Search Results (143)

The prevention and control of mine inrush water has always been a major challenge for safety. By identifying the type of water source and analyzing the real-time changes in water composition, sudden water inrush accidents can be monitored in a timely manner to avoid major accidents. This paper proposes a novel explainable machine learning model for source type identification of mine inrush water. The paper expands the original monitoring system into the XinJi No.2 Mine in Huainan Mining Area. Based on the online water composition data, using the Spearman coefficient formula, it analyzes the water chemical characteristics of different aquifers to extract key discriminant factors. Then, the Conv1D-GRU model was built to deeply connect factors for precise water source identification. The experimental results show an accuracy rate of 85.37%. In addition, focused on the interpretability, the experiment quantified the impact of different features on the model using SHAP (Shapley Additive Explanations). It provides new reference for the source type identification of mine inrush water in mine disaster prevention and control. Full article

The Chinese mitten crab (Eriocheir sinensis) industry is currently facing the challenges of origin fraud, as well as a lack of precision and interpretability of existing traceability methods. Here, we propose a high-precision origin traceability method based on a combination of stable isotope analysis and interpretable machine learning. We sampled Chinese mitten crabs from six origins representing diverse aquatic environments and farming practices, and analyzed their δ¹³C, δ¹⁵N, δ²H, and δ¹⁸O stable isotope compositions in different sexes and tissues (hepatopancreas, muscle, and gonad). By comparing the classification performance of Random Forest, XGBoost, and Logistic Regression models, we found that the Random Forest model outperformed the others, achieving high accuracy (91.3%) in distinguishing samples from different origins. Interpretation of the optimal Random Forest model, using SHAP (SHapley Additive exPlanations) analysis, identified δ²H in male muscle, δ¹⁵N in female hepatopancreas, and δ¹³C in female hepatopancreas as the most influential features for discriminating geographic origin. This analysis highlighted the crucial role of environmental factors, such as water source, diet, and trophic level, in origin discrimination and demonstrated that isotopic characteristics of different tissues provide unique discriminatory information. This study offers a novel paradigm for stable isotope traceability based on explainable machine learning, significantly enhancing the identification capability and reliability of Chinese mitten crab origin traceability, and holds significant implications for food safety assurance. Full article

In realistic hyperspectral image (HSI) cross-scene classification tasks, it is ideal to obtain target domain samples during the training phase. Therefore, a model needs to be trained on one or more source domains (SD) and achieve robust domain generalization (DG) performance on an unknown target domain (TD). Popular DG strategies constrain the model’s predictive behavior in synthetic space through deep, nonlinear source expansion, and an HSI generation model is usually adopted to enrich the diversity of training samples. However, recent studies have shown that the activation functions of neurons in a network exhibit asymmetry for different categories, which results in the learning of task-irrelevant features while attempting to learn task-related features (called “feature contamination”). For example, even if some intrinsic features of HSIs (lighting conditions, atmospheric environment, etc.) are irrelevant to the label, the neural network still tends to learn them, resulting in features that make the classification related to these spurious components. To alleviate this problem, this study replaces the common nonlinear generative network with a specific linear projection transformation, to reduce the number of neurons activated nonlinearly during training and alleviate the learning of contaminated features. Specifically, this study proposes a dimensionally decoupled spatial spectral linear extrapolation (SSLE) strategy to achieve sample augmentation. Inspired by the weakening effect of water vapor absorption and Rayleigh scattering on band reflectivity, we simulate a common spectral drift based on Markov random fields to achieve linear spectral augmentation. Further considering the common co-occurrence phenomenon of patch images in space, we design spatial weights combined with label determinism of the center pixel to construct linear spatial enhancement. Finally, to ensure the cognitive unity of the high-level features of the discriminator in the sample space, we use inter-class contrastive learning to align the back-end feature representation. Extensive experiments were conducted on four datasets, an ablation study showed the effectiveness of the proposed modules, and a comparative analysis with advanced DG algorithms showed the superiority of our model in the face of various spectral and category shifts. In particular, on the Houston18/Shanghai datasets, its overall accuracy was 0.51%/0.83% higher than the best results of the other methods, and its Kappa coefficient was 0.78%/2.07% higher, respectively. Full article

