Search Results (8,271)

Uniform application of fertilizers and pesticides continues to dominate global agriculture despite significant spatial variability in soil and crop conditions. This mismatch results in avoidable yield gaps, excessive chemical waste, and environmental pressures, including nutrient leaching and greenhouse gas emissions. The integration of Artificial Intelligence (AI) and Remote Sensing (RS) has emerged as a transformative framework for diagnosing this variability and enabling site-specific, climate-responsive management. This systematic synthesis reviews evidence from 2000–2025 to assess how AI–RS technologies optimize agrochemical efficiency. A comprehensive search across Scopus, Web of Science, IEEE Xplore, ScienceDirect, and Google Scholar were used. Following rigorous screening and quality assessment, 142 studies were selected for detailed analysis. Data extraction focused on sensor platforms (Landsat-8/9, Sentinel-1/2, UAVs), AI approaches (Random Forests, CNNs, Physics-Informed Neural Networks), and operational outcomes. The synthesized data demonstrate that AI–RS systems can predict critical soil attributes, specifically salinity, moisture, and nutrient levels, with 80–97% accuracy in some cases, depending on spectral resolution and algorithm choice. Operational implementations of Variable-Rate Application (VRA) guided by these predictive maps resulted in fertilizer reductions of 15–30%, pesticide use reductions of 20–40%, and improvements in water-use efficiency of 25–40%. In fields with high soil heterogeneity, these precision strategies delivered yield gains of 8–15%. AI–RS technologies have matured from experimental methods into robust tools capable of shifting agrochemical science from reactive, uniform practices to predictive, precise strategies. However, widespread adoption is currently limited by challenges in data standardization, model transferability, and regulatory alignment. Future progress requires the development of interoperable data infrastructures, digital soil twins, and multi-sensor fusion pipelines to position these technologies as central pillars of sustainable agricultural intensification. Full article

The high incidence of anthropogenic impacts in the Mediterranean basin raises concerns on the health and quality of commercial fish species. This study aims to evaluate the health status of the European hake, Merluccius merluccius, from three areas of the Catalan coast (NW Mediterranean Sea) with different anthropogenic impacts (i.e., chemical pollution, litter, …) and assess if hake could serve as a sentinel species. We measured biomarkers of chemical exposure including B-esterases, antioxidant enzymes (GST, GR, GPx, CAT), biotransformation markers (EROD), lipid peroxidation, and macro-parasite assemblages. Hake showed, generally, a good health status across all areas with homogeneous patterns for most parameters. Tissue-specific differences included elevated gonadal cholinesterases and higher brain and hepatic carboxylesterase activities in the south, and increased hepatic EROD but lower lipid peroxidation in the central Barcelona area. Parasite assemblages were dominated by Digenea, Cestoda, and Nematoda, with higher cestode prevalence in both central and south zones. In summary, despite a greater prevalence of environmental pollution in the central region, there was a homogeneous pattern in hake health indicators throughout the three studied fishing zones. These results establish a baseline for hake health in Mediterranean waters and suggest that the species’ high mobility and wide depth range may limit its utility to detect local-scale pollution impacts, though it may serve as a regional-scale bioindicator. Full article

Urban river floodplains function not only as zones for flood regulation and ecological buffering but have increasingly been utilized as multifunctional spaces that support leisure, waterfront, and cultural activities. However, overlapping hydraulic and geomorphic factors such as channel meandering, vegetation distribution, and flood-induced flow redistribution have amplified environmental risks, including recurrent erosion deposition, vegetation disturbance, and infrastructure damage, yet quantitative assessment frameworks remain limited. This study systematically evaluates the environmental safety of an urban floodplain by estimating vegetation variability using Sentinel-2 derived NDVI time series and deriving SEDI and TEDI through FaSTMECH two-dimensional hydraulic modeling. NDVI response cases were identified for different rainfall intensities, and interpolation-based hazard maps were generated using spatial cross-validation. Results show that the left bank exhibits higher vegetation variability, indicating strong sensitivity to hydrological fluctuations, while outer meander bends repeatedly display elevated SEDI and TEDI values, revealing concentrated structural vulnerability. Integrated analyses across rainfall conditions indicate that overall safety remains high; however, low-safety zones expand in the upstream meander and several outer bends as rainfall intensity increases. Full article

