pH, electric conductivity and sulfate as base parameters to estimate the concentration of metals in AMD using a fuzzy inference system

An adaptive neuro-fuzzy inference system (ANFIS) was developed to model an acid mine drainage (AMD) affected river. The inference system uses three AMD relatively expeditious indicators (pH, electric conductivity and sulfate) and gives, as response, concentration values for Fe, Mn, Cu, Zn, Cd, and As. It was based on a large database, which resulted from monitoring an AMD stream located in the Iberian Pyrite Belt (Chorrito stream, which flows into the Cobica River, SW Spain). The results indicate a high correlation between the analyzed values and the concentrations obtained by fuzzy estimation. Therefore, this modeling approach afforded a simple and cost-effective system capable of monitoring the affected river, since it avoids or simplifies the analyses of the most expensive and time consuming chemical parameters.

► ANFIS based neuro-fuzzy system to model an acid mine drainage (AMD) affected river
► This method is an expeditious procedure to deduce metal loads.
► This tool is effective to optimize the laboratory routines.
► The approach may be applied to monitoring other aquatic systems.