A multivariate statistical approach to the integration of different land-uses, seasons, and water quality as water resources management tool.

Abstract

The externalities generated by disorderly urbanization and lack of proper planning becomes one of the main factors that must be considered in water resource management. To address the multiple uses of water and avoid conflicts among users, decision-making must integrate these factors into quality and quantity aspects. The water quality index (WQI), using the correlation matrix and the multivariate principal component analysis (PCA) and cluster analysis (CA) techniques were used to analyze the surface water quality, considering urban, rural, and industrial regions in an integrated way, even with data gaps. The results showed that the main parameters that impacted the water quality index were dissolved oxygen, elevation, and total phosphorus. The results of PCA analysis showed 86.25% of the variance in the data set, using physicochemical and topographic parameters. In the cluster analysis, the dissolved oxygen, elevation, total coliforms, E. coli, total phosphorus, total nitrogen, and temperature parameters showed a significant correlation between the data's dimensions. In the industrial region, the characteristic parameter was the organic load, in the rural region were nutrients (phosphorus and nitrogen), and in the urban region was E. coli (an indicator of the pathogenic organisms' presence). In the classification of the samples, there was a predominance of "Good" quality, however, samples classified as "Acceptable" and "Bad" occurred during the winter and spring months (dry season) in the rural and industrial regions. Water pollution is linked to inadequate land use and occupation and population density in certain regions without access to sanitation services.