Leaf area index (LAI) is essential for understanding vegetation dynamics, ecosystem processes, and land–atmosphere interactions. Various measurement methods exist, but gap-fraction-based indirect methods are preferred due to their reduced labor and field survey time in comparison to direct methods. However, gap-fraction-based field observations, often referred to as plant area index (PAI), frequently overestimate LAI because they include woody components. To mitigate this issue, the woody-to-total-area ratio (α) can be utilized to exclude these woody components from PAI, yielding more accurate LAI estimates (hereafter referred to as LAI_adjusted). In this study, we demonstrate a novel method to estimate α using Sentinel-2-based normalized difference vegetation index (NDVI) and time-series PAI measurements. The α estimates effectively reduce the influence of woody components in PAI within deciduous broadleaf forests (DBF). Moreover, LAI_adjusted shows good agreement with the Sentinel-2 LAI, which represents effective LAI derived from the PROSAIL model. Notably, the spatial distribution of α effectively captured the expected seasonal dynamics across various forest types. In DBF, α values increased during winter due to leaf fall when compared to the growing season, while seasonal variations were relatively small in evergreen needleleaf forest (ENF). We further confirmed that our method demonstrates greater robustness with NDVI than with other vegetation indices that are more susceptible to topographic variation. Ultimately, this framework presents a promising pathway to mitigate biases in most gap-fraction-based PAI measurements, thereby enhancing the accuracy of vegetation structural assessments and supporting broader ecological and climate-related applications. Full article

Monitoring surface water quality is essential for assessing water resources and identifying their quality patterns. Traditional monitoring methods, based on conventional point-sampling stations, are reliable but costly and limited in frequency and spatial coverage. These constraints hinder the ability to evaluate water quality parameters at the temporal and spatial scales required to detect the effects of extreme events on aquatic systems. Satellite imagery offers a viable complementary alternative to enhance the temporal and spatial monitoring scales of traditional assessment methods. However, limitations related to spectral, spatial, temporal, and/or radiometric resolution still pose significant challenges to prediction accuracy. This study aimed to propose a methodology for predicting optically active and inactive water quality parameters in lotic and lentic environments using remote-sensing data and machine-learning techniques. Three remote-sensing datasets were organized and evaluated: (i) data extracted from Sentinel-2 imagery; (ii) data obtained from raw PlanetScope (PS) imagery; and (iii) data from PS imagery normalized using the methodology developed by Dias. Data on water quality parameters were collected from 24 monitoring stations located along the Paraopeba River channel and the Três Marias Reservoir, covering the period from 2016 to 2023. Four machine-learning algorithms were applied to predict water quality parameters: Random Forest, k-Nearest Neighbors, Support Vector Machines with Radial Basis Function Kernel, and Cubist. Model performance was evaluated using four statistical metrics: root-mean-square error, mean absolute error, Lin′s concordance correlation coefficient, and the coefficient of determination. Models based on normalized PS data achieved the best performance in parameter estimation. Additionally, decision-tree-based algorithms showed superior generalization capability, outperforming the other models tested. The proposed methodology proved suitable for this type of analysis, confirming not only the applicability of PS data but also providing relevant insights for its use in diverse environmental-monitoring applications. Full article

The Kubuqi Desert represents a key ecologically fragile region in northern China, primarily functioning as a windbreak and sand-fixation barrier while also contributing to regional ecological balance. However, the area’s ecological vulnerability is pronounced, and investigating the spatiotemporal dynamics of vegetation carbon storage and associated driving mechanisms is essential for the scientific formulation of ecological restoration strategies. This research incorporates multi-source remote-sensing datasets (including Landsat 8 OLI/TIRS Level 2, Sentinel-1 Synthetic Aperture Radar (SAR), ERA5 daily meteorological data, GEDI Level 4B, SRTM GL1 v003, and ESA WorldCover v100) based on the Google Earth Engine (GEE) platform, and employs multiple machine-learning algorithms (validation metrics of the machine learning model: $R^2$ = 0.917, RMSE = 0.251) to develop a dynamic monitoring model of vegetation carbon storage in the Kubuqi Desert during the period 2019–2023. The analysis systematically evaluates the influence of climatic variables and anthropogenic activities on the spatiotemporal differentiation of carbon storage. The results indicate a slight upward trend in overall carbon storage across the study area (average annual increase of 0.4%), with high values predominantly concentrated in vegetated regions (up to 5.22 Mg/Ha) and low values distributed in bare lands and desert zones (0.5–0.7 Mg/Ha). Altitude, temperature, and slope serve as the primary driving factors governing carbon-storage variability. The findings suggest that scientifically guided vegetation restoration and optimized water-resource management can enhance the carbon-sink capacity of the Kubuqi Desert, offering a robust scientific basis for ecological governance and carbon budget assessment in arid and semi-arid desert ecosystems. Full article

