Search Results (137)

Early prediction of common bean (Phaseolus vulgaris L.) yield is essential for improving productivity in tropical agricultural systems. In this study, we integrated canopy structural metrics obtained with the Tracing Radiation and Architecture of Canopies (TRAC) system, unmanned aerial vehicle (UAV)-based multispectral measurements (normalized difference vegetation index—NDVI, projected canopy area), and phenological variables collected from stages R6 to R8 under non-limiting nitrogen conditions. Exploratory analyses (correlation, variance inflation factors—VIF), dimensionality reduction (principal component analysis—PCA), and regularized regression (Elastic Net/LASSO), combined with bootstrap stability selection, were applied to identify a parsimonious subset of robust predictors. The final model, composed of six variables, explained approximately 72% of the variability in plant-level grain yield, with acceptable errors (RMSE ≈ 10.67 g; MAE ≈ 7.91 g). The results demonstrate that combining early vigor, radiation interception, and canopy architecture provides complementary information beyond simple spectral indices. This non-destructive framework delivers an efficient model for early yield estimation and supports site-specific management decisions in common bean with high spatial resolution. By enhancing input-use efficiency and reducing waste, this approach contributes to sustainable development and aligns with the global Sustainable Development Goals (SDGs) for climate-resilient agriculture. Full article

Plant growth-promoting bacteria (PGPB) associated with Coffea arabica L. and Coffea canephora Pierre ex Froehner offer a viable strategy to reduce synthetic inputs and enhance resilience in coffee agroecosystems. This review synthesizes evidence from the past decade on rhizosphere-associated and endophytic taxa, their plant growth-promotion and biocontrol mechanisms and the resulting agronomic outcomes. A compartment-specific core microbiome is reported, in the rhizosphere of both hosts, in which Bacillus and Pseudomonas consistently dominate. Within endophytic communities, Bacillus predominates across tissues (roots, leaves and seeds), whereas accompanying genera are host- and tissue-specific. In C. arabica, endophytes frequently include Pseudomonas in roots and leaves. In C. canephora, root endophytes recurrently include Burkholderia, Kitasatospora and Rahnella, while seed endophytes are enriched for Curtobacterium. Functionally, coffee-associated PGPB solubilize phosphate; fix atmospheric nitrogen via biological nitrogen fixation; produce auxins; synthesize siderophores; and express 1-aminocyclopropane-1-carboxylate deaminase. Indirect benefits include the production of antifungal and nematicidal metabolites, secretion of hydrolytic enzymes and elicitation of induced systemic resistance. Under greenhouse conditions, inoculation with PGPB commonly improves germination, shoot and root biomass, nutrient uptake and tolerance to drought or nutrient limitation. Notable biocontrol activity against fungal phytopathogens and plant-parasitic nematodes has also been documented. Key priorities for translation to practice should include (i) multi-site, multi-season field trials to quantify performance, persistence and economic returns; (ii) strain-resolved omics to link taxa to functions expressed within the plant host; (iii) improved bioformulations compatible with farm management and (iv) rationally designed consortia aligned with production goals and biosafety frameworks. Full article

Air pollution forecasting remains a critical challenge for urban public health management, with traditional approaches struggling to balance accuracy and interpretability. This study introduces a novel PM₁₀ forecasting framework combining physics-informed feature engineering with interpretable ensemble fusion using the Choquet integral, the first application of this non-linear aggregation operator for air quality forecasting. Using hourly data from 11 monitoring stations in Budapest (2021–2023), we developed four specialized feature sets capturing distinct atmospheric processes: short-term dynamics, long-term patterns, meteorological drivers, and anomaly detection. We evaluated machine learning models including Random Forest variants (RF), Gradient Boosting (GBR), Support Vector Regression (SVR), K-Nearest Neighbors (KNN), and Long Short-Term Memory (LSTM) architectures across six identified pollution regimes. Results revealed the critical importance of feature engineering over architectural complexity. While sophisticated models failed when trained on raw data, the KNN model with 5-dimensional anomaly features achieved exceptional performance, representing an 86.7% improvement over direct meteorological input models. Regime-specific modeling proved essential, with GBR-Regime outperforming GBR-Stable by a remarkable effect size. For ensemble fusion, we compared the novel Choquet integral approach against conventional methods (mean, median, Bayesian Model Averaging, stacking). The Choquet integral achieved near-equivalent performance to state-of-the-art stacking while providing complete mathematical interpretability through interaction coefficients. Analysis revealed predominantly redundant interactions among models, demonstrating that sophisticated fusion must prevent information over-counting rather than merely combining predictions. Station-specific interaction patterns showed selective synergy exploitation at complex urban locations while maintaining redundancy management at simpler sites. This work establishes that combining domain-informed feature engineering with interpretable Choquet integral aggregation can match black-box ensemble performance while maintaining the transparency essential for operational deployment and regulatory compliance in air quality management systems. Full article

