Bull Environ Contam Toxicol DOI 10.1007/s00128-015-1473-0
Heavy Metals in Waters and Suspended Sediments Affected by a Mine Tailing Spill in the Upper San Lorenzo River, Northwestern Me´xico F. Pa´ez-Osuna • H. Bojo´rquez-Leyva • M. Berge´s-Tiznado O. A. Rubio-Herna´ndez • J. F. Fierro-San˜udo • J. Ramı´rez-Rochı´n • J. A. Leo´n-Can˜edo
•
Received: 10 August 2014 / Accepted: 22 January 2015 Ó Springer Science+Business Media New York 2015
Abstract Concentrations of arsenic (As), cadmium (Cd), copper (Cu), lead (Pb), mercury (Hg), silver (Ag) and zinc (Zn) were evaluated in water and suspended sediments of the upper waters of San Lorenzo River in NW Mexico following a mine tailing spill. Except As (6.64–35.9 lg L-1), dissolved metal concentrations were low (Ag \0.06–0.22; Cd 0.01–0.34; Cu 4.71–10.2; Hg 0.02–0.24; Pb\0.15–0.65; Zn 86–1,080 lg L-1) and were less than the upper limits established by UNEP (Water quality for ecosystem and human health, 2nd edn. United Nations Environment Programme Global Environment Monitoring System/Water Programme, Burlington, 2008), EPA (2014) and the Mexican regulation (NOM 1994). In contrast, the suspended metal concentrations were high (As 91.4–130; Ag 22.1–531; Cd 3.14–6.30; Cu 65–123; Hg 0.47–1.09; Pb 260–818; Zn 742–1,810 mg kg-1) and most of samples exceeded the probable effect level of the Canadian Sediment Quality Guidelines for the Protection of Aquatic Life. F. Pa´ez-Osuna (&) H. Bojo´rquez-Leyva Unidad Acade´mica Mazatla´n, Instituto de Ciencias del Mar y Limnologı´a, Universidad Nacional Auto´noma de Me´xico, Joel Montes Camarena s/n, 82040 Mazatla´n, Sinaloa, Mexico e-mail: [email protected] F. Pa´ez-Osuna Miembro de El Colegio de Sinaloa, Culiaca´n, Sinaloa, Mexico M. Berge´s-Tiznado O. A. Rubio-Herna´ndez J. F. Fierro-San˜udo J. A. Leo´n-Can˜edo Posgrado en Ciencias del Mar y Limnologı´a, Universidad Nacional Auto´noma de Me´xico, Joel Montes Camarena s/n, 82040 Mazatla´n, Sinaloa, Mexico J. Ramı´rez-Rochı´n Posgrado en Ciencias en Recursos Acua´ticos, Facultad de Ciencias del Mar, Universidad Auto´noma de Sinaloa, Culiaca´n, Sinaloa, Mexico
Keywords Arsenic Mercury Silver Lead Tailing spill NW Mexico
Increased concentrations of contaminant metals in fluvial waters, soils and sediments as a result of the mining and processing of metal ores have been widely reported in river watersheds (e.g., Villarroel et al. 2006; Cabrera et al. 2008). The release of contaminant metals into river environments within mining-affected river systems generally results from (Bird et al. 2010): acid mine drainage; the release of waste slurries that contain solute and particulateassociated metals; dumping of mining waste that is subsequently leached or dispersed downstream; and the occurrence of mine tailings dam failures. The frequency of spills associated with mine tailings dam failures is surprisingly high; e.g., Villarroel et al. (2006) reported that during the period 1970–2003, there were at least 59 major metal-related tailings impoundments that failed worldwide, with a volume of material released ranging from 5,500 to 1,300,000 m3. One such failure occurred on January 21, 2013, at the ‘‘El Herrero’’ processing plant in the Santa Maria de Otae´z mining region (Durango, Mexico) located approximately 125 km from the upper San Lorenzo River in Sinaloa NW Mexico (Fig. 1). This plant has an output of 1,500 t/day; the deposit is a type of vein, whose mineralogy is characterized by the presence of Au, Ag, Pb, Zn and Cu (SGM 2011). During the failure of the mine tailings dam, a volume of approximately 300,000 m3 of waste entered the Los Remedios River, from which the waste material was transported through the upper San Lorenzo River to the dam known as El Comedero. This spill event has the potential to negatively affect both ecological and human health over a considerable distance. However, in this context, geochemical data defining the magnitude of
123
Bull Environ Contam Toxicol
