Abstract

Multivariate statistical techniques were applied to evaluate the temporal and spatial variation of water quality in the Cuiabá River basin that could be affected by urban effluent discharge and diffuse pollution. After data treatment, 75.74% of the total variance was explained in the 28 parameters determined in water samples, collected at 13 stations from July 2010 to May 2011. The first component (29.08%) was represented by dissolved ions from natural processes and samples collected during rainy season, and the second component (15.42%) was related mainly to agricultural and urban effluent discharges and with samples collected during dry season (positive loadings). Hierarchical cluster analysis revealed the existence of five groups, of which three were characterized as regions of low, moderate, and high impact in the basin respectively. The other two sites, Marzagão and Porto Cercado, were considered unique groups. Marzagão is located in the headwaters of the Cuiabá River and is not subject to flow regulation caused by a reservoir, and Porto Cercado is within the flood plain of Pantanal. These data will contribute to the optimization of monitoring programs and to the understanding of other similar watersheds.

Key words: Principal component analysis, environmental monitoring, pollution, watershed, hierarchical cluster analysis.