Search Results (132)

Pasture quality monitoring is essential for optimizing grazing management and reducing the incidence of nutritional and metabolic disorders in ruminants, yet conventional field-based measurements remain labor-intensive and limited in spatial coverage. This review examines how remote sensing (RS) technologies can support pasture-based livestock systems by providing timely, scalable assessments of biomass, botanical composition, and nutritive attributes. Data from multispectral, hyperspectral, radio detection and ranging (RADAR), and light detection and ranging (LiDAR) sensors, acquired via satellite, unmanned aerial vehicle (UAV), and proximal platforms, are combined with machine learning (ML) methods and radiative transfer models to derive pasture biophysical and quality indicators. The reviewed evidence shows that RS reliably estimates pasture biomass and structural traits, while advances in spectral unmixing, data fusion, and artificial intelligence (AI) improve the characterization of heterogeneous swards and support emerging indicators related to forage quality. Integrating these remotely sensed metrics into grassland decision-support frameworks can enhance grazing allocation, inform fertilization and irrigation decisions, and help detect conditions associated with nutritional imbalances. Overall, the synthesis demonstrates that RS, particularly when combined with advanced modelling and cloud-based processing, offers a robust pathway for improving pasture monitoring and strengthening the nutritional management of ruminants, thereby supporting more sustainable and animal welfare-focused grazing systems. Full article

Warm-season precipitation over Sichuan, China, is jointly modulated by complex terrain, monsoon water vapor transport, and local convective activities, leading to significant spatiotemporal heterogeneity. However, radar observations over mountainous areas are frequently impaired by terrain blockage, beam shielding, and insufficient network coverage, which cause missing data and spatial discontinuity, thereby restricting the accurate monitoring of precipitation systems. To alleviate these problems, this study develops an Efficient Multi-Scale Attention (EMA) U-Net model integrated with Digital Elevation Model (DEM) information, termed EMA-U-Net-DEM, to retrieve radar composite reflectivity by utilizing multi-channel observations from the Fengyun-4A (FY-4A) Advanced Geostationary Radiation Imager (AGRI). In the experiments, FY-4A AGRI multi-spectral measurements were used as model inputs, while radar composite reflectivity products from the Severe Weather Automatic Nowcasting (SWAN) system were applied as reference labels. The modeling and validation were carried out using warm-season (June–August) datasets over Sichuan Province. The results indicate that the proposed EMA-U-Net-DEM exhibits better performance than the traditional U-Net and several typical attention-based benchmark models. Quantitatively, the model achieves a root mean square error (RMSE) of 6.728 dBZ, a mean absolute error (MAE) of 4.788 dBZ, a coefficient of determination $R^2$ of 0.656, a peak signal-to-noise ratio (PSNR) of 25.243 dB, and a structural similarity index measure (SSIM) of 0.793. Categorical verification further reveals that the model yields the highest critical success indices (CSI) of 0.850, 0.560, and 0.364 in the reflectivity ranges of 0–25 dBZ, 25–45 dBZ, and 45–70 dBZ, respectively, demonstrating its superior ability in characterizing weak precipitation backgrounds, moderate precipitation structures, and intense convective cores. The performance enhancements are mainly attributed to the strengthened multi-scale feature extraction by the EMA module and the effective topographic constraints introduced by DEM data. This study confirms that the fusion of FY-4A multi-spectral observations and topographic information can effectively improve radar composite reflectivity retrieval over complex terrain, providing a feasible solution for precipitation monitoring, quantitative precipitation estimation, and severe weather nowcasting in mountainous regions with limited radar coverage. Full article

Drought stress severely limits foxtail millet yield and quality, yet current drought-resistance indices are exclusively yield-oriented and ignore grain-filling quality. Our two-year (2024–2025) experiments with 24–48 varieties revealed that yield and blighted grain rate (BGR) are partially decoupled (e.g., Zhangzagu 18: yield 2307 kg/ha, BGR 0.444; Zhonggu 19: yield 1622 kg/ha, BGR 0.280). We therefore constructed the Yield–Quality Synergy Index (YQSI = DYI − BGR), which penalizes varieties with poor grain filling. The YQSI tied for first place with DYI in comprehensive screening performance and achieved the highest inter-annual stability (Spearman ρ = 0.823, Jaccard = 0.438, composite score = 1.261). Sensitivity analysis confirmed robustness of the equal-weight formula across a 4-fold range of quality-penalty weights. Six strongly drought-resistant germplasms with balanced yield and quality were identified. Using UAV multimodal data (RGB, multispectral, and thermal infrared) acquired during grain filling, a Random Forest model predicted a YQSI with overall R² = 0.819 and an F1 score of 0.933 for variety screening. Feature-importance analysis highlighted NDVI, WDRVI, and red-edge texture as key predictors. This study provides a quality-constrained drought-resistance evaluation framework and demonstrates the potential of UAV-based high-throughput phenotyping for foxtail millet breeding. Full article