Reliable differentiation of crops and weeds is essential for precision agriculture, where real-time detection can minimize chemical inputs and support site-specific interventions. This study presents the large-scale and systematic benchmark of 19 YOLO-family variants, spanning YOLOv3 through YOLOv11, for cotton–weed detection using the Cotton–8 dataset. The dataset comprises 4440 annotated field images with five categories: broadleaf weeds, grass weeds, and three growth stages of cotton. All models were trained under a standardized protocol with COCO-pretrained weights, fixed seeds, and Ultralytics implementations to ensure reproducibility and fairness. Inference was conducted with a confidence threshold of 0.25 and a non-maximum suppression (NMS) IoU threshold of 0.45, with test-time augmentation (TTA) disabled. Evaluation employed precision, recall, mAP@0.5, and mAP@0.5:0.95, along with inference latency and parameter counts to capture accuracy–efficiency trade-offs. Results show that larger models, such as YOLO11x, achieved the best detection accuracy (mAP@0.5 = 81.5%), whereas lightweight models like YOLOv8n and YOLOv9t offered the fastest inference ( 27 msper image) but with reduced accuracy. Across classes, cotton growth stages were detected reliably, but broadleaf and grass weeds remained challenging, especially under stricter localization thresholds. These findings highlight that the key bottleneck lies in small-object detection and precise localization rather than architectural design. By providing the first direct comparison across successive YOLO generations for weed detection in cotton, this work offers a practical reference for researchers and practitioners selecting models for real-world, resource-constrained cotton–weed management. Full article

This study collected 427 cultivated topsoil samples from the Mohe watershed in Tangcheng County, eastern China. By integrating positive matrix factorization (PMF) for quantitative source apportionment with self-organizing maps (SOMs) for spatial clustering, we effectively identified pollution factors and conducted a systematic evaluation of pollution sources, environmental capacity, and health risks. The results show that: (1) the soils were slightly acidic and enriched in Cd, Cr, Cu, Hg, and Pb, with Cd and Hg showing high spatial variability linked to anthropogenic inputs. (2) Quantitative source apportionment indicated that 25.9% of heavy metals (As, Cr, Ni, Pb) originated mainly from natural pedogenic sources, while agricultural activities contributed 20.8% (Cd) and 42.8% (Cu, Zn). Hg (10.5%) enrichment was attributed to residential coal combustion and wind patterns, demonstrating source-specific anthropogenic influences. (3) The environmental capacity assessment indicated a moderate capacity level across the study area. However, the improved index (PI_min) revealed overload phenomena at localized sites, and these overloaded areas exhibited high spatial consistency with the distributions of agricultural and mixed sources. (4) Health risk evaluation indicated that hand-to-mouth ingestion was the dominant exposure pathway, with children facing significantly higher risks than adults. Non-carcinogenic risks remained within safe limits, but carcinogenic risks were non-negligible, with 86.7% of sites exceeding the threshold for children, especially in cultivated lands and riverbank villages. Findings underscore the importance of addressing synergistic effects of natural and agricultural sources in watershed management and prioritizing children’s health protection. Full article

Agricultural management in Europe faces increasing pressure to reduce its environmental footprint. Implementing precision agriculture for weed management could offer a solution and minimize the use of chemical products. High spatial resolution imagery from real time kinematic (RTK) unmanned aerial vehicles (UAV) in combination with supervised convolutional neural network (CNNs) models have proven successful in making location specific treatments. This site-specific advice limits the amount of herbicide applied to the field to areas that require action, thereby reducing the environmental impact and inputs for the farmer. To develop performant CNN models, there is a need for sufficient high-quality labelled data. To reduce the labelling effort and time, a new labelling method is proposed whereby image subsection pairs are labelled based on their relative differences in weed pressure to train a CNN ordinal regression model. The model is evaluated on detecting weed pressure in potato (Solanum tuberosum L.). Model performance was evaluated on different levels: pairwise accuracy, linearity (Pearson correlation coefficient), rank consistency (Spearman’s (r_s) and Kendal (τ) rank correlations coefficients) and binary accuracy. After hyperparameter tuning, a pairwise accuracy of 85.2%, significant linearity (r_s = 0.81) and significant rank consistency (r_s = 0.87 and τ = 0.69) were found. This suggests that the model is capable of correctly detecting the gradient in weed pressure for the dataset. A maximum binary accuracy and F1-score of 92% and 88% were found for the dataset after thresholding the predicted weed scores into weed versus non-weed images. The model architecture allows us to visualize the intermediate features of the last convolutional block. This allows data analysts to better evaluate if the model “sees” the features of interest (in this case weeds). The results indicate the potential of ordinal regression with relative labels as a fast, lightweight model that predicts weed pressure gradients. Experts have the freedom to decide which threshold value(s) can be used on predicted weed scores depending on the weed, crop and treatment that they want to use for flexible weed control management. Full article

