprehensive synthesis of artificial intelligence (AI) and distributed deep learning methodologies applied to soil–water interactions in precision agriculture. The physical and hydraulic foundations of soil–water systems—including water retention, unsaturated flow governed by the Richards equation, and soil degradation processes—are examined and situated within a unified framework of AI-based modeling and decision support. Classical machine learning (ML) algorithms (Random Forests, Support Vector Machines, gradient boosting) and deep learning architectures (convolutional neural networks, long short-term memory networks, transformers) are evaluated with respect to their capacity to predict soil moisture dynamics, estimate hydraulic properties, support smart irrigation scheduling, and generate digital soil maps at field-to-regional scales. Distributed training paradigms, federated learning for privacy-preserving multi-farm analytics, and edge AI deployment on low-power IoT hardware are assessed as enabling infrastructures for scalable agricultural intelligence. This review further addresses explainability, uncertainty quantification, and ethical dimensions inherent to AI-driven agricultural systems. Key challenges—including training data scarcity in data-poor regions, model interpretability, integration with physics-based hydrological models, and real-time deployment constraints—are critically discussed. Prospective research directions encompass physics-informed neural networks, foundation models for earth observation, autonomous digital twins of soil–water systems, and federated learning architectures aligned with data sovereignty frameworks. The synthesis underscores AI’s transformative potential for sustainable agricultural water management while delineating the technical and sociotechnical barriers that must be resolved to realize this potential at a global scale. Full article

Global agriculture faces a dual imperative: increase food production to meet rising demand while simultaneously reducing environmental impacts and resource inefficiencies. Addressing this challenge requires repositioning ruminant nutrition as the intelligent nexus linking crop and livestock production within Integrated Crop–Livestock Systems (ICLS). In this role, nutrition becomes central to restoring ecological, nutritional, and economic synergies that have been fragmented by decades of agricultural specialization. While ICLS provides the ecological foundation, Precision Livestock Farming delivers the technological and analytical infrastructure necessary to operationalize integration at the individual-animal level. Real-time sensing, Internet of Things platforms, and Artificial Intelligence (AI) enable dynamic monitoring of animal physiology, behavior, and environmental interactions across scales. A key advancement in this evolution is the development of Hybrid Intelligent Mechanistic Models (HIMM), which integrate biologically grounded mechanistic models with data-driven AI approaches. By combining interpretability with adaptive learning, HIMM enhances predictive accuracy, extrapolative capacity, and decision transparency, enabling the creation of digital twins that simulate biological responses before management interventions are implemented. Such architectures extend precision nutrition beyond feed efficiency and methane mitigation to include nutrient density and product quality, thereby linking different ecosystem processes directly to human dietary needs. Integrating nutrition with advanced modeling and monitoring tools can help livestock systems move beyond static “net-zero” benchmarks toward sustainable strategies that are responsive to local production contexts. In this reframed paradigm, nutrition is not merely a production input but the central analytical framework that computationally links biological mechanisms, environmental stewardship, technological innovation, and human health within sustainable ruminant systems. Full article

As extreme weather events increasingly threaten global food systems, accurately assessing climate risks and predicting regional crop yields remains a critical challenge. Conventional prediction models often rely on direct weather-to-yield relationships, bypassing continuous crop physiological responses and limiting their capacity to capture fine-grained temporal impacts of meteorological anomalies. To address this, we propose a novel two-stage spatiotemporal functional framework that integrates high-resolution daily weather trajectories with satellite-derived indicators, utilizing the Enhanced Vegetation Index (EVI) and Land Surface Water Index (LSWI) to represent canopy structural vigor and hydraulic status, respectively. In the first stage, a Historical Functional Linear Model (HFLM) dynamically maps daily meteorological trajectories (temperature, precipitation, and solar radiation) onto continuous physiological curves under strict temporal causality constraints. This generates bivariate coefficient surfaces that reveal dynamic windows of vulnerability and capture divergent, lagged physiological responses to climate stress. In the second stage, a spatially heterogeneous functional additive model integrates these weather-shaped physiological trajectories alongside raw meteorological dynamics as joint predictors for county-level yields. By extracting functional principal components and modeling flexible non-linear biological responses while accounting for continuous spatial heterogeneity, this dual-channel frameworkcaptures key aspects of both chronic physiological stress and acute meteorological shocks. Validated across a 25-year (2000–2024) U.S. Corn Belt panel, the proposed DC-FAM achieves a mean weighted mean squared prediction error (WMSPE) of 242.33 (bu/acre)² and a median out-of-sample $R_{\mathrm{cv}}^2$ of 0.422, outperforming all benchmarks including a random forest. Attribution of the 2012 flash drought further demonstrates the framework’s capacity to mechanistically trace the complete disaster propagation chain from anomalous spring warming to mid-summer hydraulic failure. The proposed framework provides a transparent, biophysically grounded tool for decoding dynamic climate stress trajectories and disaster propagation chains, offering potential implications for adaptive farm management and precision agricultural insurance. Full article