Collisions with overhead power lines are an important source of avian mortality, particularly for migratory and large-bodied birds. Although warning devices are widely installed to increase line visibility, their image-based visibility across viewing distances and directions remains poorly quantified. We evaluated eight power-line bird-collision warning devices using UAV-based multispectral imaging and an avian spectral-response weighting model. Images were acquired at 5, 10, 20, 30, and 50 m under upward-looking, horizontal, and downward-looking viewing geometries. A composite visibility score quantified target–background separation within the multispectral images. Raw mean visibility increased from 0.351 at 5 m to 0.484 at 50 m; however, the distance effect was not significant after accounting for viewing angle and heterogeneous residual variance. Viewing angle was the dominant source of variation at 20–50 m, with substantially higher visibility under downward-looking views than under horizontal and upward-looking views. A significant distance × viewing-angle interaction further showed that distance-response patterns differed among viewing geometries: visibility increased with distance under downward-looking views but remained consistently low under horizontal and upward-looking views. The relative performance of the tested devices also changed under low-visibility conditions, with the self-luminous tag and circular tag B attaining the highest condition-standardized mean scores. These findings identify viewing geometry as a critical determinant of warning-device visibility and demonstrate that averages pooled across observation conditions can obscure important performance differences. Evaluation and deployment should therefore prioritize device visibility from below and during near-horizontal viewing, particularly for overhead ground or shield wires at medium and long distances. This condition-specific framework provides a quantitative basis for improving the design and placement of power-line bird-collision warning devices. Full article

Predicting plantation establishment-failure prior to planting remains a major operational challenge due to the strong spatial variability in post-harvest environmental conditions. This study developed a spatially explicit modelling framework that integrated pre-planting unmanned aerial vehicle (UAV)-derived environmental, structural, terrain, and operational-treatment data to predict establishment risk across plantation landscapes. Environmental, terrain, vegetation, and structural predictors were derived from pre-planting multispectral UAV imagery, while plantation establishment outcomes were quantified approximately 21 months later using an automated tree-detection and assessment framework. The datasets were integrated within a ridge-regularised logistic regression model incorporating interaction terms, multi-scale predictors, operational treatment masks, and blocked spatial cross-validation. The model achieved strong predictive performance under within-site blocked spatial cross-validation, with moisture-related variables, vegetation condition, and structural metrics contributing most strongly to establishment-failure prediction. Predicted risk surfaces closely matched observed patterns of reduced stocking density and suppressed growth. Beyond predicting establishment-failure, the framework enables plantation managers to screen model-predicted outcomes under alternative treatment encodings before planting and to integrate the composite stocking-density and height response within a spatially explicit Establishment Index. The framework therefore demonstrates that future plantation establishment can be predicted from environmental conditions measured before planting and provides a scalable pathway for translating high-resolution UAV data into operational decision support. Full article

