Search Results (6,182)

Karst aquifers, vital freshwater resources, are highly vulnerable to agricultural pollution, yet their hydro-geochemical responses remain poorly understood due to high spatial heterogeneity. This study aimed to unravel these complex responses in a subtropical karst agricultural catchment to provide a basis for its sustainable management. We employed high-frequency monitoring at a headwater spring (background), a depression well (hotspot), and the catchment outlet (integrated) in Southwest China. Using hydrological and geochemical data from 2017, we applied Principal Component Analysis (PCA) to apportion natural and anthropogenic sources. The main findings revealed significant spatial heterogeneity, with the depression well acting as a contamination hotspot characterized by rapid hydrological responses and elevated SO₄²⁻ and Cl⁻ concentrations. PCA successfully decoupled an “anthropogenic factor” (PC1, 40.5%) from a “natural weathering factor” (PC2, 25.2%). Critically, agricultural SO₄²⁻ at the hotspot was counter-intuitively higher during the wet season than the dry season, opposing the typical dilution pattern of background ions and revealing that depressions act as contaminant-concentrating pathways, whose risks are severely underestimated by traditional outlet monitoring. The anomalous sulfate dynamics reveal a cross-seasonal “storage-and-release” mechanism (legacy effect) within the karst Critical Zone, demonstrating that these systems can buffer and “remember” contaminants. Full article

Groundwater quality in the semi-arid region of Oum El Bouaghi, Northeastern Algeria, was assessed for irrigation suitability using hydrogeochemical analyses, water quality indices, and GIS techniques. The study analyzed 23 groundwater samples during dry and wet seasons in 2022–2023, several physicochemical parameters were measured. Results revealed neutral to slightly alkaline pH levels, except for one acidic sample, with salinity (EC: 527–5001 µS·cm⁻¹) exceeding WHO guidelines, particularly during the dry season due to evaporation and anthropogenic activities. Hydrogeochemical facies showed dominance of Na⁺-HCO₃⁻ and Ca²⁺-Cl⁻/SO₄²⁻ water types, indicating rock–water interactions and evaporation control, as confirmed by Gibbs plots. The IWQI classified water into five categories, with severe restrictions (IWQI < 40) in 13% of samples during the dry season, improving slightly in the wet season. Indices such as SAR, Na%, and RSC indicated low to moderate sodium hazard, while KR and PS highlighted salinity risks in specific areas. Spatial analysis revealed localized pollution hotspots, with the (GPI) identifying minimal to high contamination levels, linked to agricultural and geogenic sources. These findings underscore needs for sustainable groundwater management, including monitoring, optimized irrigation practices, and mitigation of anthropogenic impacts, to ensure long-term agricultural viability in this water-scarce region. Full article

Soil moisture plays a key role in the critical zone of the Earth and has extensive value in the understanding of hydrological, agricultural, and environmental processes (among others). Long-term (in situ) monitoring of soil moisture measurements is generally not practical; however, short-term measurements are often found. Limited soil moisture measurements can be employed to develop a numerical model for long-term and accurate soil moisture estimations. A key input variable to the model is precipitation, which is also not easily accessible, particularly at a finer spatial resolution; hence, publicly available remote sensing data can be used as an alternative. This study, therefore, aims to develop a numerical model HYDRUS-1D to estimate soil moisture in the data-scarce state of the Northern Territory, Australia, with a land cover of shrubland and a Tropical-Savannah type climate. The HDYRUS-1D is based on the numerical solution of Richards’ equation of variably saturated flow that relies on information about the soil water retention characteristics. This study utilized the van Genuchten model parameters, which were optimized (against measured soil moisture) through parameter optimization with initial estimates obtained from the HYDRUS catalogue. Initial estimates from different sources can differ for the same soil texture (e.g., loamy sand) and can induce uncertainties in the calibrated model. Therefore, a comprehensive uncertainty analysis was conducted to address potential uncertainties in the calibration process. The HYDRUS-1D was calibrated for a period between March 2012 and February 2013 and was independently validated against three different periods between March 2013 and October 2016. Root Mean Square Error (RMSE), Pearson’s correlation coefficient (R), and Mean Absolute Error (MAE) were used to assess the efficiency of the model in simulating the measured soil moisture. The model exhibited good performance in replicating measured soil moisture during calibration (RMSE = 0.00 m³/m³, MAE = 0.005 m³/m³, and R = 0.70), during validation period 1 (RMSE = 0.035 m³/m³ and MAE = 0.023 m³/m³, and R = 0.72), validation period 2 (RMSE = 0.054 m³/m³ and MAE = 0.039 m³/m³, and R = 0.51), and validation period 3 (RMSE = 0.046 m³/m³ and MAE = 0.032 m³/m³, and R = 0.61), respectively. Remotely sensed precipitation data were used from the CHRS-PERSIANN, CHRS-CCS, and CHRS-PDIR-Now to assess their capabilities in estimating soil moisture. Efficiency evaluation metrics and visual assessment revealed that these products underestimated the soil moisture. The CHRS-CCS outperformed other products in terms of overall efficiency (average RMSE of 0.040 m³/m³, average MAE of 0.023 m³/m³, and an average R of 0.68, respectively). An integrated approach based on numerical modelling and remote sensing employed in this study can help understand the long-term dynamics of soil moisture and soil water balance in the Northern Territory, Australia. Full article