Fig. 1 Map of the spill-affected zone along the Los Remedios River (RR)-San Lorenzo River (SL)-El Comedero dam (CM), and location of the sampling sites (unfilled circles). TD is the location of the discharge point where the mine tailings dam failure occurred
contamination resulting from this spill is nonexistent. Thus, the first study on the composition of the waters was initiated in June of 2013, after various fishermen from the Las Can˜adas de Jacopa cooperative requested help. The purpose of this study was to evaluate the levels of As, Ag, Cd, Cu, Hg, Pb and Zn in the waters of the tributary river Los Remedios, the upper San Lorenzo River, and the El Comedero dam just 3 months after the spill event.
Materials and Methods The El Comedero dam is located in the Cosala´ municipality in Sinaloa state, Mexico. This dam receives waters from the upper San Lorenzo River, which is formed in the Sierra Madre Occidental at Durango State with water contributions from three tributaries: San Juan Camarones, San Gregorio and, most importantly, Los Remedios River (Fig. 1). The El Comedero dam has a total water capacity of 3,966.2 9 106 m3 and covers an area of 9,000 ha. This dam was constructed to control the flow of the San Lorenzo River for harnessing its waters to irrigate approximately 105,000 ha of land in the San Lorenzo and Culiacan valleys and to generate electricity. The climate in the region is warm sub-humid with summer rains (800–1,000 mm), and the average annual temperature is ´ lvarez et al. 2012). 24–26°C (Beltra´n-A A total of 16 surface water samples were collected on June 18, 2013, from (a) the Los Remedios River, which is the main tributary of the San Lorenzo River (three duplicates were collected from sites RR1, RR2 and RR3);
123
(b) upstream of the San Lorenzo River (two duplicates were collected from sites SL4 and SL5); and (c) from the El Comedero dam (three duplicates were collected from sites CD6, CD7 and CD8) (Fig. 1). The water samples were collected using 3-L polyethylene bottles that had previously been acid-washed (2 M HNO3) and rinsed with Milli-Q purified water. Before sampling, the bottles were rinsed two times with river water, and then two 3-L samples were collected at 10 min intervals as close as possible to the main flow and 10–20 cm below the surface. The samples were maintained on ice during transport to the laboratory, where they were filtered through 0.45 lm acidwashed (0.5 M HCl) filters (HA, Millipore). The filtrates were acidified (HNO3 trace metal analysis, J. T. Baker) and used for analysis of the dissolved fraction. The settled and suspended material was obtained from the water samples following a 72 h standing period. This material was then lyophilized (-49°C and 132 9 10-3 mbar for 72 h) and ground. The suspended sediment samples (0.25 g dry weight) were digested with 5 mL of concentrated acids (3 volumes HNO3: 1 volume HCl: 1 volume HF) in capped Teflon (Savillex) vials (UNEP 1995). The mixture was heated at 140°C for 3 h. The samples were diluted with Milli-Q water to a final volume of 25 mL. Blank samples and the standard reference material BCSS-1 were digested (one set) using the same procedure to determine the accuracy and precision. The digested samples were stored in polyethylene containers for further analysis. Depending on the concentration level, analyses of Cd, Cu and Zn were
Bull Environ Contam Toxicol
performed using flame or graphite-atomic absorption spectrometry; Hg determinations were performed using cold-vapor atomic absorption spectrophotometry. The measurements were performed using a Varian SpectrAA 220 instrument. The recovery values from the analysis of the standard reference material BCSS-1 (UNEP 1995) were: As 94.1, Cd 91.6, Cu 101.1, Hg 85.4, Pb 101.7, and Zn 90.3 % (Ag was not certified). The precision, as variation coefficient, varied from 5.0 to 11.2 %. The mean concentrations determined in each river and the dam were compared by one-way repeated measures ANOVA with a p \ 0.05 followed by Tukey´s multiple comparison test (Zar 1999).