Accurate land cover mapping is a fundamental tool for environmental management and ecosystem monitoring. This study presents a comparative evaluation of four deep learning architectures, namely a Convolutional Neural Network (CNN), UNet++, ConvNeXt and Vision Transformer (ViT), for land cover classification into five primary classes: water, cropland, forest, low/natural vegetation and built-up. The broader Lake Kerkini basin was selected as the primary training and evaluation area. The multispectral input data were generated through Google Earth Engine and consisted of multi-seasonal Sentinel-2 composite mosaics for the 2021 mapping year, covering winter, spring, summer and autumn. To obtain a more reliable performance estimate and mitigate the effects of spatial autocorrelation, a four-fold spatial cross-validation approach was implemented. Under this spatial validation framework, the convolution-based architectures achieved the strongest performance. CNN obtained the highest numerical fold-mean performance, with an overall accuracy of 81.53% and a macro-averaged F1 score (Macro-F1) of 80.09%, followed closely by UNet++ and ConvNeXt. Non-parametric repeated-measures statistical testing indicated a significant overall architecture effect, with CNN, UNet++ and ConvNeXt showing broadly comparable fold-level Macro-F1 performance, while the tested ViT configuration trained from scratch ranked last across all spatial folds. Regional transferability was further evaluated in the nearby independent Lake Doirani region, where the convolutional architectures, particularly CNN and UNet++, showed strong agreement with the WorldCover-derived reference labels without fine-tuning. Finally, feature-importance analysis indicated that specific spectral-seasonal channels, especially the Blue band (B2) in winter and the Short-Wave Infrared band (B12) in summer, were consistently influential in the models’ predictions. Overall, under the tested 2021 Mediterranean case-study conditions, the results highlight the importance of spatially rigorous validation and show that the evaluated convolution-based configurations achieved stronger performance than the tested ViT configuration trained from scratch. Full article

Microplastic monitoring needs methods that operate directly in water with minimal sample handling. Conventional techniques such as infrared and Raman spectroscopy and pyrolysis–GC/MS provide polymer-specific information but require sample preparation and delayed laboratory analysis. We propose an optical sensor concept for real-time, in situ microplastic assessment, based on multispectral pulsed transmission in the visible range using synchronized laser-diode lines and the directly transmitted signal through an active sensor volume. After calibration on particle-free water, each particle event reduces to a water-normalized transmission whose deficit is set by geometrical beam–particle overlap and the wavelength-dependent extinction efficiency. The weak polymer absorption is represented by the Urbach-tail formalism, the refractive-index-related redirection of light by a Fresnel-based, surface- and orientation-averaged probability of direct transmission, and particle size and shape are decoupled through an effective optical length. The coupled nonlinear system is solved for the bounds of the polymer absorption coefficient per candidate geometry. Because each polymer occupies a bounded region in multi-wavelength absorption space fixed by its band gap and structural state, the method can, in principle, separate structural modifications of identical composition, such as low- and high-density polyethylene. This is a sensor concept with a model-based proof of concept, not full environmental validation. Experimental verification on real reference particles is reported separately; the present article establishes the measurement model and inversion scheme that this verification builds on. Full article

Real-time monitoring of cotton above-ground biomass (AGB) is crucial for monitoring crop growth and optimizing management practices. This study evaluated UAV-based RGB and multispectral (MS) imagery for cotton AGB estimation across multiple growth stages under different planting densities and sowing dates in Anyang, China. Spectral features, vegetation indices (VIs), and Gray Level Co-occurrence Matrix (GLCM) texture metrics were extracted and organized into three scenarios: RGB + MS, RGB-only, and MS-only. Recursive feature elimination with cross-validation (RFECV) was applied for feature selection, and six machine learning models were evaluated using both baseline and selected feature sets. Results showed that model performance was strongly influenced by growth stage, sensor configuration, and feature composition. Accuracy was highest at the seedling and squaring stages and decreased at flowering due to canopy complexity and spectral saturation. MS-only and fused features generally performed best at the seedling stage, while RGB-only features were competitive or superior at the squaring stage, highlighting the importance of high-resolution structural information. At flowering, fused RGB–MS features provided the most stable performance, although improvements were limited. RFECV exhibited stage-dependent behavior, improving performance mainly at early growth stages but showing inconsistent benefits later. SHAP analysis revealed a shift from texture-dominated predictors at the seedling stage to balanced feature contributions at squaring and vegetation index (VIs) dominance at flowering. Overall, cotton AGB estimation is a stage-dependent process requiring adaptive sensor and feature selection strategies. Full article

