Avaliação da qualidade da água da sub-bacia do Rio Cuiabá-MT aplicando análise multivariada

Abstract

Considering complex databases, which contain values and information from various chemical, physical and biological parameters, the selection of parameters which reflect the influence of natural factors and human activities on water resources is fundamental to the definition of environmental indicators by the competent governing bodies. In this context, this study aimed at the selection of the main attributes (physical, chemical and microbiological parameters) which represent the quality of the water in the sub-basin of the Cuiabá River, as well as to evaluate the spatial and temporal variations using the multivariate statistical analysis techniques. In order to achieve this goal, water samples were collected bimonthly at 13 stations along the Cuiabá River, from July 2010 to May 2011. Eleven pesticides were derminated in samples collected in 8 stations. Thirty-two physical, chemical and biological parameters were determinated. In some months, the variables color, Escherichia coli, dissolved oxygen, total phosphorus, turbidity and pH were found outside the limits established by CONAMA Resolution 357/2005 for water bodies of class 2, especially in the stations close to the urban areas. Among the studied pesticides, metolachlor and flutriafol were detected and quantified. Factor Analysis/Principal Component Analysis (FA/PCA), using the orthogonal rotation - Varimax method, indicate that, in the best model to express the water quality in the Cuiabá River basin, the parameters have been aggregated into seven factors, explaining 76.11% of the total variance of the original variables. The first two factors together explain about 44.5% of the total variance. The first factor (29.08%) represents the component ions dissolved in water (higher correlation for calcium, potassium, total hardness, conductivity, salinity and total dissolved solids), thus expressing the natural process of weathering of geological components of the soil in the region, characterized by regions containing limestone (mainly in the upper course). The second factor (15.42%) is defined by a component associated with human activities which produce diffuse pollution (agricultural and urban) and with point sources of effluent discharge. Samples showed a wide variation in color and in turbidity parameters, mainly influenced by the carrying of sediments, which occur mostly during the rainy season, and aggravated by the processes of human occupation in the basin. The analysis indicated that it is possible to reduce from 28 to 19 parameters the number of analyzed variables in the sub-basin, which means a significant reduction. The hierarchical cluster analysis revealed the existence of five groups of stations that exhibit homogeneous characteristics. Marzagão and Porto Cercado stations presented high degree of dissimilarity among the other stations and were considered unique clusters. Three other homogeneous clusters were identified, characterizing regions of low, moderate and high impaction in the basin, respectively.