Combining self-organizing maps with WQI and PCA for assessing surface water quality – a case study, Kor River, southwest Iran

ABSTRACT

In comprehensive environmental assessments, numerous physical, chemical, and biological parameters are involved in the environment–organism relationship. So, better techniques to analyse environmental data seem to be necessary. In this study, the use of three integrated approaches of self-organizing map neural network, water quality index, and principal components analysis is proposed to obtain the classification, pollution status, and relations among many variables in water samples from Kor River in Fars province, Iran. Altogether, the impact of eight physicochemical parameters (pH, electrical conductivity, alkalinity, sodium adsorption ratio, NO₃, PO₄, SO₄, and Cl) and eight heavy metal contents (As, Cr, Cu, Hg, Mo, Ni, Pb, and Zn) are taken into account. The results indicate that the proposed computational approach presents a valuable tool for characterizing the environmental status of similar water courses.

Keywords:

• Environmental assessment
• self-organizing map
• water quality index
• principal components analysis
• pollution status
• Kor River

Acknowledgements

The authors would like to thank the medical geology research center, Shiraz University for financial support. Thanks are also extended to the research committee of Shiraz University for logistical support.

Funding

This work was supported by the medical geology research center, Shiraz University.