info:eu-repo/semantics/article
Geochemical behavior of an acid drainage system: the case of the Amarillo River, Famatina (La Rioja, Argentina)
Fecha
2017-01Registro en:
Lecomte, Karina Leticia; Maza, Santiago; Collo, Gilda; Sarmiento, A. M.; Depetris Gallino, Pedro Jose; Geochemical behavior of an acid drainage system: the case of the Amarillo River, Famatina (La Rioja, Argentina); Springer Heidelberg; Environmental Science and Pollution Research; 24; 2; 1-2017; 1630-1647
0944-1344
1614-7499
CONICET Digital
CONICET
Autor
Lecomte, Karina Leticia
Maza, Santiago
Collo, Gilda
Sarmiento, A. M.
Depetris Gallino, Pedro Jose
Resumen
The Amarillo River (Famatina range, Argentina, ~29° S and ~67° W) is unusual because acid mine drainage (AMD) is superimposed on the previously existing acid rock drainage (ARD) scenario, as a Holocene paleolake sedimentary sequence shows. In a markedly oxidizing environment, its water is currently ferrous and of the sulfate-magnesium type with high electrical conductivity (>10 mS cm−1 in uppermost catchments). At the time of sampling, the interaction of the mineralized zone with the remnants of mining labors determined an increase in some elements (e.g., Cu ~3 to ~45 mg L−1; As ~0.2 to ~0.5 mg L−1). Dissolved concentrations were controlled by pH, decreasing significantly by precipitation of neoformed minerals (jarosite and schwertmannite) and subsequent metal sorption (~700 mg kg−1 As, 320 mg kg−1 Zn). Dilution also played a significant role (i.e., by the mixing with circumneutral waters which reduces the dissolved concentration and also enhances mineral precipitation). Downstream, most metals exhibited a significant attenuation (As 100 %, Fe 100 %, Zn 99 %). PHREEQC-calculated saturation indices (SI) indicated that Fe-bearing minerals, especially schwertmannite, were supersaturated throughout the basin. All positive SI increased through the input of circumneutral water. PHREEQC inverse geochemical models showed throughout the upper and middle basin, that about 1.5 mmol L−1 of Fe-bearing minerals were precipitated. The modeling exercise of mixing different waters yielded results with a >99 % of correlation between observed and modeled data.