EUVIMEDEUVIMED entdecken
Crossref

Evaluating Heavy Metal and Physicochemical Pollution in the Karnaphuli River Estuary through Remote Sensing and Machine Learning Approaches

Sheikh Samanin Tasnim, Mahima Yusuf Tarana, Dr. Mohammad Moniruzzaman, Sadia Afrin, Shamiha Shafinaz Shreya

Journal of Earth Observation and Geospatial Applications · 01.11.2026 · 2 (2) · 44-63

DOI: 10.65372/qgarvs44 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Earth Observation and Geospatial Applications, online erschienen am 01.11.2026. Verfasst von Sheikh Samanin Tasnim und 4 weiteren Autor:innen. Die Arbeit nennt 77 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The Karnaphuli River has critical impacts on the environmental and socio-economic situation of the surrounding catchment region, and it is crucial to analyze and monitor pollution for managing river water quality. The objective of the research was to evaluate and establish correlations between physicochemical and heavy metal pollution in water and a combination of laboratory testing, Sentinel-2A satellite imagery analysis, and machine learning for the Karnaphuli River estuary. High concentrations of heavy metals such as 12.8 mg/L Aluminum (Al), 15.5 mg/L Iron (Fe) and other pollutants have been found in the water samples of Karnaphuli. Continuous water pollution values were visualized using GIS interpolation maps, which showed a higher level of pollution in shipwrecking regions near the coast. Multiple linear regression (MLR) and principal component regression (PCR) were identified as the most effective models for predicting the variability of water quality parameters for this study based on remote sensing reflectance data. The satellite reflectance of Sentinel-2A bands produced significant results and predicted 58% (highest) of the variability of total suspended solids (TSS) and 43% (lowest) of the variability of pH in the Karnaphuli River. The interpolation maps aim to represent pollution levels more accurately than the raster calculation maps, as they incorporate spatial autocorrelation techniques based on observed values throughout the modeling process. This analysis provides insights into the dynamics of pollution, aids in prioritizing attention to reducing industrial discharges, encourages necessary infrastructural investments for sewage treatment, and further study and digitization of water quality measurement using remote sensing and machine learning.

Zitieren

APA

Tasnim, S. S., Tarana, M. Y., Moniruzzaman, D. M., Afrin, S., & Shreya, S. S. (2026). Evaluating Heavy Metal and Physicochemical Pollution in the Karnaphuli River Estuary through Remote Sensing and Machine Learning Approaches. Journal of Earth Observation and Geospatial Applications, 2(2), 44-63. https://doi.org/10.65372/qgarvs44

Vancouver

Tasnim SS, Tarana MY, Moniruzzaman DM, Afrin S, Shreya SS. Evaluating Heavy Metal and Physicochemical Pollution in the Karnaphuli River Estuary through Remote Sensing and Machine Learning Approaches. Journal of Earth Observation and Geospatial Applications. 2026;2(2):44-63. doi:10.65372/qgarvs44

BibTeX
@article{Tasnim2026,
  author = {Tasnim, Sheikh Samanin and Tarana, Mahima Yusuf and Moniruzzaman, Dr. Mohammad and Afrin, Sadia and Shreya, Shamiha Shafinaz},
  title = {Evaluating Heavy Metal and Physicochemical Pollution in the Karnaphuli River Estuary through Remote Sensing and Machine Learning Approaches},
  journal = {Journal of Earth Observation and Geospatial Applications},
  year = {2026},
  volume = {2},
  number = {2},
  pages = {44--63},
  doi = {10.65372/qgarvs44}
}
Kostenlos starten →