Early-generation sesame (Sesamum indicum L.) breeding requires high-throughput phenotyping of large unreplicated populations across contrasting environments. A DJI Phantom 4 Multispectral UAV was flown at 40, 80, and 120 m above ground level (AGL) over 588 M₂ genotypes under full irrigation (ENV1) and terminal drought (ENV2; irrigation withheld from reproductive onset) on four dates (July–September 2025). Structure-from-motion canopy height models were compared with ground measurements, and four spectral reflectance indices—Normalised Difference Vegetation Index (NDVI), Normalised Difference Red Edge (NDRE), Green Normalised Difference Vegetation Index (GNDVI), and Leaf Chlorophyll Index (LCI)—were derived from 40 m imagery. Ordinary least squares (OLS), Random Forest, and Gradient Boosting were evaluated under leave-one-genotype-out (LOGO), leave-one-environment-out (LOEO), and leave-one-date-out (LODO) cross-validation; genotypic repeatability was quantified by intraclass correlation (ICC), and drought performance was ranked by a composite Stress Tolerance Index (STI) validated against an independent breeder assessment. The 40 m altitude gave the highest height accuracy (R² = 0.812 in ENV1; 0.663 in ENV2). LOGO accuracy (R² ≈ 0.83) fell to R² ≈ 0.55 under LODO—the operationally relevant figure for a new phenological stage—and the full structural–spectral OLS model collapsed (R² = −0.203) where tree ensembles remained stable. Spectral-index repeatability was up to ~2-fold higher under stress (ICC(3,4) > 0.84). The composite STI flagged 38 elite genotypes (7.6% of 498); 10 of its top 30 were confirmed in the breeder’s 48-best selection from all 588 rows—a 4.1-fold enrichment over chance (hypergeometric p = 4.5 × 10⁻⁵). Full article

In recent years, there has been growing interest in applying vision–language models (VLMs) to quantitative remote sensing. This study evaluates whether three commercial VLMs (GPT-4o, GPT-5.5, and Claude Sonnet 4.6) can detect and classify the severity of harmful algal blooms (HABs) from Sentinel-2 satellite imagery of western Lake Erie and compares them against classical machine learning classifiers (Random Forest (RF), Support Vector Machine (SVM), and eXtreme Gradient Boosting (XGBoost)) trained on both a three-band red, green, blue (RGB) composite representation of the imagery and a 10-band multi-spectral reflectance representation. Forty bloom events identified from the National Oceanic and Atmospheric Administration (NOAA) Harmful Algal Bloom Operational Forecast System (HAB-OFS) severity assessments were assembled into the evaluation dataset, spanning seven bloom seasons (2019–2025). For binary bloom detection, the VLMs did not match the classical RGB classifiers; their F1 scores (0.69–0.75) fell below the best RGB classifier (Random Forest, 0.76) and below a trivial always-present baseline (F1 = 0.77), and they carried false positive rates of 73–93% on bloom-absent images, against 27–40% for the RGB classifiers. The VLMs reached high recall by labeling most scenes as bloom-positive, which makes them operationally unreliable in this configuration. For severity classification, the VLMs assigned 60–70% of their predictions to the “moderate” category regardless of actual conditions and identified at most one of the two severe blooms, whereas the classical classifiers tracked the ground-truth distribution and delivered two to nearly three times the exact-match accuracy (0.44–0.59 vs. 0.20–0.225). The strongest method across all metrics was the multi-spectral SVM (F1 = 0.833, false positive rate 27%, accuracy 0.795). Switching the same SVM from RGB to multi-spectral features raised accuracy from 0.675 to 0.795, a 12-percentage-point gain that measures the spectral information carried by red-edge and shortwave infrared bands that are accessible through multi-spectral sensors but unavailable to standard VLM vision encoders. Feature-importance analysis showed that the multi-spectral classifiers ranked chlorophyll-specific indices, the Normalized Difference Chlorophyll Index (NDCI) and the Floating Algae Index (FAI), among their top predictors, the same signatures used in established operational algorithms, while the RGB classifiers relied on red-channel variability and green-dominant pixel fractions because RGB inputs cannot compute those indices. Two compounded limitations therefore constrain off-the-shelf VLMs for aquatic remote sensing: the limited spectral information available through standard RGB channels and a mismatch between the land-dominated training distributions of these models and aquatic optical conditions. Domain-specific classifiers operating on multi-spectral data remain the more suitable tools for continued development of HAB monitoring and water-quality retrieval. Full article

