EUVIMEDEUVIMED entdecken
Crossref

Spatiotemporal water quality assessment of Xingqing Lake, China: Integrating water quality index, principal component analysis, and machine learning for SDG 6 reporting

Soundous Dahlab, Hafidha Rahmoune, Zhou Yuqin

Journal of Environmental Quality · 2026-11 · 55 (6)

DOI: 10.1002/jeq2.70333 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Environmental Quality, online erschienen am 18.11.2026. Verfasst von Soundous Dahlab und 2 weiteren Autor:innen. Die Arbeit nennt 27 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Urban lakes are increasingly affected by stormwater runoff, wastewater inputs, and intensive human activities, while water‐quality assessments increasingly combine indices, multivariate statistics, and machine learning (ML) but often emphasize different stages of assessment, diagnosis, prediction, or management. This study introduces an integrated water quality index (WQI)–principal component analysis (PCA)–ML framework for comprehensive water‐quality assessment, demonstrated through a case study at Xingqing Lake, China (2022–2025). The study analyzed seven parameters (pH, dissolved oxygen, permanganate chemical oxygen demand, dichromate chemical oxygen demand, 5‐day biochemical oxygen demand [BOD 5 ], NH 4 + ‐N, and total phosphorus) at five monitoring sites, using WQI for classification, PCA to identify dominant multivariate water‐quality patterns, and three ML models (random forest [RF], extreme gradient boosting [XGBoost], and long short‐term memory) for predictive modeling. Results revealed clear spatial gradients, with S4 and S5 showing comparatively favorable WQI conditions, while S1–S3 exhibited greater deterioration and temporal variability. RF and XGBoost achieved strong contemporaneous WQI‐mapping performance (RF: R 2 = 0.924; XGBoost: R 2 = 0.918). PCA identified two primary drivers: PC1 (57.6% of annual variance) representing nutrient‐organic pollution and PC2 (18.5%) representing oxygenation (where PC represents principal component). PCA and ML findings showed strong convergence ( r = 0.873, p < 0.01), with NH 4 + ‐N and BOD 5 identified as dominant factors. This integrated monitoring approach aligns with Sustainable Development Goal (SDG) Target 6.3, which emphasizes improving ambient water quality through pollution reduction and long‐term water‐quality assessment. Although WQI, PCA, and ML have been used individually or in pairs, their coordinated use for simultaneous classification, multivariate diagnosis, and predictive forecasting in urban lakes remains underexplored. The integrated WQI–PCA–ML framework provides a robust diagnostic and forecasting tool for continuous local water‐quality management relevant to SDG Target 6.3 and complementary to formal Indicator 6.3.2 reporting.

Zitieren

APA

Dahlab, S., Rahmoune, H., & Yuqin, Z. (2026). Spatiotemporal water quality assessment of Xingqing Lake, China: Integrating water quality index, principal component analysis, and machine learning for SDG 6 reporting. Journal of Environmental Quality, 55(6). https://doi.org/10.1002/jeq2.70333

Vancouver

Dahlab S, Rahmoune H, Yuqin Z. Spatiotemporal water quality assessment of Xingqing Lake, China: Integrating water quality index, principal component analysis, and machine learning for SDG 6 reporting. Journal of Environmental Quality. 2026;55(6). doi:10.1002/jeq2.70333

BibTeX
@article{Dahlab2026,
  author = {Dahlab, Soundous and Rahmoune, Hafidha and Yuqin, Zhou},
  title = {Spatiotemporal water quality assessment of Xingqing Lake, China: Integrating water quality index, principal component analysis, and machine learning for SDG 6 reporting},
  journal = {Journal of Environmental Quality},
  year = {2026},
  volume = {55},
  number = {6},
  doi = {10.1002/jeq2.70333}
}
Kostenlos starten →