Precision agriculture technologies (PATs) are revolutionizing the agricultural sector by minimizing the reliance on plant protection products (PPPs) in crop management. This approach integrates a broad range of advanced solutions employed to help farmers in optimizing PPP application, while minimizing input and maintaining effective crop protection. These technologies include sensors, drones, robotics, variable rate systems, and artificial intelligence (AI) tools that support site-specific pesticide applications. The objective of this review was to perform a bibliometric analysis to identify scientific trends and gaps in this field. The analysis was conducted using Scopus and Web of Science databases for the period of 2015–2024, by applying a data filtering process to ensure a clean and reliable dataset. The methodology involved citation, co-authorship, co-citation, and co-occurrence analysis. VOSviewer software (version 1.6.20) was used to generate maps and assess global research developments. Results identified AI, sensor, and data processing categories as the most central and interconnected scientific topics, emphasizing their vital role in the evolution of precision spraying technology. Bibliometric analysis highlighted that China, the United States, and India were the most productive countries, with strong collaborations within Europe. The co-occurrence and co-citation analyses revealed increasing interdisciplinarity and the integration of AI tools across various technologies. These findings help identify key experts and research leaders in the precision agriculture domain, thus underscoring the shift toward a more sustainable, data-driven, and synergistic approach in crop protection. Full article

Cultivated land quality is critical for soil productivity and scientific fertilization. This study analyzed its evolution and impact on soil productivity across three ecological regions (southern, central, and northern Shanxi) in Shanxi Province, China, from 1998 to 2021). Using data from 8 long-term experimental sites (1998–2021) and 50 monitoring stations (2016–2021), we employed random forest analysis to evaluate temporal trends in key soil indicators. The results show the following: (1) Northern Shanxi exhibited the greatest improvement in soil fertility, with organic matter increasing by 98.2%, total nitrogen by 57.2%, available phosphorus by 131.7%, and available potassium by 17.1%. (2) Nitrogen fertilizer application increased across all regions, while phosphorus and potassium inputs generally declined. (3) Crop yields improved substantially—southern Shanxi wheat and maize increased by 15.3% and 20.9%, respectively, while central and northern Shanxi maize yields rose by 30.9% and 75.4%. Random forest models identified regional characteristics (40%), nitrogen fertilization (20%), and available phosphorus (18%) as primary influencing factors. Although cultivated land quality improved overall, soil fertility remained medium to low. Region-specific management strategies are recommended: rational nitrogen use in all regions; nitrogen control with phosphorus supplementation in the south; focused improvement of available phosphorus and potassium in the center; and increased organic fertilizer in the north. These measures support scientific nutrient management and sustainable agricultural production. Full article

Weed detection is critical for precision agriculture, enabling targeted herbicide application to reduce costs and enhance crop health. This study utilized UAV-acquired RGB imagery from cotton fields to develop and evaluate deep learning models for weed detection. As sustainable resource management gains importance in rainfed agricultural systems, precise weed identification is essential to optimize yields and minimize herbicide use. However, distinguishing weeds from crops in complex field environments remains challenging due to their visual similarity. This research employed YOLOv7, YOLOv7-w6, and YOLOv7-x models to detect and classify weeds in cotton fields, using a dataset of 9249 images collected under real field conditions. To improve model performance, we enhanced the annotation process using LabelImg and Roboflow, ensuring accurate separation of weeds and cotton plants. Additionally, we fine-tuned key hyperparameters, including batch size, epochs, and input resolution, to optimize detection performance. YOLOv7, achieving the highest estimated accuracy at 83%, demonstrated superior weed detection sensitivity, particularly in cluttered field conditions, while YOLOv7-x with accuracy at 77% offered balanced performance across both cotton and weed classes. YOLOv7-w6 with accuracy at 63% faced difficulties in distinguishing features in shaded or cluttered soil regions. These findings highlight the potential of UAV-based deep learning approaches to support site-specific weed management in cotton fields, providing an efficient, environmentally friendly approach to weed management. Full article