Rapid and accurate landslide mapping is essential for disaster risk reduction and sustainable land management in landslide-prone mountainous regions. This study presents a U-Net semantic segmentation framework for pixel-wise landslide classification in the Laonung Creek Watershed of southern Taiwan using 96 annotated tiles derived from a very high-resolution DEM and SPOT-6 multispectral imagery. An input modality ablation experiment compares four configurations: a fused DEM–optical composite matching the visual input used by the annotators (annotation-coherent input), SPOT-6 natural color imagery, a DEM-derived terrain stack, and a six-channel multi-source stack combining all SPOT-6 bands with slope and curvature. All configurations use an identical EfficientNet-B0 U-Net architecture under a spatially blocked train/validation/test design with a fixed held-out test set of 29 tiles. The multi-source stack achieves the highest test Average Precision (AP) of 0.556 (95% CI: 0.463–0.643), whereas the annotation-coherent fused composite achieves AP = 0.511 (95% CI: 0.404–0.601); overlapping confidence intervals indicate that neither modality is definitively superior at this test-set size. The terrain-only configuration (AP = 0.152) confirms that optical information is essential for reliable delineation. A key methodological finding is that differential encoder–decoder learning rates caused rapid decoder overfitting; matched rates of ${10}^{-5}$ substantially stabilized training and are recommended as a conservative default for small-sample segmentation with pretrained encoders. At matched pixel positions, the best DL model achieves AP comparable to a companion Random Forest (DL: 0.847, RF: 0.824), while producing spatially coherent probability maps that support scalable landslide inventory compilation for sustainable hazard management. Full article

The worldwide spread of semiconductor devices has driven a surge in electronic waste (e-waste), which reached 62 million metric tons in 2022 and is projected to exceed 80 million metric tons by 2030. E-waste contains hazardous substances such as cadmium and mercury, yet also represents a $57 billion annual opportunity through the recovery of valuable and critical raw materials (CRMs). However, formal recycling rates remain stagnant at 22.3%, largely due to limitations of current automated sorting methods. These systems primarily rely on visible-light (RGB) imaging, which lacks the spectral resolution needed to distinguish chemically similar polymers, complex metal alloys, and composite substrates on printed circuit boards (PCBs). This paper presents a multidisciplinary synthesis of AI-driven detection and classification for e-waste, bridging materials science and computer vision through three interconnected themes. 1. Material and Economic Context: The toxicological risks and economic drivers of semiconductor recycling are characterized, framing fine-grained material identification as essential for a circular economy. 2. Multispectral Sensing & Fusion: Sensing modalities such as near-infrared (NIR), hyperspectral imaging (HSI), and X-ray fluorescence (XRF) are assessed, and sensor fusion strategies, including early, late, and intermediate fusion, are reviewed for high-throughput industrial settings. 3. Deep Learning Benchmarking: 11 publicly available PCB datasets are analyzed, and the YOLO series (YOLOv3–YOLOv12) is compared with leading non-YOLO detectors, including Faster R-CNN, RT-DETR-L, and RetinaNet. The results show that while YOLOv9s achieves a peak mAP@0.5 of 56.5% and YOLOv11s offers an optimal industrial profile (37.2% mAP@0.5:0.95 at 115 ms edge inference), all RGB-based models fail to detect visually ambiguous surface-mount devices (SMDs), with mAP values below 12%. This confirms a performance ceiling for purely visual systems. The review concludes that transitioning from RGB-centric to multispectral fusion architectures is the primary research frontier and proposes a roadmap for standardized multimodal datasets and edge-deployable fusion models to enable next-generation, high-recovery automated recycling. Full article

Context: Accurate citrus crop mapping is essential for agricultural monitoring, production planning, and supply-chain management, particularly in Brazil, one of the world’s leading orange producers and the leading orange-juice exporter. Satellite image time series from Sentinel-2 provide rich spectral and temporal information for crop identification. However, citrus mapping remains challenging due to fragmented agricultural landscapes, cloud contamination, class imbalance, and spectral overlap with other vegetation classes. Problem: Conventional machine learning models often depend on handcrafted vegetation indices, while attention-based deep learning models may require larger datasets and can become unstable under geographically constrained conditions. Therefore, there is a need for a compact and robust deep learning architecture capable of extracting citrus phenological signatures directly from multispectral time-series data. Methods: This study evaluates a Spatio-Temporal Pixel-Set Encoder Convolutional Neural Network (PSE-CNN) for citrus crop classification in the immediate geographic regions of São João da Boa Vista and Mogi Guaçu, São Paulo, Brazil. MapBiomas Collection 10.1 data from 2019 to 2024 were used to derive reference polygons, and Sentinel-2 imagery was processed into cloud-masked, 15-day temporal composites using ten spectral bands. The proposed PSE-CNN was benchmarked against PSE-TAE, PSE-Transformer, Random Forest, and XGBoost using spatially grouped data partitioning and temporal test years. Results: The proposed PSE-CNN achieved the highest Unified F1-Score of $0.704$ and the lowest coefficient of variation of $3.03\%$, indicating stronger inter-annual stability across test years and random seeds among the evaluated models. It also outperformed classical models that relied on handcrafted vegetation indices and demonstrated greater overall stability than attention-based deep learning alternatives. Conclusions: The results indicate that combining pixel-set encoding with temporal convolution provides a resource-aware and stable framework for retrospective citrus crop mapping from Sentinel-2 satellite image time series. These findings suggest that PSE-CNN can support scalable agricultural monitoring, contributing to sustainable crop inventory systems in regions where labeled data and computational infrastructure are limited. Full article