Accurate pig counting during herd transfers is fundamental to effective livestock management in large-scale swine production, yet existing methods struggle with bidirectional passages, boundary oscillations, and occlusion in real corridor environments. This study proposes an integrated system combining an improved YOLO-based detection model with a Hysteresis-based Multi-frame Temporal Confirmation Counting Strategy (HMTC). The YOLO11s baseline was enhanced using lightweight RepViT blocks, dynamic upsampling (DySample), and shape-aware bounding box regression (Shape-IoU). The resulting model achieves a mAP₅₀ of 0.982 with a compact architecture of 8.28M parameters, representing a 12.3% reduction relative to the baseline while improving detection accuracy. To address bidirectional counting challenges, the HMTC strategy utilizes hysteresis-based region classification, temporal confirmation, and trajectory verification to suppress boundary jitter and ensure directional correctness. Evaluated on nine videos from a single transfer corridor, the proposed system achieves an overall counting accuracy of 99.21% on this test set and runs in real time on an embedded edge device at over 30 FPS without loss of counting accuracy. Together, the improved detection model and HMTC counting strategy provide a cohesive approach to pig passage counting, validated here under a single transfer-corridor condition; these results offer a promising basis for automated animal inventory management, pending further validation across more diverse farm environments. Full article

Listeria monocytogenes is a ubiquitous Gram-positive bacterium responsible for listeriosis, a foodborne zoonotic disease affecting humans and animals. Although infection in immunocompetent individuals is often asymptomatic or limited to mild self-limiting gastroenteritis, Listeria monocytogenes may cause severe invasive disease in vulnerable groups, including pregnant women, neonates, elderly individuals, and immunocompromised patients. Although the incidence of listeriosis is relatively low compared with many other foodborne pathogens, the high hospitalization and mortality rates associated with clinical cases make this bacterium a major concern for food safety and public health. The evolutionary success of L. monocytogenes reflects the interaction between a conserved core genome and a dynamic accessory genome shaped by horizontal gene transfer (HGT), ecological selection, and expansion of specific clones. Transient intestinal carriage in humans and animals, potentially influenced by gut microbiome composition, creates ecological interfaces where plasmids, transposons, prophages, and integrative conjugative elements contribute to the exchange of antimicrobial resistance determinants, virulence factors, and stress tolerance systems. Virulence diversification is further influenced by the differential distribution of pathogenicity islands such as LIPI-1, LIPI-3, and LIPI-4 across specific clonal lineages. These evolutionary processes occur across interconnected farm, food-production, environmental, and clinical ecosystems consistent with the One Health framework. Advances in whole-genome sequencing have clarified lineage-specific gene flow, expansion of specific clones, and the dynamics of the resistome and mobilome in L. monocytogenes populations. This narrative review aims to synthesize current knowledge on the mobile genetic elements and ecological interfaces that shape horizontal gene transfer in L. monocytogenes. Its novelty lies in integrating antimicrobial resistance, virulence-associated genomic islands, stress adaptation, and gut microbiome-mediated selection within a One Health and metapopulation framework. The main message of this review is that HGT should be interpreted as a context-dependent contributor to L. monocytogenes adaptation, acting together with clonal background, ecological selection, and mobile genetic elements. Full article

One of the keys to improving feed conversion rates in Precision Livestock Farming (PLF) is the early identification of growth impediments. However, the swine farming data collected by Electronic Feeding Station (EFS) are often disorganized and lack effective labeling. Data from healthy pigs are frequently intermixed with that from sick pigs, leading to label leakage and survivor bias in models, particularly when age is included as a feature. To address these known issues, this study breaks away from traditional modeling methods. First, we clean and classify the time-series data from electronic feeding stations, using age-cohort baselines as one of the criteria for determining high and low productivity, thereby avoiding problems such as label leakage. Next, we construct a high-dimensional feature matrix that captures dynamic derivatives such as feeding acceleration and weight gain acceleration, which together serve as behavioral feature fingerprints. To test the system, we optimized the mixed-model algorithm and evaluated the model based on behavioral deviations among individual pigs after removing all absolute age labels. Our results indicate that the full-feature model achieved an ROC-AUC of 0.778 and an F1-score of 0.4137 at the optimal threshold. Interestingly, SHAP attribution analysis revealed that “intake peer deviation,” “Cumulative Intake and Lifetime Avg Intake,” and “feeding acceleration” served as precursors to low productivity and growth retardation in this dataset, with these factors proving more significant than absolute feed intake or age. Our ablation experiments confirmed that a model based solely on behavioral features (excluding age labels) maintained an ROC-AUC of 0.773, successfully decoupling pig growth performance from growth stage. Our model can detect changes in feeding dynamic signatures at an average of 12.3 days, thereby providing insights for pig growth assessment, health monitoring, or more informed culling decisions. Full article