Grassland degradation threatens ecosystem function and livelihoods, especially in alpine regions where ecosystems are highly sensitive to disturbance. To compare the effectiveness of common restoration measures at fine spatial scales, we examined four household-level practices in the Ruoergai alpine grassland: year-round grazing exclusion (GE), seeding with grazing exclusion (SGE), seasonal grazing rest (GR), and balancing grazing capacity (BG). Using Sentinel-2 remote sensing data, we monitored vegetation dynamics (NDVI, EVI2, and NIRv) and applied a Propensity Score Matching–Difference-in-Differences (PSM–DID) framework, which constructs comparable control areas without any restoration measures and evaluates whether treatment sites experienced greater pre-to-post restoration changes than their matched controls, thereby strengthening causal inference. All four measures produced statistically significant pre-to-post increases in vegetation indices relative to their matched controls, with GE and SGE showing the largest DID-estimated effects. However, these DID-estimated gains did not persist beyond the implementation year, and in some cases (e.g., SGE, BG), the vegetation indices in treated areas fell below those of the controls, indicating limited persistence. GR and BG yielded smaller DID-estimated effects, reflecting the potential influence of socioeconomic incentives and regulatory challenges on restoration outcomes. These findings highlight the need for sustained management and incentive-aligned policies to maintain restoration benefits in alpine grasslands. Full article

Artisanal and Small-scale Mining (ASM) significantly impacts the Democratic Republic of Congo’s (DRC) socio-economic landscape and environmental integrity, yet its dynamic and informal nature makes monitoring challenging. This study addresses this challenge by implementing a novel deep learning approach to map ASM sites across the DRC using satellite imagery. We tackled key obstacles including ground truth data scarcity, insufficient spatial resolution of conventional satellite sensors, and persistent cloud cover in the region. We developed a methodology to generate a pseudo-ground truth dataset by converting point-based ASM locations to segmented areas through a multi-stage process involving clustering, auxiliary dataset masking, and manual refinement. Four model configurations were evaluated: Planet-NICFI standalone, Sentinel-1 standalone, Early Fusion, and Late Fusion approaches. The Late Fusion model, which integrated high-resolution Planet-NICFI optical imagery (4.77 m resolution) with Sentinel-1 SAR data, achieved the highest performance with an average precision of 71%, recall of 75%, and F1-score of 73% for ASM detection. This superior performance demonstrated how SAR data’s textural features complemented optical data’s spectral information, particularly improving discrimination between ASM sites and water bodies—a common source of misclassification in optical-only approaches. We deployed the optimized model to map ASM extent in the Mwenga territory, achieving an overall accuracy of 88.4% when validated against high-resolution reference imagery. Despite these achievements, challenges persist in distinguishing ASM sites from built-up areas, suggesting avenues for future research through multi-class approaches. This study advances the domain of ASM mapping by offering methodologies that enhance remote sensing capabilities in ASM-impacted regions, providing valuable tools for monitoring, regulation, and environmental management. Full article

Zoological gardens represent unique sites for vector and vector-borne disease studies. They offer suitable breeding habitats for vector development and a diverse range of vertebrate hosts for blood feeding of insect vectors. This study aimed to assess the prevalence of avian blood parasites (Haemosporida, Trypanosomatida) in wild-caught mosquitoes (Culicidae) and Culicoides biting midges (Ceratopogonidae) from the largest and oldest zoo in Lithuania. Insects were collected in May–August 2023 using UV-light, CDC and BG-Sentinel traps; collected material was analysed using both microscopy and PCR-based methods for parasite detection. Overall, 504 parous biting midges (10 species) and 59 mosquitoes (three species) were investigated. Haemosporidians (Haemoproteus minutus (hTURDUS2), H. homogeneae (hSYAT16), and H. asymmetricus (hTUPHI01)) were identified in 5.4% of the 174 tested biting midges. Haemoproteus asymmetricus hTUPHI01 sporozoites were seen in only one individual of Culicoides kibunensis. Of 108 Culicoides females, 3.7% carried trypanosomatids—parasites infecting birds (Trypanosoma bennetti group) and mammals (T. theileri group). Among the 59 tested mosquitoes, two (3.4%) Cx. pipiens/torrentium mosquitoes were found to be PCR-positive for trypanosomatids (T. culicavium and Crithidia brevicula). No haemosporidian parasite DNA was detected in the mosquitoes examined. This pilot study indicates that avian blood parasites circulate within the Lithuanian Zoo, highlighting the need for further research on transmission pathways, vector–host interactions, and potential risks. Full article

