Search Results (4,049)

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

As the Italian hop industry undergoes consolidation, assessing the environmental pressure of diverse cultivation and processing models is essential for sustainable growth. This study characterizes the Product Environmental Footprint (PEF) of Italian hop production through a multi-case analysis of eight representative farms. A primary data collection tool was utilized to quantify resource inputs, including water management, nutritional strategies, and phytosanitary defense. Following a rigorous thermodynamic consistency screening of the field data to eliminate unrepresentative parameters, the life cycle inventory focused on two validated regional anchor cases. The findings reveal a high degree of management heterogeneity, with dry cone yields ranging from 400 to 1673 kg of dry matter per hectare. Two functional units were defined: 1 kg of fresh hop cones (FU1) to assess cultivation impacts, and 1 kg of processed products (FU2) at the brewery gate to evaluate the full supply chain. Integrating deterministic life cycle impact outputs with a probabilistic Monte Carlo uncertainty analysis, the results indicate that the environmental impact varies significantly across commercial formats: Cryogenic Powder (2.33 ± 0.34 mPt/kg) represents the most resource-intensive format, while Raw Bales and T90 Pellets from high-yield models exhibit scores as low as 1.36 and 1.55 mPt/kg, respectively. The study identifies the agricultural phase as the primary environmental hotspot, driven predominantly by water deprivation. To address these burdens, a Sustainable Italian Hop (SIH) integrated scenario was developed. By combining precision irrigation, thermal decarbonization via biomass valorization, and a direct-to-pellet processing flow, this model achieved a 70% total reduction in the environmental footprint score (0.465 ± 0.076 mPt/kg) and an 86% reduction in water use impacts. Finally, the socio-technical and financial barriers to implementing the SIH framework are qualitatively evaluated. These results provide actionable benchmarks for aligning the emerging Italian hop supply chain with European Union climate neutrality objectives. Full article

Horticultural intensification in Mediterranean areas has increased the risk of nitrate pollution due to inefficient irrigation and nitrogen fertilisation management. The availability of simulation models aimed at rational nitrogen management in outdoor crops is limited. The objective of this study is to adapt the VegSyst model, initially developed for greenhouse vegetables, for use in open-field conditions in relevant crops, such as processing peppers and broccoli in Extremadura. VegSyst simulates dry matter production and nitrogen uptake by incorporating the influence of evaporative demand (TUE approach) in addition to the effect of radiation (RUE approach). Experimental field data obtained in five campaigns (peppers: 2020–2022; broccoli: 2020 and 2022) under different nitrogen doses were used. The model was calibrated, and critical N dilution curves were developed for each crop. Subsequently, the simulation of fi-PAR, dry matter production (DMP) and N uptake was validated using statistical indices (RMSE, RE, d, EF) and regression analysis. The model showed a high predictive capacity for N uptake in both crops, with values of d ≥ 0.98 and EF ≥ 0.90 in the validation campaigns. The fi-PAR simulation was acceptable in peppers and excellent in broccoli. In contrast, the DMP prediction showed notable deviations in peppers, especially in 2022, attributable to interannual variations in weather conditions and physiological limitations not considered by the model. In both crops, the TUE-based strategy was a better fit for the measurements than the RUE-based strategy, indicating that under semi-arid Mediterranean conditions, transpiration is the limiting factor for biomass production. The adaptation of the VegSyst-Outdoors model proved to be robust for simulating N uptake and sufficiently accurate to be integrated into decision support tools aimed at efficient fertilisation and irrigation management. Full article

Although pathogenic Leptospira DNA has been detected in water and soil from different climatic regions, information from Mediterranean-climate agricultural systems remains limited. This study characterized the environmental detection of pathogenic Leptospira DNA in water and soil samples from irrigated agroecosystems of central Chile, evaluating spatial and seasonal variation and associations with selected physicochemical variables. A total of 605 samples were collected from eight agricultural sites during spring 2019, summer 2020, and winter 2021. Samples were analyzed by real-time PCR targeting lipL32. Overall, 29.1% of samples were PCR-positive, and pathogenic Leptospira DNA was detected in all sites and seasons. Soil samples showed higher positivity than water samples (34.5% vs. 21.4%), and positivity was higher in summer (41.7%) than in spring (22.7%) or winter (19.3%). Water temperature and turbidity were the only physicochemical variables that differed between positive and negative samples, whereas the binomial generalized linear mixed model (GLMM) showed that season and sample type were associated with PCR positivity after accounting for site-level clustering. These results show that pathogenic Leptospira DNA can be widely detected in irrigated agricultural systems from a Mediterranean-climate region, suggesting that soil, seasonality, irrigation practices, and other site-level characteristics should be considered in future studies on the environmental ecology of pathogenic Leptospira. Full article