Results and Discussion The concentrations of metals measured in water samples from the El Comedero dam and associated rivers are shown in Fig. 2. Data for downstream of the San Lorenzo River (sampled during 2009–2010) were taken from Frı´as-Espericueta et al. (2014). The concentrations of metals in the dissolved fractions from the ecosystems studied here were low likely due to adsorption by suspended sediments, complexation by suspended species and by low solubility of elements in aquatic environments. Dissolved concentrations (in lg L-1) in water ranged from \0.065 to 0.218 for Ag, 6.64 to 35.9 for As, 0.015 to 0.338 for Cd, 4.71 to 10.2 for Cu, 0.019 to 0.242 for Hg, \0.15 to 0.65 for Pb, and 86 to 1,080 for Zn. Figure 2 (upper row) shows the dissolved levels of the analyzed elements as a function of distance; Cd and Ag, although irregularly, tend to increase from the Los Remedios River to downstream of the San Lorenzo River; similarly, Zn exhibited the same tendency, but only until El Comedero dam. The concentrations of As, Hg, Pb and Cu exhibited a tendency to decrease with an irregular pattern. Considering the mean concentrations, the order (highest to lowest) was Zn [ As [ Cu [ Pb [ Cd [ Ag [ Hg. With the exception of As (5 of 8 sites), these concentrations were relatively low and were less than the upper limit value for drinking water established by the World Health Organization (UNEP 2008): As (10 lg L-1), Cd (3.0 lg L-1), Cu (2,000 lg L-1), Hg (1.0 lg L-1), Pb (10 lg L-1), and Zn (3,000 lg L-1). Similarly, these concentrations were less than the maximum contaminant levels established by the Mexican Official Norm (NOM 1994): As (50 lg L-1), Cd (5.0 lg L-1), Cu (2,000 lg L-1), Hg (1.0 lg L-1), Pb (25 lg L-1), and Zn (5,000 lg L-1). Waters from the San Lorenzo River and El Comedero dam are used for drinking, irrigation for agriculture, livestock and fish culture. Considering the maximum safe concentrations proposed for the culture of warm-
water species (Boyd 2009), Ag (0.05–0.10 lg L-1), Cd (25–70 lg L-1), Cu (25–70 lg L-1), Hg (0.1 lg L-1), Pb (200–4,000 lg L-1) and Zn (50–100 lg L-1), the waters from the El Comedero dam are, with respect to Cd, Pb and Cu, suitable for fish farming. However, considering Hg, and particularly Ag and Zn, these waters appear to be unsuitable. In general, the concentrations of dissolved Hg found in this study are comparable to or within the range reported for small streams in Europe (\0.001–1.08 lg L-1) (Protano et al. 2014), Indonesia (0.08–0.22 lg L-1) (Yasuda et al. 2011) and China (\0.01–0.69 lg L-1) (Lin et al. 2012). The Ag levels found in this study were below or close to the mean dissolved Ag level reported in major unpolluted world rivers (0.3 lg L-1) (Martı´n and Meybeck 1979), which indicates that the influence of the spill on the dissolved Ag was not significant. In contrast the As concentrations were within of ranges for affected rivers from polluted European rivers (4.5–45 lg L-1) and for rivers mining influenced (\0.2–7,900 lg L-1) (Plant et al. 2005). Similarly, the dissolved concentrations of Cd, Cu and Pb were comparable to and within the ranges found in rivers from the Danube and Marisa drainage basins (Cd, \0.3–9.4 lg L-1; Cu, \0.8–250 lg L-1; Pb, \0.3–200 lg L-1), with the exception of the Zn levels, which were two times higher than those from Eastern Europe (Zn,\1–460 lg L-1) (Bird et al. 2010). However, when the levels are compared to the maximum