Accurate land cover mapping is essential for environmental monitoring and agricultural management. Sentinel-2 imagery, with high spatial resolution and open access, provides valuable opportunities for operational classification. Convolutional neural networks (CNNs) have demonstrated state-of-the-art results, yet their adoption is limited by high computational demands and limited methodological transparency. This study proposes a lightweight CNN for soil–vegetation classification, in Dolj County, Romania. The architecture integrates three convolutional blocks, global average pooling, and dropout, with fewer than 150,000 trainable parameters. A fully documented workflow was implemented, covering preprocessing, patch extraction, training, and evaluation, addressing reproducibility challenges common in deep leaning studies. Experiments on Sentinel-2 imagery achieved 91.2% overall accuracy and a Cohen’s kappa of 0.82. These results are competitive with larger CNNs while reducing computational requirements by over 90%. Comparative analyses showed improvements over an NDVI baseline and a favorable efficiency–accuracy balance relative to heavier CNNs reported in the literature. A complementary ablation analysis confirmed that the adopted three-block architecture provides the optimal trade-off between accuracy and efficiency, empirically validating the robustness of the proposed design. These findings highlight the potential of lightweight, transparent deep learning for scalable and reproducible land cover monitoring, with prospects for extension to multi-class mapping, multi-temporal analysis, and fusion with complementary data such as SAR. This work provides a methodological basis for operational applications in resource-constrained environments. Full article

Subsoil compaction is a major constraint in sugarcane production, limiting yields and reducing resource-use efficiency. This study presents WireDepth, an innovative cloud-connected monitoring system that leverages edge computing and IoT technologies for real-time, spatially aware analysis and visualization of subsoiling depth. The system integrates ultrasonic, laser, inclinometer, and potentiometer sensors mounted on the subsoiler, with on-board microcontroller processing and dual wireless connectivity (LoRaWAN and NB-IoT/LTE-M) for robust data transmission. A cloud platform delivers advanced analytics, including 3D depth maps and operational efficiency metrics. System accuracy was assessed using 300 reference depth measurements, with Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) calculated per sensor. The inclinometer and potentiometer achieved the highest accuracy (MAPE of 0.92% and 0.84%, respectively), with no significant deviation from field measurements (paired t-tests, p > 0.05). Ultrasonic and laser sensors exhibited higher errors, particularly at shallow depths, due to soil debris interference. Correlation analysis confirmed a significant effect of depth on sensor accuracy, with laser sensors showing the strongest association (Pearson r = 0.457, p < 0.001). Field validation in commercial sugarcane fields demonstrated that WireDepth improves subsoiling precision, reduces energy waste, and supports sustainable production by enhancing soil structure and root development. These findings advance precision agriculture by offering a scalable, real-time solution for subsoiling management, with broad implications for yield improvement in compaction-affected systems. Full article