The authentication of fish maw is of considerable importance for preventing product substitution and protecting market confidence in high-value aquatic foods. This study developed a rapid and nondestructive authentication strategy by combining hyperspectral imaging (HSI) with wavelength selection and a Kolmogorov–Arnold Network (KAN) to discriminate 10 commercially representative fish maw varieties. Hyperspectral datasets were collected in the visible and near-infrared (VNIR, 400–1000 nm) and short-wave infrared (SWIR, 900–1700 nm) regions. To improve spectral quality and model robustness, four preprocessing methods (SG, SG−MeanNor, SG−DT, and SG−SNV) were evaluated, followed by the construction of PLS-DA, SVM, MLP, CNN, and KAN models. Feature wavelengths were subsequently selected separately from the VNIR and SWIR spectra using CARS, iVISSA, and SPA to establish reduced-variable authentication models. The results showed that SG-DT achieved the best overall preprocessing effect, confirming its ability to reduce spectral noise and baseline variation. In addition, SWIR-based models consistently outperformed VNIR-based models, suggesting that compositional information captured in the SWIR region played an important role in fish maw authentication. Among all tested models, the SWIR@SG-DT-SPA-KAN model exhibited the best performance, achieving 98.67% accuracy, 98.75% precision, 98.67% recall, and 98.64% F1-score using only 16 SPA-selected wavelengths from the SG-DT-preprocessed SWIR spectra. This study demonstrates that HSI coupled with feature wavelength and KAN modeling can provide an accurate and efficient tool for fish maw authentication. More importantly, the reduced-wavelength model offers practical potential for developing fast and cost-effective multispectral systems for authenticity screening in the aquatic food market. Full article

In saline–alkali environments, soil salinity imposes severe abiotic stress on maize growth by inhibiting root activity and nutrient uptake. Traditional destructive sampling methods struggle to enable cross-growth stage, large-scale dynamic fertilizer effect assessment. This study, conducted in saline–alkali farmlands of Inner Mongolia, utilized UAV multispectral remote sensing to extract 20 vegetation indices and 40 texture parameters, constructing a multi-source feature set. An ensemble learning framework integrating Random Forest (RF), Decision Tree (DTR), AdaBoost and Gradient Boosting Regression (GBR) was developed to achieve precise monitoring of maize plant height, leaf area index (LAI), and yield. In addition, the study aimed to evaluate the dynamic effects of seven fertilizer treatments (six controlled-release composite fertilizers, T1–T6, and conventional CK) and to identify the optimal fertilization scheme, with particular emphasis on comparing the two best-performing treatments, T1 and T2. Results showed that: (1) The ensemble model improved prediction robustness, with R² values of 0.88, 0.76, and 0.76 for plant height, LAI, and yield across the entire growth cycle, respectively. The integration of texture features effectively mitigated spectral saturation during peak growth stages (e.g., tasseling and filling). (2) For fertilizer evaluation, T1 performed best in growth and yield at jointing, tasseling, and filling stages, with a yield increase rate of up to 40.18% at the jointing stage. Although T2 slightly outperformed T1 in yield increase at maturity (15.42%), T1 was identified as the optimal fertilizer scheme for the region based on whole-growth-stage growth performance, measured yield, LAI, and yield increase rate. These results demonstrate that UAV-based multi-source feature fusion combined with ensemble learning provides an effective and non-destructive approach for fertilizer evaluation and precision nutrient management in saline–alkali regions. Full article