Precise lagoon bathymetry remains scarcely available for most tropical islands despite its importance for navigation, resource assessment, spatial planning, and numerical hydrodynamic modeling. Hydrodynamic models are increasingly used for instance to understand the ecological connectivity between marine populations of interest. Island remoteness and shallow waters complicate in situ bathymetric surveys, which are substantially costly. A multisource strategy using historical point sounding, multibeam surveys and well calibrated satellite-derived bathymetry (SDB) can offer the possibility to map entirely extensive and geomorphologically complex lagoons. The process is illustrated here for the rugose complex lagoon of Gambier Islands in French Polynesia. The targeted bathymetry product was designed to be used in priority for numerical larval dispersal modeling at 100 m spatial resolution. Spatial gaps in in situ data were filed with Sentinel-2 satellite images processed with the Iterative Multi-Band Ratio method that provided an accurate bathymetric model (1.42 m Mean Absolute Error in the 0–15 m depth range). Processing was optimized here, considering the specifications and the constraints related to the targeted hydrodynamic modeling application. In the near future, a similar product, possibly at higher spatial resolution, could improve spatial planning zoning scenarios and resource-restocking programs. For tropical island countries and for French Polynesia, in particular, the needs for lagoon hydrodynamic models remain high and solutions could benefit from such multisource coverage to fill the bathymetry gaps. Full article

The 2025 large-scale forest fire in North Gyeongsang Province (Gyeongbuk) caused habitat fragmentation and disrupted ecological networks. This study quantitatively assessed both structural and functional connectivity loss and derived scientifically grounded restoration priorities. Fire intensity was assessed using Sentinel-2-based dNBR, and connectivity changes before and after the fire were analyzed by integrating MSPA (Morphological Spatial Pattern Analysis) and Omniscape (circuit theory-based model). MSPA captured extreme fragmentation, showing an 84% reduction in core habitats and a 976% increase in isolated patches, but failed to reflect functional movement flows. Omniscape approximated this using circuit theory, quantifying a 60% loss in cumulative current flow within the fire boundary and confirming that structural disconnection led to functional connectivity collapse. The restoration priority assessment (53 patches), based on source–sink theory, identified 14 high-priority patches (66% of total area). These patches were characterized by their adjacency to undamaged external cores, which serve as potential population sources for post-restoration recolonization. Notably, the top-priority areas were identified as key connection points within the national ecological corridor where Juwangsan National Park, the Nakdong Ridge, and Grade 1 Ecological Natural Areas overlap. This study demonstrated that integrating MSPA with Omniscape can simultaneously quantify both morphological fragmentation and functional disconnection caused by forest fires. This framework suggests that restoration planning should consider connectivity with broader ecological networks, in addition to recovering lost habitat area. Full article

The mineral deposits in Namibia’s Erongo region are renowned and frequently associated with complex geological environments, including calcrete-hosted paleochannels and hydrothermal alteration zones. Mineral extraction is hindered by high operational costs, restricted accessibility and stringent environmental regulations. To address these challenges, this study proposes an integrated approach that combines satellite remote sensing and machine learning to map and identify mineralisation-indicative zones. Sentinel 2 Multispectral Instrument (MSI) and Landsat 8 Operational Land Imager (OLI) multispectral data were employed due to their global coverage, spectral fidelity and suitability for geological investigations. Normalized Difference Vegetation Index (NDVI) masking was applied to minimise vegetation interference. Spectral indices—the Clay Index, Carbonate Index, Iron Oxide Index and Ferrous Iron Index—were developed and enhanced using false-colour composites. Principal Component Analysis (PCA) was used to reduce redundancy and extract significant spectral patterns. Supervised classification was performed using Support Vector Machine (SVM), Random Forest (RF) and Maximum Likelihood Classification (MLC), with validation through confusion matrices and metrics such as Overall Accuracy, User’s Accuracy, Producer’s Accuracy and the Kappa coefficient. The results showed that RF achieved the highest accuracy on Landsat 8 and MLC outperformed others on Sentinel 2, while SVM showed balanced performance. Sentinel 2’s higher spatial resolution enabled improved delineation of alteration zones. This approach supports efficient and low-impact mineral prospecting in remote environments. Full article