Reservoir construction and agricultural irrigation have substantially altered the natural hydrological regimes of many Mediterranean watersheds. This study aims to reconstruct the natural flow conditions of the Egirdir Lake Basin (Türkiye) and quantify the impacts of irrigation and reservoir operations on water inflows to Egirdir Lake using the Soil and Water Assessment Tool (SWAT). The SWAT model consisted of 14 subbasins and 274 hydrologic response units (HRUs) and initially calibrated and validated using naturalized flow data provided by the State Hydraulic Works (DSI) for the period from 1990 to 2014. The same model structure and parameters were then applied to simulate a regulated condition representing the combined effects of irrigation and reservoir operation. Results showed a considerable reduction in annual streamflows under the regulated condition. This study demonstrated the significant impact of irrigation water use and reservoir operation on the hydrological dynamics of semi-arid basins. Full article

Here, we consider a nonlinear hydrodynamic model with mixed dispersion–temporal evolution as the scalar version of the generalized shallow-water wave equation, which specifically provides a comprehensive and versatile framework for studying energy propagation in nonlinear fluids of constrained depth. This equation is acknowledged as an integrable model in the analysis of tidal wave dynamics and in simulations of weather variations, tsunami prediction, and irrigation flows. We also investigate a few of its singular and periodic solitary wave solutions by employing various Riccati-based ansatzes. These results highlight the necessity of studying various nonlinear wave phenomena, which may have potential applications in various domains of physics and applied mathematics. These results extend the variety of its solutions and also enrich the existing knowledge about its solutions with various profiles. To improve visual clarity and to facilitate structural understanding, the solution profiles are represented graphically using Maple software in 3D, 2D, and contour plots.We also discuss its invariance under infinitesimal transformations, which yields a one-dimensional Hamilton–Jacobi-like equation. Full article

Incomplete spatio-temporal (ST) data from sensor networks in precision agriculture often limits environmental modeling and decision-making accuracy. To address this, we propose the Spatio-Temporal Conditional Diffusion Framework (ST-CDF), a generative approach for high-fidelity data reconstruction. The framework’s core is a deep denoising network that integrates a Graph Attention Network (GAT) to explicitly model non-Euclidean spatial correlations, a Differential Attention Transformer to capture abrupt temporal dynamics, and an Inverse Discrete Wavelet Transform (IDWT) module to preserve multi-scale signal details. The generative process is constrained by a physics-informed training objective, which injects known physical laws (i.e., the Penman–Monteith equation for reference evapotranspiration, $E{T}_0$) as an inductive bias, ensuring the imputed data maintains physical consistency. For privacy-preserving deployment on resource-constrained IoT devices, we extend the framework with a Federated Cluster-Guided Distillation (Fed-CGD) strategy. We conducted extensive experiments against established methods on two real-world agricultural datasets. ST-CDF demonstrated improved imputation accuracy across evaluated metrics. Its efficacy was most pronounced in the physically-demanding $E{T}_0$ calculation task, where data imputed by ST-CDF at an 80% missing rate achieved a Root Mean Square Error (RMSE) of 0.3485 and a Coefficient of Determination ($R^2$) of 0.7558, outperforming the baseline models. Furthermore, we explore ST-CDF as an explainable (XAI) framework for active agricultural decision support, demonstrating its utility in performing counterfactual simulations of “what-if” interventions, such as irrigation. The findings highlight ST-CDF as an effective, physically-grounded, and interpretable tool for data-driven scientific computation and precision agriculture. Full article