Through highly detailed data acquisition, a precision agriculture approach leads to the optimization of inputs, improving, for instance, water and nutrient use efficiency. High-resolution vigor mapping in perennial orchards provides the spatial detail required to achieve such targeted management. This exploratory case study characterizes the spatial variability of vegetative vigor in a young SHD almond orchard in southern Sardinia by integrating high-resolution unmanned aerial vehicle (UAV) imagery and Normalized Difference Vegetation Index (NDVI) mapping with two consecutive seasons of ground measurements; the NDVI raster was subsequently used to delineate three distinct vigor zones. The NDVI was selected as a reference index because of its well-assessed performance in field-variability studies. Field measurements, during the kernel-filling period, included physiological assessments (stem water potential (Ψ_stem), SPAD, photosynthetic rates), morphological evaluations, soil properties, yield, and quality analyses. High vigor zones exhibited better physiological conditions (Ψ_stem = −1.60 MPa in 2023, SPAD = 38.77 in 2022), and greater photosynthetic rates (15.31 μmol CO₂ m⁻² s⁻¹ in 2023), alongside more favorable soil conditions. Medium vigor zones showed intermediate characteristics, and balanced soil textures, producing a higher number of smaller almonds. Low vigor zones exhibited the poorest performance, including the most negative water status (Ψ_stem of −1.94 MPa in 2023), lower SPAD values (30.67 in 2023), and coarse-textured soils, leading to reduced yields. By combining UAV-based NDVI mapping with ground measurements, these results highlight the value of precision agriculture in intra-field variability identification, providing a basis for future studies that will test site-specific management strategies in SHD orchards. Full article

The present study explored the seasonal dynamics of spat settlement and growth of the Mediterranean mussel (Mytilus galloprovincialis) in the semi-enclosed and eutrophic Maliakos Gulf (Central Aegean, Greece), a coastal system within the Natura 2000 network (GR 2440002, Natura 2000). Spat collectors were deployed at three mussel farms representing different locations in the gulf (north, south, and inner west) and at multiple depths over a year. The results revealed a clear reproductive cycle, with spawning initiated in early January and spat settlement occurring from March to June. Settlement intensity was highest in shallower waters during the beginning of the season (March) and in the end (June), while depth had no significant effect mid-season. Mussel size and weight varied significantly with season and location, with the largest individuals observed during spring and early summer at the north and south sites. Environmental monitoring depicted strong nitrogen enrichment and phosphorus limitation, driven by inputs from the Spercheios River and surrounding agricultural activities. During winter, elevated chlorophyll-a concentrations likely supported early larval development, while nutrient imbalances threaten long-term ecosystem stability. These findings underscore the importance of area- and season-specific management of spat collectors and call for integrated monitoring and regulation of nutrient inputs to safeguard the ecological integrity of the gulf and ensure sustainable mussel farming. Full article

Optimized water-saving irrigation (WSI) practices are critical for enhancing resource use efficiency and ensuring sustainable wheat production in water-scarce regions. This meta-analysis quantitatively assessed the effects of various WSI methods on wheat yield, water use efficiency (WUE), and partial factor productivity of nitrogen (PFPN) across China’s wheat regions. The results showed that optimized irrigation, particularly drip and micro-sprinkler systems, significantly reduced irrigation water and nitrogen inputs by 35.1% and 7.2%, respectively, without yield penalties. Drip and micro-sprinkler irrigation, which together accounted for over 97% of observations, improved WUE by 18.7% and 10.1%, respectively, and increased PFPN by 6.8% and 5.5%, highlighting their dominant role in current WSI practices. Moderate deficit irrigation (60–100% of full irrigation) optimized WUE and PFPN while maintaining stable yields, whereas severe deficit irrigation (<40%) caused substantial yield losses. Soil texture and bulk density strongly modulated WSI effectiveness. Climatic factors, particularly growing season precipitation, negatively correlated with WSI benefits, highlighting enhanced efficiency gains under drier conditions. These findings emphasize the need to prioritize drip and micro-sprinkler irrigation in national water-saving strategies and advocate for integrated approaches combining WSI with soil health management and site-specific irrigation scheduling to promote sustainable wheat intensification under variable agroecological conditions. Full article