The study aimed to analyze how peasant economies in the municipality of Caicedonia recognize, classify, and manage functional biodiversity associated with coffee, plantain, and orange production systems to propose a contextualized framework for species accounting and ecological cost assessment within the coffee ecosystem. A qualitative interpretive approach with exploratory quantitative support was adopted through an exploratory descriptive design and participatory action research methodology. The study integrated 21 semi structured interviews conducted with producers managing approximately 61 associated crop units distributed across diversified farming systems. Data collection included field visits, direct observation, participatory species identification exercises, and thematic interviews focused on ecological functions, agricultural practices, biodiversity management, and perceived environmental impacts. The methodological framework additionally incorporated thematic coding, functional species classification, ecological cost identification, process and strategy mapping, descriptive frequency analysis, and multiple correspondence analysis to explore relationships among crop systems, species, ecological functions, management practices, and environmental pressures. The findings indicate that producers develop consistent empirical classifications regarding pests, pollinators, biological control organisms, and ecological indicators while recognizing cumulative ecological impacts associated with intensive agricultural practices. Quantitative exploration analysis revealed differentiated ecological configurations according to crop system and biodiversity management dynamics, supporting contextualized biodiversity accounting for sustainable agronomic decision making. Full article

Antimicrobial resistance is a major global health challenge that requires a One Health approach integrating humans, animals, wildlife, food systems and the environment. Among resistant bacteria, extended-spectrum β-lactamase-producing Escherichia coli (ESBL-producing E. coli) is particularly relevant because it is widely distributed across hosts and ecosystems, may carry mobile resistance genes and is commonly used as an indicator for antimicrobial resistance surveillance. This narrative review examines the occurrence, characteristics and transmission dynamics of ESBL-producing E. coli at the wildlife–livestock interface, with emphasis on its public health relevance and strategies for mitigation and control. The reviewed evidence indicates that livestock, wildlife and environmental matrices can be interconnected reservoirs of resistant E. coli and resistance genes. Transmission should not be interpreted as a simple linear process from livestock to wildlife or humans but rather as a bidirectional and ecological phenomenon shaped by antimicrobial use, farm management, biosecurity, wildlife ecology, environmental contamination and mobile genetic elements. Wildlife may function as a sentinel, reservoir or disperser of resistant bacteria, although detection alone does not demonstrate direct transmission. Integrated surveillance combining livestock, wildlife, food-chain and environmental sampling, supported by genomic analysis, is essential to clarify transmission pathways and guide effective control measures. Full article

The integration of wind energy into power systems relies on forecasting technologies to address operational challenges caused by its volatility and intermittency. This paper proposes a computing architecture for ultra-short-term wind power forecasting. The methodology integrates an adaptive dual-stage signal processing technique with an optimized deep learning model. To manage the non-stationarity of meteorological variables, the Pearson and Maximal Information Coefficient (MIC) analyses are employed for feature selection. The ICEEMDAN algorithm is then used for initial decomposition, followed by sample entropy and K-Means clustering to assess component complexity. Variational Mode Decomposition (VMD) is applied only to the high-frequency component to further separate stochastic fluctuations while preserving relatively stable trend components. A Convolutional Neural Network-Bidirectional Long Short-Term Memory (CNN-BiLSTM) network is constructed to forecast the resulting multi-scale components. To reduce reliance on manual empirical tuning, the Crested Porcupine Optimizer (CPO) is used to fine-tune key network hyperparameters. Evaluations using operational wind-farm data indicate that the developed hybrid method captures the temporal dynamics of wind power and yields lower prediction errors than the tested benchmark models. This research provides a data-driven computing framework for renewable-energy forecasting and related operational analysis. Full article