levels found in streams from southern Ecuadorian rivers impacted by gold mining (Carling et al. 2013) or miningaffected tributary streams from the Danube basin (Bird et al. 2010), our values are low. The available information on metal concentrations in the rivers of NW Mexico is relatively scarce. In waters from the San Pedro River (Cananea mining district), which is affected by drainage from mine tailings, extremely variable ´ lvarez et al. 2009): 10–63,950 levels were found (Gomez-A for Cd, 60–1,200,000 for Cu, \2–75,000 for Pb, and 10–54,000 lg L-1 for Zn. Frı´as-Espericueta et al. (2014) examined eight rivers in the downstream portion of Sinaloa state (including the San Lorenzo River), in which the levels of the dissolved fraction were higher for Cd (0.32–1.06 lg L-1) and Pb (3.43–8.27 lg L-1) and lower for Cu (1.36–2.10 lg L-1) and Zn (8.0–16.9 lg L-1) than the levels found in this study. This rough comparison reveals that dissolved Cu and Zn are enriched in the waters of the El Comedero dam, upstream of the San Lorenzo River and in the Los Remedios River. This situation is more clearly evidenced by comparing the waters from the downstream versus the upstream region of the San Lorenzo River; although the levels of Cd are comparable in both waters, the levels of Zn and Cu are 77 and four times greater in the upper waters than in the lower waters. It is interesting the contrary behavior of Pb, the waters are around 14 times greater in the lower waters than in the
123
Bull Environ Contam Toxicol
Fig. 2 Concentrations of Ag, As, Cd, Cu, Hg, Pb and Zn in water samples from El Comedero dam and associated rivers: dissolved (top row) (lg L-1) and suspended fractions (bottom row) (lg g-1). Sampling sites are from Fig. 1; SLD corresponds to a site in the San
123
Lorenzo River 60 km downstream of El Comedero dam that was analyzed by Frı´as-Espericueta et al. (2014). An asterisk indicates a concentration that exceeds the PEL for sediments (CCME 2001) or the WHO for water (UNEP 2008)
Bull Environ Contam Toxicol
upper waters. Results from an isotopic study of Pb in the region (Soto-Jime´nez et al. 2008), revealed that the Pb contained in environmental matrices is mostly derived from gasoline (in use until 1997) and US emissions (aerosols from industrial activity), while natural Pb weathered from mother rock represented a minimal contribution. In contrast to the dissolved fraction, concentrations of metals in suspended sediments were relatively high. The concentrations (in mg kg-1) ranged from 22.1 to 531 for Ag, 91.4 to 130 for As, 3.14 to 6.30 for Cd, 65 to 123 for Cu, 0.47 to 1.09 for Hg, 260 to 818 for Pb and 742 to 1,810 for Zn. Considering the mean concentrations, the order of the decrease was Zn [ Pb [ Ag [ As [ Cu [ Cd [ Hg. Unlike the dissolved fraction, the seven elements in the suspended sediments exhibited a gradient that was enriched from the upper sites (RR1 or RR2) through upstream of the San Lorenzo River, the El Comedero dam to downstream of the San Lorenzo River (Fig. 2, bottom row). The concentrations of suspended Cd, Cu, Pb and Zn were higher in comparison to those from downstream waters of the San Lorenzo River (SLD) (Frı´as-Espericueta et al. 2014). As, Ag and Hg data were not available for SLD. In general, the levels of all elements in the suspended sediments from the rivers and dam were relatively high. Particularly, the concentration of As (100 % of samples), Pb (100 %), Zn (100 %), Cd (87.5 %) and