EUVIMEDEUVIMED entdecken
Crossref

Insights into the prediction and driving factors of perfluorinated and polyfluoroalkyl substances in sediments from Taihu Lake of China using machine learning models

Hui Xiao, Kai Liu, Ligang Deng, Wanru Yang, Lijian Sun, An Yi, Yuanhang Zhang, Xin Qian, Tong Liu, Huiming Li

Environmental Pollution · 2025

DOI: 10.1016/j.envpol.2025.127294 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Environmental Pollution, 2025. Verfasst von Hui Xiao und 9 weiteren Autor:innen. Beteiligte Einrichtung: School of Environment, Nanjing Normal University, Nanjing, 210023, China; State Key Laboratory of Water Pollution Control and Green Resource Recyclin…. Hauptthemen laut MEDLINE: Fluorocarbons, Lakes, Geologic Sediments, Water Pollutants, Chemical. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 58 Literaturangaben. Gefördert von: Natural Science Foundation of Jiangsu Province, National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The spatiotemporal changes and environmental driving factors governing the distribution of perfluorinated and polyfluoroalkyl substances (PFAS) in the sediment cores of shallow lakes remain inadequately understood. This study investigated the occurrence and predicted future trends of PFAS in sediment cores from a shallow lake in China, and revealed the key influencing factors using time-series prediction and interpretable machine learning (IML) models. A decreasing trend of PFAS with depth in the three representative areas was found, with long-chain PFAS, such as perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS), predominating. The RQ of the top (0-4 cm) sediment cores indicated that they were in at a moderate ecological risk. Time-series forecasts using the Attention-CL model predicted increasing PFOA and PFOS concentrations from 2023 to 2034, suggesting potential consequences of delays in effective policy implementation. Feature importance and Shapley analyses revealed the importance of pH, total organic carbon (TOC), and oxidation-reduction potential (ORP) for PFAS. Partial dependence plots and individual conditional expectation showed the threshold effect of pH on PFOA and perfluoroundecanoic acid (PFUdA) to be 6.5, the threshold effect of ORP to be 200 mV, and the threshold effect of electrical conductivity to be 300 μS/cm. The threshold effect of TOC on PFOA was 3.0 mg/g, whereas that on PFUdA was 3.5 mg/g. In general, the IML model highlights the importance of environmental parameters, such as pH, ORP, and TOC, in shaping PFAS distribution. These findings elucidate the complex interactions between sediment characteristics and PFAS distribution, informing targeted monitoring and remediation efforts in a broader environmental science field.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

FluorocarbonsLakesGeologic SedimentsWater Pollutants, ChemicalMachine LearningEnvironmental MonitoringChinaAlkanesulfonic AcidsCaprylatesPredictive Learning Models

Stichwörter der Autor:innen

Ecological risk assessmentInterpretable machine learningLake sediment coresPFASTime-series prediction

Substanzen

FluorocarbonsWater Pollutants, ChemicalAlkanesulfonic AcidsCaprylatesperfluorooctanoic acidperfluorooctane sulfonic acid

Autor:innen und Institutionen

  • School of Environment, Nanjing Normal University, Nanjing, 210023, China; State Key Laboratory of Water Pollution Control and Green Resource Recycling, Nanjing University, Nanjing, 210023, China.
  • State Key Laboratory of Water Pollution Control and Green Resource Recycling, Nanjing University, Nanjing, 210023, China.
  • School of Environment, Nanjing Normal University, Nanjing, 210023, China.
  • State Key Laboratory of Water Pollution Control and Green Resource Recycling, Nanjing University, Nanjing, 210023, China; Jiangsu Collaborative Innovation Center of Atmospheric Environment and Equipment Technology (CICA…
  • Faculty of Environmental Earth Science, Hokkaido University, Sapporo, Japan.
  • School of Environment, Nanjing Normal University, Nanjing, 210023, China; Jiangsu Province Engineering Research Center of Environmental Risk Prevention and Emergency Response Technology, Nanjing, 210023, China. Electron…

Förderung

  • Natural Science Foundation of Jiangsu Province
  • National Natural Science Foundation of China

Zitieren

APA

Xiao, H., Liu, K., Deng, L., Yang, W., Sun, L., Yi, A., Zhang, Y., Qian, X., Liu, T., & Li, H. (2025). Insights into the prediction and driving factors of perfluorinated and polyfluoroalkyl substances in sediments from Taihu Lake of China using machine learning models. Environmental Pollution. https://doi.org/10.1016/j.envpol.2025.127294

Vancouver

Xiao H, Liu K, Deng L, Yang W, Sun L, Yi A, et al. Insights into the prediction and driving factors of perfluorinated and polyfluoroalkyl substances in sediments from Taihu Lake of China using machine learning models. Environmental Pollution. 2025. doi:10.1016/j.envpol.2025.127294

BibTeX
@article{Xiao2025,
  author = {Xiao, Hui and Liu, Kai and Deng, Ligang and Yang, Wanru and Sun, Lijian and Yi, An and Zhang, Yuanhang and Qian, Xin and Liu, Tong and Li, Huiming},
  title = {Insights into the prediction and driving factors of perfluorinated and polyfluoroalkyl substances in sediments from Taihu Lake of China using machine learning models},
  journal = {Environmental Pollution},
  year = {2025},
  doi = {10.1016/j.envpol.2025.127294}
}
Kostenlos starten →