Objectives: This study investigated the effects of 2-week ingestion of hemp fiber (high and low doses) versus placebo bars on gut permeability and plasma metabolite shifts during recovery from 2.25 h intensive cycling. Hemp hull powder is a rich source of two bioactive compounds, N-trans-caffeoyl tyramine (NCT) and N-trans-feruloyl tyramine (NFT), with potential gut health benefits. Methods: The study participants included 23 male and female cyclists. A three-arm randomized, placebo-controlled, double-blind, crossover design was used with two 2-week supplementation periods and 2-week washout periods. Supplement bars provided 20, 5, or 0 g/d of hemp hull powder. Participants engaged in an intensive 2.25 h cycling bout at the end of each of the three supplementation periods. Five blood samples were collected before and after supplementation (overnight fasted state), and at 0 h-, 1.5 h-, and 3 h-post-exercise. Five-hour urine samples were collected pre-supplementation and post-2.25 h cycling after ingesting a sugar solution containing 5 g of lactulose, 100 mg of ¹³C mannitol, and 1.9 g of mannitol in 450 mL of water. An increase in the post-exercise lactulose/¹³C mannitol ratio (L:¹³CM) was used as the primary indicator of altered gut permeability. Other outcome measures included muscle damage biomarkers (serum creatine kinase, myoglobin), serum cortisol, complete blood cell counts, and shifts in plasma metabolites using untargeted metabolomics. Results: No trial differences were found for L:¹³CM, cortisol, blood cell counts, and muscle damage biomarkers. Orthogonal partial least-squares discriminant analysis (OPLSDA) showed distinct trial differences when comparing high- and low-dose hemp fiber compared to placebo supplementation (R2Y = 0.987 and 0.995, respectively). Variable Importance in Projection (VIP) scores identified several relevant metabolites, including 3-hydroxy-4-methoxybenzoic acid (VIP = 1.9), serotonin (VIP = 1.5), 5-hydroxytryptophan (VIP = 1.4), and 4-methoxycinnamic acid (VIP = 1.4). Mummichog analysis showed significant effects of hemp fiber intake on multiple metabolic pathways, including alpha-linolenic acid, porphyrin, sphingolipid, arginine and proline, tryptophan, and primary bile acid metabolism. Conclusions: Hemp fiber intake during a 2-week supplementation period did not have a significant effect on post-exercise gut permeability in cyclists (2.25 h cycling bout) using urine sugar data. On the contrary, untargeted metabolomics showed that the combination of consuming nutrient-rich hemp fiber bars and exercising for 135 min increased levels of beneficial metabolites, including those derived from the gut in healthy cyclists. Full article

The deep-water area of the Qiongdongnan basin is currently a hot topic for exploration. The discovery of gas fields in the Baodao sag confirms its abundant oil and gas resources and potential, making it of significant economic and strategic importance. The complexity of sedimentary structural evolution within the Baodao sag makes the process of oil and gas accumulation in the area extremely complex, and the law of natural gas enrichment is difficult to grasp, resulting in unclear exploration directions. Therefore, this study focuses on the third member of the Lingshui Formation in the Paleogene of the Baodao sag. Based on the abundant thin section, scanning electron microscopy, 3D seismic and geochemical analysis data in the area, through analyzing the density of natural gas, the proportion of hydrocarbon and non-hydrocarbon components, the dryness coefficient carbon, and the isotopic characteristics, combined with the deep natural gas genesis discrimination chart, the types and genesis types of natural gas and organic matter in the sag are clarified. In addition, combined with the package and BasinMod 2009 software, the filling period and reservoir-filling process were clarified and restored. At the same time, the reservoir formation characteristics of the different fault-step zones inside the sag were dissected and the primary and secondary migration of natural gas were analyzed in order to clarify the types and characteristics of different fault-step zone transport systems. Finally, the research findings indicate that there are two reservoir formation modes developed within the depression, as follows: “multiple hydrocarbon generation and control sources—continuous vertical control of large faults—lateral sand body convergence (T + Z-type transport)—multiple cap layer closure” and “mixed-source hydrocarbon supply—continuous vertical control of large faults—short lateral sand body convergence (Z-type transport)—multiple cap layer closure”, providing an important basis for the next exploration of the basin. Full article