Global population growth, intensifying climate change, and escalating food security demands are mounting. In response, modern agriculture must transcend the limitations of traditional experience-based cultivation models to address issues such as low resource utilization, poor environmental adaptability, and significant yield fluctuations. As the core technical support of smart agriculture, agricultural sensors have become the key to transformation. This review systematically introduces the classification and working principles of current mainstream agricultural sensors: according to the monitoring parameters, they can be divided into humidity sensors, light sensors, gas sensors, pressure sensors, nutrient sensors, etc. At the same time, breakthroughs in emerging technologies such as microneedle sensing, nanosensing, and wireless sensor networks are being explored, which are breaking the application limitations of traditional sensors in complex agricultural environments. Combined with specific cases, the practical value of sensor technology is improving in agricultural drought monitoring, soil detection, and agricultural product quality assessment. Looking ahead, if agricultural sensors can overcome existing limitations through breakthroughs in material innovation, multi-sensor unit integration, and artificial intelligence algorithm fusion, this will provide stronger technological support for the further advancement of smart agriculture. Full article

Camouflaged Object Detection (COD) is a challenging computer vision task aimed at accurately identifying and segmenting objects seamlessly blended into their backgrounds. This task has broad applications across medical image segmentation, defect detection, agricultural image detection, security monitoring, and scientific research. Traditional COD methods often struggle with precise segmentation due to the high similarity between camouflaged objects and their surroundings. In this study, we introduce a Boundary-Guided Differential Attention Network (BDA-Net) to address these challenges. BDA-Net first extracts boundary features by fusing multi-scale image features and applying channel attention. Subsequently, it employs a differential attention mechanism, guided by these boundary features, to highlight camouflaged objects and suppress background information. The weighted features are then progressively fused to generate accurate camouflage object masks. Experimental results on the COD10K, NC4K, and CAMO datasets demonstrate that BDA-Net outperforms most state-of-the-art COD methods, achieving higher accuracy. Here we show that our approach improves detection accuracy by up to 3.6% on key metrics, offering a robust solution for precise camouflaged object segmentation. Full article

Leaf water content (LWC) is a vital physiological indicator reflecting crop water status, crucial for precision irrigation and water management. Traditional monitoring methods are labor-intensive and costly, while unmanned aerial vehicle (UAV) remote sensing offers an efficient alternative with high spatiotemporal resolution. This study developed an inversion model for winter wheat LWC based on a stacking ensemble learning framework integrating multispectral and texture features to improve estimation accuracy. UAV multispectral images collected at different growth stages were used to extract 17 vegetation indices (VIs) and 32 texture features (TFs). The top 10 features most correlated with LWC were selected to construct a fused dataset, and five machine learning models (SVM, RF, XGB, PLSR, RR) were combined within a base–meta stacking architecture. Results showed that: (1) Using only multispectral features yielded R² values of 0.526–0.718 and rRMSE of 22.795–29.536%, while texture-only models performed worse (R² = 0.273–0.425, rRMSE = 34.7–36.6%), indicating that single data sources cannot fully represent LWC variability. (2) Combining multispectral and texture features notably improved accuracy (R² = 0.748–0.815; rRMSE = 18.5–21.6%), demonstrating the complementary advantages of spectral and spatial information. (3) Stacking ensemble learning outperformed all single models, achieving the highest precision under fused features (R² = 0.865; rRMSE = 16.3%). (4) LWC distribution maps derived from the stacking model effectively revealed field-scale moisture differences and spatial heterogeneity during different periods. This study confirms that multi-source feature fusion combined with ensemble learning enhances UAV-based crop water estimation, offering a reliable and scalable approach for precision agricultural water monitoring. Full article

This study addressed the persistent limitation of discontinuous and labor-intensive compost monitoring procedures by developing and field-validating a low-cost sensor system for monitoring oxygen (O₂), carbon dioxide (CO₂), and methane (CH₄) under tropical windrow conditions. In contrast to laboratory-restricted studies, this framework integrated rigorous calibration, multi-layer statistical validation, and process optimization into a unified, real-time adaptive design. Experimental validation was performed across three independent composting replicates to ensure reproducibility and account for environmental variability. Calibration using ISO-traceable gas standards generated linear correction models, confirming sensor accuracy within ±1.5% for O₂, ±304 ppm for CO₂, and ±1.3 ppm for CH₄. Expanded uncertainties (U₉₅) remained within acceptable limits for composting applications, reinforcing the precision and reproducibility of the calibration framework. Sensor reliability and agreement with reference instruments were statistically validated using analysis of variance (ANOVA), intraclass correlation coefficient (ICC), and Bland–Altman analysis. Validation against a reference multi-gas analyzer demonstrated laboratory-grade accuracy, with ICC values exceeding 0.97, ANOVA showing no significant phase-wise differences (p > 0.95), and Bland–Altman plots confirming near-zero bias and narrow agreement limits. Ecological interdependencies were also captured, with O₂ strongly anticorrelated to CO₂ (r = −0.967) and CH₄ moderately correlated with pH (r = 0.756), consistent with microbial respiration and methanogenic activities. Nutrient analyses indicated compost maturity, marked by increases in nitrogen (+31.7%), phosphorus (+87.7%), and potassium (+92.3%). Regression analysis revealed that ambient temperature explained 25.8% of CO₂ variability (slope = 520 ppm °C⁻¹, p = 0.021), whereas O₂ and CH₄ remained unaffected. Overall, these findings validate the developed sensors as accurate and resilient tools, enabling real-time adaptive intervention, advancing sustainable waste valorization, and aligning with the United Nations Sustainable Development Goals (SDGs) 12 and 13. Full article

