Multivariate statistical analysis of water quality of the Densu River, Ghana

ABSTRACT

We present here an evaluation of the water quality of an inland watercourse, the Densu River, using multivariate statistical approaches; Principal Component and Cluster Analyses. Sixteen (16) physicochemical parameters were analysed from river water samples for ten (10) different stations. The sampling campaigns were done for four months, October 2018–January, 2019. Results from the Principal Component Analysis (PCA) revealed that five components with eigenvalues >1 were extracted from the water quality data, depicting 78% of total variation in the data. The first Principal Component (PC) explained 37% variance in the data and correlated highly with ions conductivity (0.97), temperature (0.70), total hardness (0.95), sodium (0.93), sulphate (0.80) and chloride (0.93). The second PC explained 12% variance and was loading high with Turbidity (0.84) and PO₄³⁻ (0.80). The third components explained 11% of the total variance and correlated highly with five-day Biological Oxygen Demand (BOD₅) (0.90) and Chemical Oxygen Demand (COD) (0.72). Cluster analysis (CA) classified similar sampling stations into 3 clusters based on their water quality characteristics. Cluster 1 consisted of S1 (Aplaku), and described as fairly good quality but highly influenced by natural activities such as the sea water intrusion. Cluster 2 comprised of S2, S10, S9, S7 (Weija, Potroase, Koforidua, Mangoase) and described as fairly good water quality, but highly influenced by both natural and anthropogenic activities. Cluster 3 consisted of S3, S4, S5, S6 and S8 (Ashalaja, Afuaman, Pakro, Nsawam and Akwadum) and described as poor water quality was mainly influenced by human activities. The study revealed that pollution sources in the Densu River were mainly from anthropogenic and natural activities; including agricultural, urban pollution, domestic waste discharge, geology and the influx of sea water. It is recommended that, effluents must be treated before discharged into the river and buffer zone policies should be enforced.

KEYWORDS:

• Anthropogenic activities
• Cluster Analysis (CA)
• Densu Basin
• Principal Component Analysis (PCA)

Acknowledgements

This work was supported with funding from the World Bank and the Government of Ghana through the Africa Centre of Excellence (ACE) Project hosted by the Regional Water and Environmental Sanitation Centre (RWESCK), KNUST. CSIR-Water Research Institute is duly acknowledged for availing their facilities and supporting the field work as well as the laboratory analysis. Authors are grateful to Mr. Victor Mante for his assistance during the field sampling and laboratory analysis phase of this work.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported with funding from the World Bank and the Government of Ghana through the Africa Centre of Excellence (ACE) Project hosted by the Regional Water and Environmental Sanitation Centre (RWESCK), KNUST