Controversy persists on a global scale regarding the trade-offs between greenhouse gas (GHG) emissions, yield, the global warming potential (GWP), and GHG intensity (GHGI) following organic fertilizer substitution within vegetable cropping systems. This study aimed to quantify these effects under diverse conditions and elucidate the direct and indirect drivers governing these outcomes through a meta-analysis and structural equation modeling (SEM). We synthesized 655 paired observations from 69 published studies using random-effects meta-analysis, finding that organic fertilizer substitution significantly increased CH₄ emissions and GWP compared to inorganic fertilizer controls. Although this was the general trend, organic fertilizer could reduce GWP under specific climatic and soil conditions by reducing N₂O emissions, such as mean annual precipitation <400 mm or soil total nitrogen ≥3 g kg⁻¹. These conditions were also associated with substantially higher yield and lower GHGI. Furthermore, SEM demonstrated that field management practices exerted significant direct effects on N₂O emissions, GWP, and GHGI. Reductions in N₂O emissions, GWP, and GHGI could be achieved with fertilizer application duration ≥10 years, total N application rate ≥300 kg ha⁻¹, and field cultivation or plowing. GHGI was also reduced through yield enhancement under a moderate organic substitution rate (33–66%) or irrigation ≥300 mm. Our study provides a scientific basis for moving beyond universal recommendations towards precision organic management, which is essential for optimizing fertilization strategies to mitigate agricultural GHG emissions. Full article

Background/Objectives: This investigation was performed because corticosteroid injections are commonly used for symptomatic relief in patients with meniscal deficiency, yet their effect on graft survivorship and postoperative outcomes following meniscal allograft transplantation (MAT) remains poorly understood, with limited literature specifically addressing this topic. The aim of this study is to evaluate whether preoperative intra-articular corticosteroid injections (ICS) are associated with reoperation after MAT. Secondary aims included comparing reoperation-free survival, patient-reported outcome measures (PROMs), and patient acceptable symptom state (PASS) achievement. Methods: A retrospective review of 130 adults undergoing meniscal allograft transplantation (MAT) between 2011 and 2023 was performed. Patients with documented corticosteroid injection (CSI) status and ≥2 years of follow-up were included. Exclusion criteria included prior meniscal allograft transplantation, receipt of non-corticosteroid injections (e.g., hyaluronic acid or platelet-rich plasma), concomitant osteotomy procedures, multi-ligament knee reconstruction or inadequate follow-up. Propensity score matching (2:1 no steroid: steroid) based on age, sex, body mass index, fixation technique, operative compartment, and concomitant procedures yielded 54 matched patients (35 no steroid, 19 steroid). The primary outcome was ipsilateral knee reoperation, categorized as major reoperation (revision MAT, anterior cruciate ligament reconstruction, osteochondral allograft transplantation, conversion to total knee arthroplasty, meniscectomy and meniscus repair). Minor reoperations included irrigation and debridement, lysis of adhesions or manipulation under anesthesia, hardware removal, chondroplasty, and synovectomy. Reoperation-free survival was assessed using Kaplan–Meier analysis. PROMs and PASS were compared using adjusted regression models. Statistical significance was set at p < 0.05. Results: Baseline characteristics and follow-up were comparable between groups (7.6 ± 3.5 vs. 6.6 ± 3.2 years; p = 0.30). Overall reoperation occurred in 37.1% of patients in the no-steroid group and 31.6% in the steroid group (p = 0.771). Major reoperation rates were similar (17.1% vs. 15.8%; p = 1.000. There was no significant difference in minor reoperations between groups (20.0% vs. 10.5%; p = 0.468). Kaplan–Meier analysis demonstrated no difference in reoperation-free survival (p = 0.903), with comparable survival at the 1-, 2-, and 5-year time points. No individual subtypes differed significantly between groups. PROMs and PASS achievement were also similar, with no statistically significant differences observed. Conclusions: Preoperative corticosteroid injection was not associated with increased reoperation risk, inferior reoperation-free survival, or worse patient-reported outcomes following meniscal allograft transplantation. However, given the study’s limited power, lack of detailed injection characteristics, and the use of a heterogeneous complication outcome, these findings should be interpreted cautiously, as further investigation is warranted. Full article

The quality of rice grain impacts milling yield, market acceptance, and product value. Physical quality is determined by many traits, such as chalkiness, whiteness, vitreous whiteness, caryopsis length, and width. Breeding for these traits is challenging due to their quantitative nature, environmental effects, and time and labor requirements to evaluate these traits. The digital grain analyzer (S21) equipment determines rice grain physical quality by image-based analysis; however, its use remains restricted. Thus, here we aimed to evaluate S21 efficiency to determine the physical quality of rice grains and estimate the genetic diversity of the trait using a Brazilian panel of 152 irrigated rice genotypes as a working model. We accessed total whiteness, vitreous whiteness, chalkiness degree, chalky grain rate, white belly, grain length, width, and length/width ratio. Our results demonstrated that S21 allowed the characterization of the genotypes according to physical traits, facilitating grouping and separation of accessions and correlation analyses between quality traits. It was also possible to estimate the heritability of quality traits. S21 was efficient in characterizing the physical quality of rice grains and determining their genetic diversity. The equipment is an effective tool exhibiting potential application by breeder programs. Full article

