Model of behaviour of conductivity versus pH in acid mine drainage water, based on fuzzy logic and data mining techniques

Abstract

Acid Mine Drainage is a water pollution type characterized by several topics such as high acidity, sulfate and heavy metal concentrations. One of the chemical characteristics is the absence of correlation between pH and conductivity, as could be expected. This last parameter is well correlated with other variables such as sulfate concentration and can be used as a field assessment. The absence of pH/conductivity correlation is largely discussed by several authors. In this work, the use of fuzzy logic algorithms in a large temporal database (over 20,000 records) has allowed us to study the "hidden" relation between them. This work finds this correlation, with some conditions such as the range of pH where it happens. Maybe the study of the usual range of pH values in previous studies has disturbed the correlation because of other chemical processes.

Tinto Versus Odiel: Two A.M.D. Polluted Rivers and an Unresolved Issue. An Artificial Intelligence Approach

“…The statistical summary of the variables shows values which are characteristic of channels undergoing AMD processes within the described hydrological and mining context (Sáinz et al 2002(Sáinz et al , 2004Sánchez-Rodas et al 2005;Grande et al 2000Grande et al , 2003aNieto et al 2007;Cánovas et al 2008;de la Torre et al 2009;Jiménez et al 2009;Sarmiento et al 2009;Grande et al 2010b). The worsening of the water quality in the Tinto and Odiel rivers is the result of the intense mining activity of sulfides for over 2,000 years.…”

“…The Iberian Pyrite Belt is an extremely important AMD-producing source which has been widely described in the past years (Grande et al 2005b;España et al 2005;Nieto et al 2007;de la Torre et al 2009;Jiménez et al 2009;Sarmiento et al 2009). Sulfate, metal and acidity total input to the Atlantic Ocean as described by Sáinz (1991) amounts to 1,200,000 t/year of sulfate and 20,000 t/year of heavy metals, which gives an idea of the magnitude of the phenomenon.…”

Relationships Between Conductivity and pH in Channels Exposed to Acid Mine Drainage Processes: Study of a Large Mass of Data Using Classical Statistics

“…Rivers affected by this kind of contamination are characterized by their acidity and also by the high sulfate and heavy metal content in their waters and metal content in their sediments (Nebel and Wright 1999). The pollutant effluent from acid mines is complex and is characterized by elevated concentrations of iron and sulfate, low pH, and elevated concentrations of a wide variety of metals (Gray 1998 In river environments exposed to AMD processes, and largely described in the hydrological scene by Borrego et al (2002), Davis et al (2000), Leblanc et al (2000), Sáinz et al (2002Sáinz et al ( , 2003a, Grande et al (2000Grande et al ( , 2003Grande et al ( , 2005a, Bednar et al (2005), Neal et al (2005, Jiménez et al (2009, pH values reach extraordinarily low values of around 2, while conductivity is around 3 mS/cm.…”

Hydrochemical characterization of an acid mine drainage-affected reservoir: the Sancho Reservoir, Huelva, southwest Spain

“…The IPB is one of the largest metallogenic regions of the world, with numerous giant and supergiant massive sulphide ore deposits (Sáez et al 1999), comprising 1700 Mt of sulphide reserves (Pinedo-Vara 1963). Consequently, the regional drainage network is strongly affected by AMD and acid rock drainage (ARD), as described by several authors: Elbaz-Poulichet et al (1999,2000,2001), Davis et al (2000), Grande et al (2003aGrande et al ( , 2003bGrande et al ( , 2005aGrande et al ( , 2005bGrande et al ( , 2010aGrande et al ( , 2010bGrande et al ( , 2010cGrande et al ( , 2010dGrande et al ( , 2010e, 2011aGrande et al ( , 2011b, Leblanc et al (2000), Borrego et al (2002), Sáinz et al (2002Sáinz et al ( , 2003aSáinz et al ( , 2003bSáinz et al ( , 2004Sáinz et al ( , 2005, Olıás et al (2004Olıás et al ( , 2005aOlıás et al ( , 2005bOlıás et al ( , 2006, Aroba et al (2007), Jiménez et al (2009), de la Torre et al (2009, Carro et al (2011) and Grande (2011).…”