Hg (62.5 %) exceeded the PEL (probable effect level) established by the Canadian Sediment Quality Guidelines for the Protection of Aquatic Life. Although Cu clearly showed a concentration gradient from the upstream waters until the dam and the downstream waters, such levels did not exceed the PEL values. The PEL represents the lower limit of the range of chemical concentrations that are usually or always associated with biological effects (CCME 2001). The PEL value for Ag is not available. However, the Ag levels found were comparable or higher than those registered for unpolluted major world rivers (\0.3–60 mg/kg-1) (Martı´n and Meybeck 1979); 62.5 % of the samples exceeded the upper range value. The metal concentrations in the suspended sediments found here were comparable to or greater than, depending on the metal, those from West Java Indonesia (Yasuda et al. 2011). Levels of Hg were similar, while levels of Cd, Cu and Zn were higher than those reported in the Cikaniki River, which is affected by gold mining. However, our results are low compared to those from watersheds influenced by metal mining activities in regions of Europe or South America. In the Danube, levels as high as 33 mg kg-1 for Cd, 8,088 mg kg-1 for Cu, 1,700 mg kg-1 for Pb, and 2,010 mg kg-1 for Zn have been reported (Bird et al. 2010). In Ecuadorian rivers impacted by gold mining, concentrations as high as 46,049 mg kg-1 for As, 104 mg kg-1 for Cd,
9,134 mg kg-1 for Cu, 34 mg kg-1 for Hg, 6,646 mg kg-1 for Pb, and 9,792 mg kg-1 for Zn have been reported (Carling et al. 2013). It is generally assumed that differences between metal concentrations in the water and sediment samples are associated with long-term exposure in sediments, whereas metal concentrations in water indicate recent contamination (Protano et al. 2014). From this scenario it is possible to argue for this case that a few days after the spill the dissolved metals were most enriched with those metals liberated from the tailings. Subsequently, upstream rivers were affected by medium or long-term metal pollution associated with the suspended material from the tailings. Importantly, the dissolved concentrations represent the environmental health risk from each of the elements at the sampling time (or after few days), whereas the suspended sediment metal concentrations provide information about the risks over an extended period of time, which could be of months or years. Acknowledgments G. Ramı´rez-Rese´ndiz for the preparation of figures. This work was supported by the Universidad Nacional Autono´ma de Me´xico, PAPIIT IN210609 and PAPIIT IN208813-2.
References ´ lvarez R, Sa´nchez-Palacios J, Farı´as-Sa´nchez A, Ramı´rezBeltra´n-A Lozano JP (2012) Aspectos reproductivos de la lobina negra (Micropterus salmonides) en la presa Jose´ Lo´pez-Portillo, El Comedero, Sinaloa, Me´xico. Ciencia Pesquera 20:65–75 Bird G, Brewer PA, Macklin MG, Nikolova M, Kotsev T, Mollov M, Swain C (2010) Dispersal of contaminant metals in the miningaffected Danube and Maritsa drainage basins, Bulgaria, Eastern Europe. Water Air Soil Pollut 206:105–127 Boyd CE (2009) Trace metal toxic at high concentrations. Glob Aquacult Adv Aug pp 24–26 Cabrera F, Ariza J, Madejo´n P, Murillo JM (2008) Mercury and other trace elements in soils affected by the mine tailing spill in Aznalco´llar (SW Spain). Sci Total Environ 390:311–322 Carling GT, Dı´az X, Ponce M, Perez L, Nasimba L, Pazmino E, Rudd A, Merugu S, Ferna´ndez