With the occurrence of global climate change and the depletion of agricultural water resources, there is a growing need to develop rapid, non-destructive, and autonomous plant health monitoring systems. As an economically valuable crop, Ocimum basilicum L. (basil) is sensitive to changes in water availability and may exhibit stress-related morphological variations under drought and over-irrigation conditions. However, due to the visual similarity of leaf symptoms under drought stress, waterlogging stress, and optimal irrigation conditions, accurately distinguishing these conditions remains challenging in practical applications. To address this challenge, this paper presents an attention-based dual-branch deep learning framework designed to extract both subtle leaf details and channel-related features from high-resolution plant images. By combining the Convolutional Block Attention Module (CBAM) and Squeeze-and-Excitation (SE) mechanism in a parallel structure, the proposed network improves the analysis of high-resolution images with an input size of 720 × 720 pixels. Under controlled environmental conditions, with ground-truth labels obtained using soil moisture sensor measurements, the proposed model was compared with eight deep learning architectures, including DenseNet121, InceptionV3, and VGG16. The proposed model achieved a hold-out evaluation accuracy of 99.54%, outperforming the second-best model, DenseNet121, which achieved 96.43%. In addition, the proposed model reached a class-specific precision value of 100% for the Drought Stress category and achieved an area under the receiver operating characteristic curve of 1.00 under the controlled experimental setting. Taylor Diagram analysis also indicated that the model closely preserved the variability pattern of the reference data. These results suggest that the proposed application-specific framework may support non-destructive basil water-stress detection under controlled conditions. After further validation with larger datasets, different cultivars, variable environmental conditions, and real-world agricultural scenarios, the proposed approach may contribute to precision irrigation management and sustainable agricultural production. The contribution of this study should be interpreted as an application-specific implementation and evaluation of complementary attention mechanisms for controlled-environment basil water-stress classification, rather than as the introduction of a fundamentally new deep learning methodology. Full article

Saline irrigation is increasingly practiced in semi-arid regions to cope with freshwater scarcity; however, it strongly affects crop growth, water use, and soil salinity. This study aims to calibrate and validate the AquaCrop model to simulate key growth parameters of winter wheat (cv. Achtar) under saline irrigation conditions in the Tadla Plain, Morocco, focusing on canopy cover (CC), actual evapotranspiration (ETa), soil water content (SWC), biomass (B), and grain yield (GY). The model was first calibrated using observed data from the 2023 growing season and subsequently validated using data from the 2022 growing season. Overall, AquaCrop effectively reproduced crop growth during both calibration and validation phases. During calibration, canopy cover was accurately simulated, with average RMSE values below 1%, while biomass and grain yield were also well reproduced, with low RMSE values (0.25 t ha⁻¹ for B and 0.10 t ha⁻¹ for GY), confirming the robustness of the calibrated parameters. The model also performed well in simulating ETa and SWC, capturing the seasonal dynamics of crop water use and soil moisture. During validation, ETa was satisfactorily reproduced, with an RMSE of approximately 0.80 mm day⁻¹, while SWC showed good agreement with observations, with NRMSE values ranging from 7.9 to 10.5%. Grain yield and biomass were reliably predicted, with NRMSE values below 4%. These results demonstrate that AquaCrop is a reliable tool for simulating winter wheat under saline irrigation and for assessing crop response under salt-affected conditions, providing an integrated evaluation of crop performance, water use, and soil salinity dynamics to support improved irrigation management and water-use efficiency under semi-arid conditions. Full article

The rising global population has heightened food demand, placing pressure on agricultural systems, particularly in water-scarce regions such as South Africa. Smallholder farmers, essential to the sector, face climatic variability and resource constraints, necessitating innovative solutions to enhance sustainability and productivity. This study evaluates unmanned aerial vehicles (UAVs) for generating spatially explicit evapotranspiration (ET) estimates in a small-scale sugarcane field, supporting precision water management. Vegetation indices (VIs) derived from UAV-based multispectral imagery were used to predict actual ET (ET_a) and validated against eddy covariance measurements. Five models were assessed, including Normalised Difference Vegetation Index (NDVI)-based and Enhanced Vegetation Index (EVI)-based approaches. Machine learning was used to relate crop coefficients (K_c) to NDVI, enabling improved estimation. The two-band EVI (EVI2) model achieved the highest accuracy, with an R² of 0.63, an RMSE of 0.67, and an MAE of 0.52. ET-VI approaches, particularly EVI2, require lower data and technical complexity, making them suitable for smallholder systems. However, reducing dependence on in situ data remains essential to improve accessibility of remote sensing approaches for agricultural water management in resource-limited environments. These findings demonstrate the potential of UAV-based ET_a modelling to support field-scale irrigation decision-making while highlighting the need for further refinement to improve operational applicability across diverse smallholder farming contexts and beyond. Full article

