- Research Article
- 10.1016/j.chemgeo.2026.123343
Integrated hydrochemical and geochemical modeling of cadmium migration and source apportionment in shallow groundwater
- May 01, 2026
- Chemical Geology
- Yuxing Fan + 9 more +9
Publications from 2021 to 2026
Showing 10 of 202 papers
Integrated hydrochemical and geochemical modeling of cadmium migration and source apportionment in shallow groundwater
Assessing the role of central banks in addressing financial sector carbon emissions
How purpose-specific AI use builds organizational resilience: A dynamic capability perspective
Bioaccumulation of trace elements and ecological risks induced by petroleum extraction in Shengli oilfield, China.
Oil extraction activities generate both organic and metal pollution, however, trace elements in oilfield areas have received less attention in terms of investigation and risk assessment. This study investigated the accumulation levels of eight trace elements (Cd, Mn, Ni, Cu, Pb, Zn, Cr, and Hg) in soils, plants, and animals from three sampling lines using ICP-MS. Potential ecological risks of these elements were assessed based on the local geochemical background of soils and the hazardous concentration for 5% species (HC5) derived from the Species Sensitivity Distribution Curves of plant and animals. Biomagnification of these elements was characterized with the Biomagnification Factor (BMF). The results indicated potential soil contamination with Pb (19.8 ± 5.2mg/kg) and Hg (0.023 ± 0.012mg/kg), exceeding local environmental background values by 1.02 and 1.04 times, respectively. High accumulation of Mn, Zn, and Cu was observed in plants and animals, with concentrations ranging from 21.2 to 153, 20.4-335, and 12.7-143mg/kg, respectively. Higher elemental concentrations were observed in Goosegrass (Mn: 93.5 ± 15.3mg/kg, Zn: 32.2 ± 7.9mg/kg, Cu: 9.1 ± 1.0mg/kg) and grubs (Mn: 152 ± 38.1mg/kg, Zn: 202 ± 45.3mg/kg, Cu: 118 ± 8.1mg/kg), demonstrating a stronger capacity for accumulating these elements. Mn, Cu, and Zn posed relatively high risks to the investigated organisms especially for insects, with Hazard Quotient (HQ) value high to 12,059, 25,191 and 9017, respectively. Biomagnification was evident for Zn, Hg, Cd, and Cu through the food chain transfer, with BMF high to 6.1, 26.1, 15.2, and 12.1, respectively. These findings highlight that less-regulated trace elements like Mn, Cu, and Zn, often perceived as less hazardous, can present ecological risks in oilfields via high bioaccumulation and trophic transfer, warranting greater attention in oilfield management.
Read moreNew meshless finite difference method based on hybrid orthogonal and conventional polynomial approximation
Corrigendum to “‘Decoding LLMs’ verbal deception in online reviews” [Decision Support Systems 200 (2026) 114529
Bridging virology, public health, and food safety: a comprehensive review of hepatitis A virus.
Hepatitis A virus (HAV) remains a significant global public health concern, particularly as a foodborne pathogen. Despite advances in vaccination and sanitation, HAV continues to cause outbreaks, especially in regions with uneven immunization coverage or inadequate hygiene infrastructure. This review comprehensively examines HAV from virological, clinical, and epidemiological perspectives, highlighting its pathogenesis, transmission routes, genomic diversity, and foodborne risks. We discuss the evolving global epidemiology of HAV, emphasizing the shift from high to low endemicity in vaccinated populations and the emergence of susceptible cohorts in moderately developed regions. The review underscores the challenges in foodborne HAV detection, including the limitations of current methods and the promise of emerging technologies like reverse transcription digital polymerase chain reaction (RT-dPCR) for enhanced sensitivity. Additionally, we summarize emerging antiviral approaches infood safety control, including plant-derived bioactive compounds and edible coatings, for controlling HAV transmission through the food chain. Finally, we present an overview of WHO-recommended vaccination strategies based on endemicity levels-targeted immunization for high-risk groups in low-endemic areas, universal childhood vaccination in moderate-endemic regions, and reliance on naturally acquired immunity in high-endemic zones. Our synthesis aims to bridge gaps between virology, clinical medicine, and food safety, offering insights for interdisciplinary collaboration in HAV prevention and control.
Read more3D reconstruction and material analysis of Neolithic wattle-and-daub houses at Fenghuangzui site in China
Wattle-and-daub houses emerged in the middle Yangtze River valley between 6000 and 5000 BCE, serving as the dominant form of housing for about 4000 years. Unfortunately, their susceptibility to soil erosion, weathering, and insect damage has left few surviving examples. We present a comprehensive analysis of wattle-and-daub houses, plastered white on walls and/or floors, from the Upper Qujialing-Shijiahe period (3300–2200 BCE) at the Fenghuangzui site in China. Through three-dimensional reconstruction based on red-burnt clay chunks and wood imprints, we proposed the general construction sequence and illustrated how intentional firing enhanced the durability of these dwellings. Initial scientific characterization of white-surfaced plaster (bai hui mian) fragments using XRF, XRD, SEM-EDS, Raman spectroscopy, and FTIR indicates the potential use of a gel-like calcium aluminosilicate hydrate (C-A-S-H) binder. This study provides new insights into architectural practices, technological innovations, and social organization at Neolithic walled towns in the middle Yangtze River valley.
Read moreAn enclosing formation algorithm based on transverse functions for underactuated unmanned surface vehicle network
Table tennis trajectory prediction method based on GRU network
As computer vision technology and deep learning models develop, the demand for trajectory perception and prediction of high-speed small targets is increasing day by day. To raise the detection and trajectory prediction accuracy of table tennis targets in complex backgrounds, a trajectory prediction model combining improved YOLOv8 and gated loop units is proposed. The detection part introduces a multihead self-attention mechanism, spatial pyramid pooling module, and cross channel feature fusion structure based on YOLOv8 to achieve stable recognition of small target ping-pong balls. The trajectory prediction part constructs a sequence modeling network based on gated recurrent units to perform multistep prediction on the detection results. The findings denote that the average accuracy of the detection part is 93.5%, and the F1 value has been improved to 90.8%, which is 5.1% and 4.7% higher than the original YOLOv8, respectively. In trajectory prediction tasks, the maximum error of the trajectory prediction model in scenarios such as occlusion and high-speed changes does not exceed ± 0.02, which is significantly better than the comparison model. The findings validated the effectiveness and robustness of the designed method in intelligent recognition and prediction of table tennis trajectories, providing technical support for sports intelligence systems.
Read more