DP, Gale BK, Johnson WP (2013) Particulate and dissolved trace element concentrations in three Southern Ecuador rivers impacted by artisanal gold mining. Water Air Soil Pollut 224:1–16 CCME (2001) Canadian Council of Ministers of the Environment. Canadian sediment quality guidelines for the protection of aquatic life. http://www.ccme.ca/sourcetotap/wqi.html. Accessed 30 May 2014 EPA (2014) United States Environmental Protection Agency. Water: drinking water contaminants. http://water.epa.gov/drink/contami nants. Accessed 30 May 2014 Frı´as-Espericueta MG, Mejı´a-Cruz R, Osuna-Lo´pez I, Muy-Rangel MD, Rubio-Carrasco W, Aguilar-Jua´rez M, Voltolina D (2014) Metal discharges by Sinaloa rivers to the coastal zone of NW Mexico. Bull Environ Contam Toxicol 92:132–136 ´ lvarez A, Meza-Figueroa D, Villalba-Atondo AI, ValenzuGomez-A ela-Garcı´a JL, Ramı´rez-Herna´ndez J, Almendariz-Tapia J (2009) Estimation of potential pollution from mine tailings in the San Pedro River (1993–2005), Mexico–US border. Environ Geol 57:1469–1479
123
Bull Environ Contam Toxicol Lin Y, Vogt R, Larssen T (2012) Environmental mercury in China: a review. Environ Toxicol Chem 31:2431–2444 Martı´n JM, Meybeck M (1979) Elemental mass-balance of material carried by major world rivers. Mar Chem 7:173–206 NOM (1994) Norma Oficial Mexicana NOM-127-SSAI-1994. www. salud.gob.mx/unidades/edi/nom/127ssa14.html. Accessed 30 May 2014 Plant JA, Kinniburgh DG, Smedley PL, Fordyce FM, Klinck BA (2005) Arsenic and selenium. In: Shewood Lollar B (ed) Environmental geochemistry. Elsevier, San Diego, pp 17–66 Protano CL, Zinna S, Giampaoli V, Romano S, Chiavarini S, Vitali M (2014) Heavy metal pollution and potential ecological risks in rivers: a case study from southern Italy. Bull Environ Contam Toxicol 92:75–80 SGM (2011) Panorama Minero del Estado de Durango. Septiembre 2011. Servicio Geolo´gico Mexicano. www.sgm.gob.mx. Accessed 30 May 2014 Soto-Jime´nez MF, Pa´ez-Osuna F, Scelfo G, Hibdon S, Franks R, Aggarawl J, Flegal AR (2008) Lead pollution in subtropical
123
ecosystems on the SE Gulf of California coast: a study of concentrations and isotopic composition. Mar Environ Res 66:451–458 UNEP (1995) Manual for the geochemical analyses of marine sediments and suspended particulate matter. Reference Methods for Marine Pollution Studies No. 63, UNEP UNEP (2008) Water quality for ecosystem and human health, 2nd edn. United Nations Environment Programme Global Environment Monitoring System/Water Programme, Burlington 120 pp Villarroel LF, Miller JR, Lechler PJ, Germanoski D (2006) Lead, zinc, and antimony contamination of the Rio Chilco-Rio Tupiza drainage system, Southern Bolivia. Environ Geol 51:283–299 Yasuda M, Yustiawati MS, Syawal MT, Sikder T, Hosokawa T, Saito S, Tanaka M, Kurasaki M (2011) Metal concentrations of river water and sediments in west Java, Indonesia. Bull Environ Contam Toxicol 87:669–673 Zar JH (1999) Biostatistical analysis, 4th edn. Prentice Hall, New Jersey
Heavy Metals in Waters and Suspended Sediments Affected by a Mine Tailing Spill in the Upper San Lorenzo River, Northwestern Me´xico F. Pa´ez-Osuna • H. Bojo´rquez-Leyva • M. Berge´s-Tiznado O. A. Rubio-Herna´ndez • J. F. Fierro-San˜udo • J. Ramı´rez-Rochı´n • J. A. Leo´n-Can˜edo
•
Received: 10 August 2014 / Accepted: 22 January 2015 Ó Springer Science+Business Media New York 2015