In large-scale rice farming systems, the design of efficient cropping schedules is essential for improving labor management and operational efficiency. However, climate change, including rising temperatures and increased frequency of extreme weather events, has altered crop growth dynamics, making it difficult to achieve optimal management using conventional experience-based scheduling. In addition, the need to distribute operations across numerous fields and optimize labor allocation has increased the complexity of schedule design. In this study, we propose a decision-support method for designing rice cropping schedules using a heading date prediction model and climatological temperature data. The method adjusts transplanting dates based on predicted heading and maturity dates and determines operation periods through both forward and backward scheduling. A case study conducted on a large-scale farming system in Chiba Prefecture demonstrated that the proposed method effectively dispersed the distribution of heading and maturity dates, leading to improved temporal distribution of operations. The standard deviation of heading dates decreased from 11.7 to 8.7 days, indicating a reduction in peak labor demand. The novelty of this study lies in extending a heading date prediction model from growth prediction to practical applications in cropping schedule design and visualization. This approach enables a transition from experience-based planning to data-driven decision-making and contributes to labor distribution in large-scale farming under climate change conditions. Full article

Using a complex network evolutionary game model, this study examines the effects of policy incentives, certification mechanisms, price premiums, production costs, and neighborhood learning on farmers’ adoption of organic farming technologies. It aims to reveal the dynamic mechanisms of organic farming technology diffusion under subsidy policies and certification mechanisms. Numerical simulations are conducted to analyze the effects of the subsidy rate and the effectiveness of organic certification on the diffusion level of organic farming technologies. The results show that both subsidy policies and certification mechanisms can promote the diffusion of organic farming technologies; however, the effect of subsidy policies is relatively limited, whereas certification mechanisms play a more significant role. Furthermore, the effects of the subsidy rate and certification effectiveness are influenced by factors such as the proportion of consumers with a preference for organic products, increased production costs, and the organic price premium. Under different levels of bounded rationality and strategy updating rules, the combined “subsidy–certification” policy consistently outperforms single-policy scenarios, with certification mechanisms generally exerting a stronger promotional effect than subsidy policies. In addition, the initial adoption proportion and network size also affect the evolutionary outcomes of the system. A higher initial adoption proportion cannot sustain a higher steady-state diffusion level in the long run, while an increase in network size tends to weaken the effectiveness of policy interventions. Finally, this study proposes policy recommendations, including improving certification and market development mechanisms and strengthening information dissemination and technical service systems, thereby providing practical insights for promoting the diffusion of organic farming technologies. Full article

Lumpy skin disease (LSD), a vector-borne viral disease of cattle, re-emerged in Italy in June 2025 after six years of absence in Europe, affecting the island of Sardinia, which had previously been disease-free. The insular setting, the predominance of extensive cattle farming systems, and the rapid implementation of control measures provided a unique opportunity to investigate epidemic dynamics and evaluate vaccination effectiveness under field conditions. This study aimed to describe the epidemiological pattern of the first epidemic season (June–October 2025), estimate key transmission parameters, and assess vaccination effectiveness at the farm level. Confirmed outbreaks consistent with local transmission and notified between 20 June and 26 October 2025 were analyzed to characterize epidemic transmission dynamics, while vaccination effectiveness was assessed over an extended follow-up period through 31 December 2025. The between-farm basic reproduction number (R₀) was estimated from the early exponential growth phase using log-linear regression and doubling time calculations. Spatio-temporal clustering was assessed using Kulldorff’s scan statistic under a Poisson model, accounting for the population at risk. Vaccination effectiveness was evaluated using a time-dependent Cox proportional hazards model with a 21-day post-vaccination lag. A total of 79 outbreaks were confirmed, of which 68 were consistent with local transmission. Affected farms included a total of 3443 cattle, with morbidity, mortality, and case fatality rates of 14.4%, 7.0%, and 31.1%, respectively. The exponential growth phase lasted four weeks, with an estimated growth rate of 0.366 per week and a doubling time of 1.89 weeks. The estimated R₀ ranged from 1.55 to 1.92, depending on the assumed generation time, indicating moderate but sustained transmission. The median apparent spatial spread velocity was 4.8 km/day. Spatio-temporal analysis identified a single highly significant cluster in the central-eastern area, accounting for approximately 27% of outbreaks (RR = 58.06; p < 0.001). Vaccination was associated with a substantial reduction in outbreak risk (HR = 0.18; 95% CI: 0.06–0.51; p = 0.001), corresponding to an estimated effectiveness of approximately 82% at the farm level. The 2025 Sardinian epidemic was characterized by moderate transmissibility and strong spatial clustering during the early phase. Rapid implementation of vaccination was associated with a significant reduction in outbreak risk, even under conditions of high infection pressure. The integration of spatio-temporal analyses and time-dependent modeling proved essential to support evidence-based control strategies in newly affected regions. Full article

