Application of statistical models to estimate total dissolved solids in acid mine drainage

Abstract

The present study presents an unexplored statistical approach using linear and multiple regression to estimate total dissolved solids in acid mine waters. One hundred twenty-one samples were collected from three distinct abandoned mining areas in the Iberian Peninsula: Valdarcas, São Domingos, and Campanario. The parameters analyzed included pH, electrical conductivity, total dissolved solids, acidity, and sulfate concentration. Discriminant analysis and linear regression were used to identify the best predictor variables of total dissolved solids. The results showed a high variability between the mining areas due to the hydrochemical characteristics at each site. The most relevant variables differentiating the three mines are sulfate, acidity, and electrical conductivity. However, for these specific samples, sulfate and acidity proved to be the most relevant parameters for estimating total dissolved solids.