Thorium is a notable candidate for resolving uranium shortage caused by the global application of nuclear power generation. Uranium extraction from seawater is another attempt to handle its source deficiency, however, vanadium is one of the main competitive elements in that process. Exploration of probes which can discriminatively detect thorium and vanadium from uranium has primary significance for their further separation and for environmental protection. Herein, N′-(2,4-dihydroxybenzylidene)-4-hydroxylphenylhydrazide, AOH, is used as sensor for Th⁴⁺ and vanadyl (VO²⁺) determination. AOH demonstrates a specific “turn-on” fluorescence selectivity towards Th⁴⁺ over f-block and other foreign metal ions, with a detection limit (LOD) of 7.19 nM in acidic solution and a binding constant of 9.97 × 10⁹ M⁻². Meanwhile, it shows a “turn-off” fluorescence response towards VO²⁺ over other metal ions at the coexistence of Th⁴⁺, with a LOD of 0.386 μM in the same media and a binding constant of 4.54 × 10⁴ M⁻¹. The recognition mechanism, based on HRMS, ¹H NMR, and FT-IR results, demonstrates that VO²⁺ causes the fluorescence quenching by replacing Th⁴⁺ to coordinate with AOH. In real water detection tests, Th⁴⁺ and VO²⁺ exhibited satisfying recoveries. These findings expand the application of sensors in nuclide pollution control. Full article

Watercress (Nasturtium officinale), a freshwater aquatic plant in the Brassicaceae family, is characterized by its high content of specialized metabolites, including flavonoids, glucosinolates, and isothiocyanates. Traditionally, commercial cultivation is conducted in submerged beds using river or spring water, often on soil or gravel substrates. However, these methods have significant environmental impacts, such as promoting eutrophication due to excessive fertilizer use and contaminating water sources with pesticides. This study aimed to explore two emerging cultivation strategies, i.e., hydroponics and aquaponics, to grow watercress and evaluate its specialized metabolite content using an untargeted metabolomic approach. The goal was to characterize metabolic profiles, identify component variations, and assess changes in metabolite accumulation at two harvest times. Two culture systems (hydroponic and aquaponic) and two harvest stages (‘baby leaf’ and traditional harvest) were examined. The results revealed 23 key metabolites, predominantly glucosinolates and flavonoids, that significantly influenced the metabolic profile discrimination, with the aquaponic system yielding the highest diversity and relative abundance of metabolites (variable importance in the projection (VIP) > 1). Important condition-related compounds were identified via cross-validation (area under the curve (AUC) > 0.7), including isorhamnetin sophoroside–glucoside and gluconasturtiin at the traditional harvest in the hydroponic system and glucoarabin at the ‘baby leaf’ stage in the aquaponic system. These findings highlight the potential of aquaponic and hydroponic systems as sustainable alternatives for watercress cultivation, offering environmental benefits and enhanced metabolite quality. Full article