The construction cost of an offshore wind farm collection system accounts for 15–30% of the total investment, and its efficient design is crucial to the economy; however, traditional methods in large-scale scenarios suffer from slow convergence and local optimization problems. In this study, we propose an upper and lower topology optimization framework based on the sled dog optimizer (SDO). The upper layer adopts polar coordinate partitioning combined with dynamic minimum spanning tree (DMST) to realize wind farm partitioning, and deals with the current-carrying capacity constraints and cable no-crossing requirements in a synchronized manner. The lower layer applies the SDO algorithm to optimize the topology structure within the partitioning range. The performance of genetic algorithm (GA), immunity algorithm (IA), particle swarm optimization (PSO), and SDO approaches is compared by the dynamic minimum spanning tree method through the case of an offshore wind farm with 62 wind turbines (WTs). The results show that the SDO-DMST framework significantly outperforms the comparison algorithms in terms of computational efficiency and cost optimization, and the proposed method can stably obtain high-quality cable topology solutions, which proves its superiority in unit group partitioning and cable routing co-optimization. In this paper, the SDO is introduced to collection system optimization for the first time, providing an efficient and robust design solution for large-scale offshore wind farms. Full article

This study examined the effect of east–west orientation of willow tree (Salix alba L.) rows on soil biological activity and weed dynamics in a temperate maize (Zea mays L.) intercropped agroforestry (AF) system in Eastern Hungary. The experiment evaluated how the year (2022, 2023), location (distance from the rows), and irrigation (IR) influenced spatial patterns of earthworm (EW) parameters and weed cover. The study aimed to assess how willow-based AF systems influence soil biological and weed community dynamics under varying IR and row spacing, in comparison with monoculture cropland (MC) systems, and to evaluate their potential role in climate change adaptation in arable farming. Both soil sampling for the EW survey and vegetation studies were conducted along perpendicular transects extending from the tree rows to measure EW abundance and biomass, as well as total weed cover. Experimental results revealed clear spatial gradients in EW distribution and weed abundance near the tree rows, driven by litter input, shading, moisture, and reduced disturbance. These effects were intensified under IR at narrower row spacings. No significant differences were observed between AF-South (shaded), AF-Center, and MC plots; however, significantly higher EW abundance and biomass were found on the AF-North (sunny) side. As for the location, significantly greater total EW abundance was found at AF-North (105.0 individual m⁻²) compared with the MC plots. AF systems enhance soil biological activity and shape weed dynamics through spatial ecological gradients influenced by tree row spacing and irrigation, supporting their role as sustainable land-use systems while emphasizing the need for site-specific management and further long-term optimization. Full article

This study investigated the effects of construction design and seasonal climatic conditions on bedding dynamics in bedded-pack dairy systems with contrasting composting functionality. The study intentionally included systems representing both composting bedded-pack barns (CBP), characterized by active management (regular turning and ventilation), and non-composting bedded-pack barns (BPB), which lacked aeration and did not promote active composting, resulting in limited or absent composting activity. Nine farms were divided into three groups: CONV (large, full-time CBP), ADAP (adapted, full-time CBP), and PART (partially used BPB). Evaluations were conducted during both cold and hot seasons. Composting dynamics were assessed over 24 h by measuring bedding temperature and moisture at eight points. During daytime, additional measurements at twenty points allowed for spatial distribution analysis using the inverse distance weighting method. Bedding attributes—including pH, density, depth, and particle size—were also measured in eight points. A 2 × 3 factorial design (two seasons, three barn types) was applied, and data were analyzed using Tukey’s test and Pearson correlation. Microclimate conditions were monitored through air temperature and humidity. Bedding temperature was significantly higher in the hot season (36.55 °C) compared to the cold season (32.12 °C), and was highest in the ADAP group (40.01 °C), followed by CONV (37.39 °C) and PART (26.18 °C) (p < 0.05). The 24 h temperature curve indicated favorable composting conditions only in the CONV and ADAP groups. Spatial temperature distribution varied significantly across locations in most barns (p < 0.05). Moisture content was lower in the hot season (46.91% and 41.41%) than in the cold season (57.03% and 51.97%) for CONV and ADAP, respectively. Moisture and temperature were significantly correlated with key bedding characteristics (p ≤ 0.05). Overall, a greater combination of characteristics associated with more favorable composting conditions was observed in ADAP barns, particularly during the hot season, whereas PART systems showed conditions incompatible with active composting. Full article