Multivariate statistical analysis applied in samples of groundwater in the region of Cariri, south of the State of Ceará

Abstract

In the area constituted by the municipalities of Barbalha, Crato and Juazeiro do Norte, south of Ceará State, groundwater resources are the only source of public and private water supply, and overexploitation is already evident in the intermittency of the old perennial drainage network, now perpetuated by sewage launches. The chemical evolution of groundwater was studied by identifying of similarities between physicochemical variables, discriminating the most significant ones, through the methodology of multivariate statistical analysis to explain possible processes responsible for water quality in the area. The factorial analysis identified three components responsible for approximately 85% of the total variance; the first one as an indicator of the dissolution of the minerals present in the Rio da Batateira and Santana formations; the second and third factors as indicators of a possible mixture between the natural waters of the aquifer with polluted waters. Multivariate cluster analysis by component showed four groups in factor 1 and two groups in factors 2 and 3, identified by samples with different concentration ranges. In the discriminant analysis, the most significant variables in the processes responsible for the hydrochemical evolution of the area were Mg²⁺ and K⁺ , being able to be related to ion exchange processes and weathering of feldspars.