Ammonia (NH₃) is a major anthropogenic pollutant originating from agricultural activity, particularly livestock operations. NH₃ emissions from livestock slurry storage pose risks to environmental quality and human health. Reducing NH₃ emissions aligns with several United Nations Sustainable Development Goals (SDGs), including SDG 3, SDG 12, SDG 14, and SDG 15. This study evaluates the performance of the commercially available SOP^® LAGOON additive under real-scale farm conditions for mitigating NH₃ emissions. Two adjacent slurry storage tanks of a dairy farm in Northern Italy were monitored from 27 May to 7 September: one treated with SOP^® LAGOON and one left untreated (serving as a control). In the first month, the treated tank showed a 77% reduction in NH₃ emissions. Emissions from the treated tank remained consistently lower than those from the control throughout the monitoring period, reaching an 87% reduction relative to the baseline levels by the end of the period. The results suggest that SOP^® LAGOON is an effective and scalable strategy for reducing NH₃ emissions from liquid manure storage, with practical implications for farmers and policy makers in regard to designing sustainable manure management practices. Full article

Accurate and nondestructive monitoring of tomato growth is essential for large-scale greenhouse production; however, it remains challenging for small-fruited cultivars such as cherry tomatoes. Traditional 2D image analysis often fails to capture precise morphological traits, limiting its usefulness in growth modeling and yield estimation. This study proposes an automated phenotyping framework that integrates deep learning-based instance segmentation with high-resolution 3D point cloud reconstruction and ellipsoid fitting to estimate fruit size and ripeness from daily video recordings. These techniques enable accurate camera pose estimation and dense geometric reconstruction (via SfM and MVS), while Nerfacto enhances surface continuity and photorealistic fidelity, resulting in highly precise and visually consistent 3D representations. The reconstructed models are followed by CIELAB color analysis and logistic curve fitting to characterize the growth dynamics. When applied to real greenhouse conditions, the method achieved an average size estimation error of 8.01% compared to manual caliper measurements. During summer, the maximum growth rate (g_max) of size and ripeness were 24.14%, and 95.24% higher than in winter, respectively. Seasonal analysis revealed that winter-grown tomatoes matured approximately 10 days later than summer-grown fruits, highlighting environmental influences on phenological development. By enabling precise, noninvasive tracking of size and ripeness progression, this approach is a novel tool for smart and sustainable agriculture. Full article

This study aimed to assess the potential of machine learning models applied to high spatial resolution images from UAVs for estimating the aboveground biomass (AGB) of forage grass cultivated in the Brazilian semiarid region. The fresh and dry AGB were determined in Cenchrus ciliare plots with an area of 0.04 m². Spectral data were obtained using a multispectral sensor (Red, Green, and NIR) mounted on a UAV, from which 45 vegetation indices were derived, in addition to a structural variable representing plant height (H95). Among these, H95, GDVI, GSAVI2, GSAVI, GOSAVI, GRDVI, and CTVI exhibited the strongest correlations with biomass. Following multicollinearity analysis, eight variables (R, G, NIR, H95, CVI, MCARI, RGR, and Norm G) were selected to train Random Forest (RF), Support Vector Machine (SVM), and XGBoost models. RF and XGBoost yielded the highest predictive performance, both achieving an R² of 0.80 for AGB—Fresh. Their superiority was maintained for AGB—Dry estimation, with R² values of 0.69 for XGBoost and 0.67 for RF. Although SVM produced higher estimation errors, it showed a satisfactory ability to capture variability, including extreme values. In modeling, the incorporation of plant height, combined with spectral data obtained from high spatial resolution imagery, makes AGB estimation models more reliable. The findings highlight the feasibility of integrating UAV-based remote sensing and machine learning algorithms for non-destructive biomass estimation in forage systems, with promising applications in pasture monitoring and agricultural land management in semi-arid environments. Full article