Optimizing nitrogen management is essential for maintaining high spring maize yield while mitigating nitrous oxide (N₂O) emissions in irrigated areas. However, the interactive effects of total nitrogen application rate and starter nitrogen proportion on yield and N₂O emissions remain insufficiently quantified. Reliable assessment of these interactions requires well-calibrated DeNitrification–DeComposition (DNDC) simulations, yet existing calibration studies often emphasize crop parameters while neglecting soil parameters critical for soil hydrothermal dynamics and N₂O production. In this study, field data from shallow-buried drip-irrigated spring maize in Tongliao during 2024–2025 were used to conduct Extended Fourier Amplitude Sensitivity Test (EFAST) sensitivity analysis on 12 crop and 13 soil parameters of the DNDC model. Sensitive parameters were calibrated using the differential evolution algorithm, and 64 nitrogen management scenarios were simulated by combining eight total nitrogen application rates (100, 150, 200, 250, 300, 350, 400, and 450 kg N ha⁻¹) with eight starter nitrogen proportions (0%, 15%, 25%, 30%, 35%, 40%, 45%, and 50% of the total nitrogen rate). The results showed that DNDC outputs were jointly controlled by crop and soil parameters, among which maximum yield, leaf carbon-to-nitrogen ratio, stem fraction, grain carbon-to-nitrogen ratio, thermal degree days for maturity, grain fraction, soil organic carbon (SOC) decrease rate below topsoil, soil clay content, soil porosity, wilting point and depth of top soil with uniform SOC content were dominant. Compared with the conventional crop-parameter calibration, the sensitivity-screened parameter set improved the simulation of both cumulative N₂O emissions and yield. Across the 64 scenarios, cumulative N₂O emissions ranged from 0.42 to 4.87 kg [N]/ha, while simulated maize yield ranged from 1597 to 6347 kg [C]/ha. N₂O emissions increased with total nitrogen rate, whereas yield increased initially and then reached a plateau. Increasing the starter nitrogen proportion did not substantially enhance yield but increased N₂O emission risk under high nitrogen rates. Overall, the scenario with 300 kg/ha and no nitrogen applied at sowing achieved a relatively high yield of 5519 kg [C]/ha while maintaining a low cumulative N₂O emission of 0.98 kg [N]/ha and was therefore identified as the preferred trade-off strategy under shallow-buried drip irrigation. This study provides an EFAST–DNDC framework for optimizing nitrogen management to sustain spring maize yield while reducing N₂O emissions in the West Liaohe Plain. Full article

Precision agriculture increasingly relies on high-resolution, long-term remote sensing to delineate sub-field management zones. However, traditional spatial zonation assumes temporal stationarity, utilizing seasonal aggregates that obscure transient, intra-annual stress signals. This study develops a data-driven framework to characterize both persistent and non-stationary crop water use dynamics by integrating monthly, 30-m evapotranspiration (ET) data from OpenET (2000–2025) with zero-shot temporal anomaly detection. A pre-trained time-series foundation model (Chronos-T5-Small) generated counterfactual expectations for sub-field ET, quantifying deviations using a mean absolute error-based anomaly score. Unsupervised clustering of these anomaly scores with longitudinal ET metrics partitioned the landscape into dynamic biophysical regimes. Cross-registered against legacy persistence mapping based on seasonal totals, the foundation model showed strong directional agreement (86.1%, Cohen’s Kappa = 0.716) in identifying chronically constrained zones across 869 shared active pixels. Crucially, the framework identified 966 historically persistent pixels undergoing stability decay, of which 95.3% were statistically verified via paired t-tests to have collapsed into the field’s baseline variance pool. Furthermore, counterfactual anomaly detection isolated zones of recent acute divergence, differentiating enduring edaphic constraints from sudden system disruptions. This approach demonstrates how foundation models can transition from purely predictive engines to diagnostic instruments, advancing operational precision agriculture. Full article