Application of fuzzy logic tools for the biogeochemical characterisation of (un)contaminated waters from Aljustrel mining area (South Portugal)

Abstract

Aljustrel mining area (South Portugal) belongs to the Iberian Pyrite Belt (IPB). It is classified of high environmental risk due to its large tailings and to the Acid Mine Drainage (AMD) affected waters, generated by sulphides' oxidation. Integrating biological parameters (for the first time) in the input data matrix of the software PreFuRGe, allowed a better discrimination of the diatoms' responses to the stimuli caused by the hydrochemical changes imposed by the processes affecting water quality. Each hydrochemical scenario, was modeled by imposing maximum and minimum limits for each antecedent, according to the conditions imposed by the consequent, which in this case were the number of diatom species and pH. Thus, PreFuRGe evidenced some qualitative aspects that could not be achieved by classic statistics. pH appeared as the main discriminator of diversity and diatom species composition, nevertheless and due to the complex environment under study other chemical interactions must be considered: (a) AMD waters, with extremely low pH values, but also with extremely high hydrogeochemical complexity, represented by a mixture of metals, do not allow to associate, unequivocally, the reduction in diatom diversity to pH, but also to high metal (loid)s concentrations; (b) in the most alkaline waters, with higher abundance of diatom species, average to high concentrations of Na and Cl (due to Cenozoic sediments) do not seem to affect diatom diversity. This methodology proved to be an efficient tool to establish, for the first time, cause-effect relationships, improving the comprehension between biological (diatoms) and hydrochemical parameters.