The complex life cycle, high reproductive potential and ability to quickly develop resistance to insecticides are key factors contributing to the destructiveness of the peach–potato aphid (Myzus persicae Sulzer) among pest species. Monitoring its population dynamics at a large scale allows us to better understand M. persicae biology and take relevant measures for pest management. For this purpose, reliable molecular tools are needed. Based on the analysis of 128,362 microsatellite loci, we developed four multiplex assays including 49 comprehensively characterised SSR markers. Internal validation confirmed the species specificity and low genotyping error (e_a = 0.8%, e_l = 0.99%, e_obs = 22.7%) of the assays. A total of 194 alleles were identified (mean = 4 alleles per locus, range = 2–8 alleles per locus) within a group of 365 aphid accessions collected in the Vysočina region (Czechia). The studied aphid population showed the typical characteristics expected of the species with clonal or partially clonal reproduction (heterozygote excess, negative F_IS, moderate-to-high linkage disequilibrium (LD), and distortion of the H-W equilibrium for most of the loci), and did not exhibit any stratification on a spatiotemporal level. Owing to the high discriminatory power of the markers, we discovered that the population sample was founded upon a small number of fundatrices, as only five dominating lineages comprising over 70% of all accessions were identified. In conclusion, this study identified a significant number of new high-quality markers with the high discriminatory power necessary for revealing the population structure and dynamics of M. persicae, which holds considerable potential in both general biological and agricultural research. Full article

The use of digital tools in agroecological crop production can help mitigate current farming challenges such as labour shortage and climate change. The aim of this study was to map digital tools used in crop production, assess their impacts across economic, environmental, and social dimensions, and determine their potential as enablers of agroecology. A systematic search and screening process, following the Preferred Reporting Items for Systematic reviews and Meta-Analyses methodology, identified 453 relevant studies. The results showed that most digital tools are applied for crop monitoring (83.4%), with unmanned aerial vehicles (37.7%) and camera sensors (75.2% combined) being the most frequently used technologies. Farm Management Information Systems (57.6%) and Decision Support Systems (25.2%) dominated the tool categories, while platforms for market access, social networking, and collaborative learning were rare. Most tools addressed the first tier of agroecology, which refers to input reduction, highlighting a strong focus on efficiency improvements rather than systemic redesign. Although digital tools demonstrated positive contributions to social, environmental, and economic dimensions, studies concentrated mainly on economic benefits. Future research should investigate the potential role of digital technologies in advancing higher tiers of agroecology, emphasising participatory design, agroecosystem services, and broader coverage of the agricultural value chain. Full article

Smart Agriculture (SA) combines cutting edge technologies such as the Internet of Things (IoT), Artificial Intelligence (AI), and real-time sensing systems with traditional farming practices to enhance productivity, optimize resource use, and support environmental sustainability. A key aspect of SA is the continuous monitoring of field conditions, particularly Soil Moisture (SM), which plays a crucial role in crop growth and water management. Accurate forecasting of SM allows farmers to make timely irrigation decisions, improve field management, and conserve water. To support this, recent studies have increasingly adopted soil sensors, local weather data, and AI-based data-driven models for SM forecasting. In the literature, most existing review articles lack a structured framework and often overlook recent advancements, including privacy-preserving Federated Learning (FL), Transfer Learning (TL), and the integration of Large Language Models (LLMs). To address this gap, this paper proposes a novel taxonomy for SM forecasting and presents a comprehensive review of existing approaches, including traditional machine learning, deep learning, and hybrid models. Using the PRISMA methodology, we reviewed over 189 papers and selected 68 peer-reviewed studies published between 2017 and 2025. These studies are analyzed based on sensor types, input features, AI techniques, data durations, and evaluation metrics. Six guiding research questions were developed to shape the review and inform the taxonomy. Finally, this work identifies promising research directions, such as the application of TinyML for edge deployment, explainable AI for improved transparency, and privacy-aware model training. This review aims to provide researchers and practitioners with valuable insights for building accurate, scalable, and trustworthy SM forecasting systems to advance SA. Full article