Background and Clinical Significance: Splenic artery pseudoaneurysm (SAP) is a rare but life-threatening complication of chronic pancreatitis. Although endovascular embolization achieves high technical success, recurrence and delayed rupture may occur, particularly in patients with ongoing pancreatic inflammation or alcohol use disorder (AUD). Case Presentation: A 47-year-old woman with alcohol-associated chronic pancreatitis presented with hematochezia, melena, and syncope. CT angiography revealed a 3.6 cm SAP adjacent to a 4.2 cm pancreatic head pseudocyst, and she underwent successful coil embolization. Despite initial stability, she relapsed into heavy alcohol use, experienced recurrent pancreatitis flares, and developed progressive multisystem comorbidities. Surveillance imaging up to three months post-embolization showed pseudocyst fluctuations without early recanalization, but long-term follow-up lapsed. Eight months after embolization, she presented in hemorrhagic shock from recurrent SAP rupture and died despite massive transfusion and emergent splenic artery ligation. Conclusions: Fatal SAP rupture may occur months after technically successful embolization. Sentinel bleeding, AUD relapse, and progressive systemic decline are critical warning signs. Structured post-embolization imaging and multidisciplinary management are essential to improve long-term outcomes. Full article

Precision agriculture will continue to heavily rely on data-driven models to enable more intensive crop monitoring and data-driven decisions. The available remote sensing techniques, particularly those based on multispectral Sentinel-2 data, still have major shortcomings due to cloud cover, low temporal resolution, and time lags in data availability. To address these shortcomings, this paper proposes a hybrid approach that combines Sentinel-2 satellite data with real-time data generated by low-cost ground-based single-pixel detectors (SPDs), such as the AS7263. This hybrid approach addresses key shortcomings in existing agricultural monitoring systems and offers a cost-effective, scalable solution for real-time monitoring and prediction of end-of-season yield, moisture, and plant height using simple PLRS models implemented directly in SPDs with an energy-efficient algorithm for deployment on the STM32G030 microcontroller. Full article

Freshwater cladocerans such as Daphnia magna (D. magna) are keystone grazers whose hormone-regulated life history traits make them sensitive sentinels of endocrine-disrupting chemicals (EDCs). The organophosphate flame-retardant tris(2-butoxyethyl) phosphate (TBOEP) and perfluoroethylcyclohexane sulfonate (PFECHS) now co-occur at ng L⁻¹–µg L⁻¹ in surface waters, yet their chronic sub-lethal impacts on invertebrate endocrine networks remain unclear. We analysed two publicly available 21-day microarray datasets (TBOEP: GSE55132; PFECHS: GSE75607) using gene ontology enrichment, STRING protein interaction networks, Drosophila phenotype mapping, and KEGG (Kyoto Encyclopaedia of Genes and Genomes)-anchored frameworks to build putative adverse outcome pathways (AOPs) for D. magna. Differentially expressed genes were clustered into functional modules and hub nodes were ranked by degree and betweenness. TBOEP suppressed moulting and growth, altering 1157 genes enriched for metabolism and membrane processes; hubs VRK1, MIB2, and adenylosuccinate synthetase formed a muscle anatomical development sub-network. PFECHS down-regulated vitellogenin and shifted 879 genes dominated by oxidative-stress and glutathione-metabolism signatures; central nodes UBC9, eIF4A-III, Tra-2α, and HDAC1 linked meiotic-cycle, oogenesis, and cyclic-compound binding. Despite chemical dissimilarity, both compounds converged on Wnt-signalling nodes—TBOEP via presenilin-1, and PFECHS via CK1ε/CK2—thereby reducing TCF/LEF-dependent transcription. Predicted outcomes include impaired oocyte maturation, reduced fecundity, and stunted body size, consistent with observed decreases in length and vitellogenin protein. Our network analysis, based on high-dose, sub-lethal exposures used in the underlying microarray studies, indicates that TBOEP- and PFECHS-induced perturbations can destabilise endocrine, developmental, and metabolic pathways in D. magna without overt lethality, and highlights Wnt-centred key events and hub genes as candidate biomarkers to be evaluated in future low-dose studies that use environmentally realistic exposure scenarios. Hub genes and Wnt-mediated key events emerge as sensitive biomarkers for monitoring mixed EDC exposure. Full article