In arid areas, droughts caused by climate change seriously impact wheat production. Therefore, research on spatial and temporal variability of dry and hot wind events and drought risk under different development patterns of future climate can provide a reference for wheat cultivation planning in the study area. Based on meteorological data under three scenarios of the CMIP6 (Sixth International Coupled Model Comparison Program) shared socio-economic path (SSP), we introduced wheat dry hot wind discrimination criteria and calculated the Standardized Precipitation–Evapotranspiration Index (SPEI). Future temperature changes within the Ningxia Province were consistent, increasing at a rate of 0.037, 0.15 and 0.45 °C·(10 a⁻¹) under SSP126, 245 and 585 scenarios, respectively. Simultaneously, average annual precipitation would increase by 17.77, 38.73 and 32.12 mm, respectively. Dry hot wind frequency differed spatially, being higher in northern Ningxia and western Ningxia, and lower in southern Ningxia and eastern Ningxia. During the wheat growing period, there is an obvious increasing drought risk trend under the SSP585 model in May, and the possibility of drought risk in the middle period was highest under the SSP126 model. In June, SPEI was generally higher than in May, and the risk of alternating drought and flood was greater under the SSP585 model, while near-medium drought risk was lower under the SSP126 and SSP245 models. The influence of DHW (dry and hot wind) on wheat yield will increase with the increase of warming level. However, when DHW occurs, effective irrigation can mitigate the harm. Irrigation water can be sourced from various channels, including rainfall, diversion, and groundwater. These results provide scientific reference for sustainable agricultural production, drought risk and wheat meteorological disaster forecast in inland arid areas affected by climate change. Full article

The severe water quality pollution of the Bohai Sea impacts both the ecosystem and the economy of the region. This study assesses the water quality of the Bohai Sea using a two-year (2020–2021) dataset to investigate the spatial distribution and sources of contamination. Multivariate statistical analysis methods, including principal component analysis (PCA), cluster analysis (CA), and discriminant analysis, are employed. Thirteen chemical indicators are analyzed through PCA, resulting in the extraction of three principal components that reflect different pollution sources related to domestic, industrial, and agricultural activities. Additionally, the corresponding water quality index (WQI) is calculated to categorize the water quality into three levels using CA. The PCA-based WQI method is feasible and shows similarities to the traditional WQI method. Higher pollution levels are observed in Panjin and Tianjin, while Huludao, Yantai, and Dalian exhibit relatively good water quality. The results indicate complex, multifactorial pollution causes in the Bohai Sea, including eutrophication, heavy metal contamination, and ammonia pollution. These findings can guide region-specific water quality management: Panjin should control heavy metal discharges from industry and transportation, while Tianjin requires improvements in runoff management of ammonia-based fertilizers. Together, these strategies support the ecological and sustainable development of the Bohai Sea. Full article

Microbial contamination and the prevalence of foodborne pathogens in mutton meat and during its slaughtering process were investigated through microbial source tracking and automated pathogen identification techniques. Samples from mutton meat, cutting boards, hand swabs, knives, weighing balances, and water sources were collected from four different retail sites in Coimbatore. Total plate count (TPC), yeast and mold count (YMC), coliforms, E. coli, Pseudomonas aeruginosa, Salmonella, and Staphylococcus were examined across 91 samples. The highest microbial loads were found in the mutton-washed water, mutton meat, and cutting board samples. The automated pathogen identification system identified Staphylococcus species as the predominant contaminant and also revealed a 57% prevalence of Salmonella. Further analysis of goat meat inoculated with specific pathogens showed distinct volatile and metabolite profiles, identified using gas chromatography-mass spectrometry (GC-MS). Multivariate statistical analyses, including principal component analysis (PCA), orthogonal partial least squares discriminant analysis (OPLS-DA), and sparse partial least squares discriminant analysis (sPLS-DA), identified potential biomarkers for pathogen contamination. The results highlight the significance of cross-contamination in the slaughtering process and suggest the use of volatile compounds as potential biomarkers for pathogen detection. Full article