Understanding spatial variability in crop yields across fields is critical for developing precision agricultural strategies that optimize productivity while reducing negative environmental impacts. This variability often arises from a complex interplay of topographic features, soil characteristics, and hydrological conditions. This study investigates the influence of hydro-topographic factors on corn and soybean yield variability from 2016 to 2023 at the well-managed experimental sites in Beltsville, Maryland. A high-resolution surface digital elevation model (DEM) and subsurface DEM derived from ground-penetrating radar (GPR) were used to quantify topographic factors (elevation, slope, and aspect) and hydrological factors (surface flow accumulation, depth from the surface to the subsurface-restricting layer, and distance from each crop pixel to the nearest subsurface flow pathway). Topographic variables alone explained yield variation, with a relative root mean square error (RRMSE) of 23.7% (r² = 0.38). Adding hydrological variables reduced the error to 15.3% (r² = 0.73), and further combining with remote sensing data improved the explanatory power to an RRMSE of 10.0% (r² = 0.87). Notably, even without subsurface data, incorporating surface-derived flow accumulation reduced the RRMSE to 18.4% (r² = 0.62), which is especially important for large-scale cropland applications where subsurface data are often unavailable. Annual spatial yield variation maps were generated using hydro-topographic variables, enabling the identification of long-term persistent yield regions (LTRs), which served as stable references to reduce spatial anomalies and enhance model robustness. In addition, by combining remote sensing data with interannual meteorological variables, prediction models were evaluated with and without hydro-topographic inputs. The inclusion of hydro-topographic variables improved spatial characterization and enhanced prediction accuracy, reducing error by an average of 4.5% across multiple model combinations. These findings highlight the critical role of hydro-topography in explaining spatial yield variation for corn and soybean and support the development of precise, site-specific management strategies to enhance productivity and resource efficiency. Full article

The black soil region of Northeast China is crucial for agricultural productivity. Ammonia-oxidizing archaea (AOA) are key indicators of soil nitrification in this region, yet it remains unclear whether this process is driven by the entire community or by specific clusters. Here, we investigated the AOA community across a long-term fertilization Brown Soil Experimental Station and 15 sites in the Typical Black Soil Zone. Using Illumina MiSeq sequencing of the AOA amoA gene and cluster-specific primers, 14 OTUs were selected as core clusters based on relative abundance >0.1% and strong correlations (r > 0.7) with soil properties or PNR, and were further grouped into five distinct clusters according to phylogenetic analysis. Compared to the overall AOA community, core clusters responded more precisely to fertilization, straw addition, and spatial variation, with contrasting environmental responses reflected in their relationships with soil nitrification dynamics. Clusters G1 and G2 had positive correlations with soil PNR, while Clusters G4 and G5 had negative correlations. Moreover, AOA core clusters demonstrated stronger correlations with soil properties, including pH, C/N ratio, and NH₄⁺/NO₃⁻ ratio. These findings demonstrate that AOA core clusters are reliable microbial indicators of soil nitrification, and monitoring their abundance changes under nitrogen input can provide early insights into potential inhibition, informing predictive models and guiding more precise nitrogen management to support sustainable agricultural practices. Full article

Freshwater ecosystems are increasingly stressed by drought and anthropogenic inputs that can increase specific conductivity (SPC) and pH; however, little is known about how harsher conditions affect fish. We evaluated how fish growth and diet composition changed along a natural gradient in SPC and pH in Wyoming, USA using Northern plains killifish (Fundulus kansae) and Fathead minnows (Pimephales promelas). We surveyed 201 sites where we measured water chemistry, sampled fish, and assessed invertebrate prey availability from May to September 2024. Northern plains killifish and/or Fathead minnows inhabited 12 sites, which were the focus of our study. We measured otoliths to assess growth and stomach contents to estimate dietary selectivity. Growth decreased at higher SPC (486–23,500 µS/cm) for Fathead minnows and pH (7.2–9.0) for both species, suggesting an energy trade-off with osmoregulation. Dietary analyses revealed variable selection for Chironomidae larvae, while other taxa such as Gammaridae and Coleoptera were avoided at higher SPC and pH. Despite the extreme conditions, these fish maintained some dietary preference, highlighting behavioral plasticity. Our findings suggest that while these species can tolerate harsh environments, sublethal effects on growth and diet may limit long-term fitness. This research offers a framework for assessing the viability of fish populations inhabiting ecosystems with increasing salinity and pH that can inform conservation and management strategies under future environmental change. Full article