• exploratory data analysis;
  • hierarchical cluster analysis;
  • Kaduna River;
  • principal component analysis;
  • river water;
  • water quality


Exploratory data analysis such as hierarchical cluster analysis and principal component analysis were applied to water quality dataset of the Kaduna River, obtained during 3 years (2008–2010), monthly monitoring of eight key different sampling sites for 19 parameters to extract correlations and similarities between variables and to classify river sampling sites in groups of similar quality. Hierarchical cluster analysis grouped eight sampling sites into three statistically significant clusters of similar water composition. Six varifactors were obtained after varimax rotation of initial principal components using principal component analysis. These techniques gave an insight into the sources of pollution. Anthropogenic influence (municipal, industrial wastewater and agricultural run-off) was the major source of river water pollution.