The Waxing Mo polymetallic deposit is located in the central part of the Lesser Xing’an–Zhangguangcai Range (LXZR), NE China. The Mo (Cu) mineralization in the deposit is dominantly hosted by quartz veinlets and stockworks and is closely related to silicification and potassic alteration, while the W mineralization is most closely related to greisenization. Zircon samples from granodiorite, biotite monzogranite, granodiorite porphyry, and syenogranite in the Waxing deposit yielded U-Pb ages of 172.3 Ma, 172.8 Ma, 173.0 Ma, and 171.4 Ma, respectively. Six molybdenite samples from porphyry Mo ores yielded a Re-Os isochron age of 172.0 ± 1.1 Ma. The granitoids in the ore district are relatively high in total alkali (Na₂O + K₂O), are metaluminous to weakly peraluminous, and are classified as I-type granitoids. The zircon samples from all granitoids showed a relatively consistent Hf isotopic composition, as shown by positive ε_Hf(t) values (3.1–8.3) and young T_DM2 ages (0.69–1.25 Ga). These results, combined with the whole-rock geochemistry, suggest that the magma source of these rocks most likely derived from partial melting of a juvenile middle-lower continental crust, with a minor contribution from the mantle. These granitoids have compositional characteristics of adakites such as relatively high Sr contents (e.g., >400 ppm) and Sr/Y ratios (e.g., >33), as well as weak Eu anomalies (e.g., Eu/Eu* = 0.8–1.1), indicating extensive fractionation crystallization of a hydrous magma. The apatite geochemistry indicates that the ore-related magma in Waxing is F-rich and has a relatively low content of sulfur. The zircon geochemistry reveals that the granodiorite, biotite monzogranite, and granodiorite porphyry have relatively high oxygen fugacity (i.e., ΔFMQ = +1.1~1.3), whereas the fO₂ values of the granite porphyry and syenogranite are relatively low (i.e., ΔFMQ = +0.1~0.5). The whole-rock and mineral geochemistry suggest that the Mo mineralization in Waxing is probably genetically related to granitoids (i.e., granodiorite, biotite monzogranite, and granodiorite porphyry), with higher oxygen fugacity and a high water content, whereas the magmatic S concentration is not the key factor controlling the mineralization. A comparison of the geochemical compositions of ore-forming and barren stocks for porphyry Mo deposits in the LXZR showed that geochemical ratios, including Eu/Eu* (>0.8), 10,000*(Eu/Eu*)/Y (>600), Sr/Y (>33), and V/Sc (>8), could be effective indicators in discriminating fertile granitoids for porphyry Mo deposits from barren ones in the region. Full article

Bubble detection in water plays important roles in human exploration and management of the ocean. This research presents a detection technique based on laser polarization dual-mode fusion, aiming at solving the difficulties of light scattering intensity characteristics that are hard to extract and the small particle size of underwater bubbles that are hard to detect. To increase the precision of bubble identification, an image fusion technique based on bubble polarization degree is first presented. Second, we quantitatively investigate the grayscale undulation of bubbles with different size and number distributions in the image from both statistical and experimental aspects, introduce image grayscale fluctuation (GF) to fuse two modes of laser polarization and the image, establish an a posteriori distribution probability model of discriminating features such as the size and number of bubbles, and realize the bubble small-sample, multi-source data fitting. The findings demonstrate that dynamic bubble detection in the 50–1000 μm and 100–2000 cm⁻³ ranges can achieve more than 95%, as well as more than a 93%, accuracy in quantity distribution and bubble size change. This technique achieves the continuous perception of bubble features in complicated underwater environments, and offers a possible application scheme for the detection of marine bubble environments. Full article

Numerous scholars worldwide have conducted extensive research on the identification of water sources for mine water inflows, among which the utilization of groundwater’s chemical properties for water source discrimination is characterized by its rapidity, effectiveness, and economy. In the Gaohe Coal Mine of Shanxi Province, anomalous water discharge has been observed from boreholes in some coal-bearing aquifers. The water quality differs from both coal-bearing aquifer water and Ordovician limestone aquifer water. Analysis of K⁺, Na⁺, and SO₄^2- suggests that the water does not belong to coal-bearing aquifer water, while the analysis of Ca²⁺ indicates it is not Ordovician limestone aquifer water. Particularly, in the 8# Coal-Bearing Aquifer Observation Borehole, the concentration of Ca²⁺ is extremely low, consistent with coal-bearing aquifer water, yet the concentration of SO₄^2- is extremely high, resembling Ordovician limestone water. This is speculated to be due to Ordovician limestone water replenishing the aquifer where the observation borehole is located, triggering a series of chemical reactions. Using the PHREEQC (Version 2) hydrochemical simulation software, hydrochemical simulation experiments were conducted to model the process of different proportions of Ordovician limestone water entering the coal-bearing aquifer. This study explored the reaction mechanisms between Ordovician limestone water, coal-bearing aquifer water, and coal measure aquifer rock samples, validated the hydrochemical and water–rock interactions occurring during this process, and estimated the proportion of water sources in the anomalous borehole water discharge based on the ion concentration profiles of the simulated mixed water. These findings can be applied to the prevention and control of Ordovician limestone water hazards, especially those caused by water-conducting pathways. Full article

Hypophthalmichthys nobilis are widely distributed in the Yangtze River basin and its related lakes. They are an important economic fish species and are a famous cultured species known as the “Four Famous Domestic Fishes” in China. Currently, with the fishing ban in the Yangtze River basin, fishing for H. nobilis in the natural water bodies of the Yangtze River basin has been completely prohibited. In order to identify the sources of H. nobilis appearing in the market, further control and accountability is necessary to trace the sources of H. nobilis in the Yangtze River basin and its related water bodies. Therefore, this study identified and traced different sources of H. nobilis through muscle element fingerprint analysis (EFA). The results show that H. nobilis from different stations have characteristic element compositions. The characteristic element of H. nobilis from Wuhan (WH) is Pb, which is significantly higher than that in other stations; the characteristic element from Anqing (AQ) is Hg, which is significantly higher than that in other stations; and the characteristic element from Taihu (TH) is Al, which is significantly higher than that in other water areas. Multivariate analysis selected different spatial distribution patterns in four discriminative element ratios (Pb/Ca, Cr/Ca, Na/Ca, and Al/Ca) in the muscle of H. nobilis in the Yangtze River basin and its related lakes. This study suggests that the screened discriminative elements can be used to visually distinguish different sources of H. nobilis and to quickly trace and verify the origin of newly emerging samples. Therefore, the use of selected discriminative element fingerprint features to trace the origin of new samples has been proven to be feasible. By further discriminating and verifying the muscle element fingerprints of new samples, the discrimination rate is high. Therefore, a multivariate analysis of muscle element fingerprints can be used for tracing the origins of samples of unknown origin in market supervision. Full article

The observation of the phytoplankton distribution with a high spatiotemporal resolution is necessary to track the nutrient sources that cause algal blooms and to understand their behavior in response to hydraulic phenomena. Photography from UAVs, which has an excellent temporal and spatial resolution, is an effective method to obtain water quality information comprehensively. In this study, we attempted to develop a method for estimating the chlorophyll concentration from aerial images using machine learning that considers brightness correction based on insolation and the spatial distribution of turbidity evaluated by satellite image analysis. The reflectance of harmful algae bloom (HAB) was different from that of phytoplankton seen under normal conditions; so, the images containing HAB were the causes of error in the estimation of the chlorophyll concentration. First, the images when the bloom occurred were extracted by the discrimination with machine learning. Then, the other images were used for the regression of the concentration. Finally, the coefficient of determination between the estimated chlorophyll concentration when no bloom occurred by the image analysis and the observed value reached 0.84. The proposed method enables the detailed depiction of the spatial distribution of the chlorophyll concentration, which contributes to the improvement in water quality management in reservoirs. Full article