Abstract Concentrations of arsenic (As), cadmium (Cd), copper (Cu), lead (Pb), mercury (Hg), silver (Ag) and zinc (Zn) were evaluated in water and suspended sediments of the upper waters of San Lorenzo River in NW Mexico following a mine tailing spill. Except As (6.64–35.9 lg L-1), dissolved metal concentrations were low (Ag \0.06–0.22; Cd 0.01–0.34; Cu 4.71–10.2; Hg 0.02–0.24; Pb\0.15–0.65; Zn 86–1,080 lg L-1) and were less than the upper limits established by UNEP (Water quality for ecosystem and human health, 2nd edn. United Nations Environment Programme Global Environment Monitoring System/Water Programme, Burlington, 2008), EPA (2014) and the Mexican regulation (NOM 1994). In contrast, the suspended metal concentrations were high (As 91.4–130; Ag 22.1–531; Cd 3.14–6.30; Cu 65–123; Hg 0.47–1.09; Pb 260–818; Zn 742–1,810 mg kg-1) and most of samples exceeded the probable effect level of the Canadian Sediment Quality Guidelines for the Protection of Aquatic Life. F. Pa´ez-Osuna (&) H. Bojo´rquez-Leyva Unidad Acade´mica Mazatla´n, Instituto de Ciencias del Mar y Limnologı´a, Universidad Nacional Auto´noma de Me´xico, Joel Montes Camarena s/n, 82040 Mazatla´n, Sinaloa, Mexico e-mail: [email protected] F. Pa´ez-Osuna Miembro de El Colegio de Sinaloa, Culiaca´n, Sinaloa, Mexico M. Berge´s-Tiznado O. A. Rubio-Herna´ndez J. F. Fierro-San˜udo J. A. Leo´n-Can˜edo Posgrado en Ciencias del Mar y Limnologı´a, Universidad Nacional Auto´noma de Me´xico, Joel Montes Camarena s/n, 82040 Mazatla´n, Sinaloa, Mexico J. Ramı´rez-Rochı´n Posgrado en Ciencias en Recursos Acua´ticos, Facultad de Ciencias del Mar, Universidad Auto´noma de Sinaloa, Culiaca´n, Sinaloa, Mexico
Keywords Arsenic Mercury Silver Lead Tailing spill NW Mexico
Increased concentrations of contaminant metals in fluvial waters, soils and sediments as a result of the mining and processing of metal ores have been widely reported in river watersheds (e.g., Villarroel et al. 2006; Cabrera et al. 2008). The release of contaminant metals into river environments within mining-affected river systems generally results from (Bird et al. 2010): acid mine drainage; the release of waste slurries that contain solute and particulateassociated metals; dumping of mining waste that is subsequently leached or dispersed downstream; and the occurrence of mine tailings dam failures. The frequency of spills associated with mine tailings dam failures is surprisingly high; e.g., Villarroel et al. (2006) reported that during the period 1970–2003, there were at least 59 major metal-related tailings impoundments that failed worldwide, with a volume of material released ranging from 5,500 to 1,300,000 m3. One such failure occurred on January 21, 2013, at the ‘‘El Herrero’’ processing plant in the Santa Maria de Otae´z mining region (Durango, Mexico) located approximately 125 km from the upper San Lorenzo River in Sinaloa NW Mexico (Fig. 1). This plant has an output of 1,500 t/day; the deposit is a type of vein, whose mineralogy is characterized by the presence of Au, Ag, Pb, Zn and Cu (SGM 2011). During the failure of the mine tailings dam, a volume of approximately 300,000 m3 of waste entered the Los Remedios River, from which the waste material was transported through the upper San Lorenzo River to the dam known as El Comedero. This spill event has the potential to negatively affect both ecological and human health over a considerable distance. However, in this context, geochemical data defining the magnitude of
123
Bull Environ Contam Toxicol
Fig. 1 Map of the spill-affected zone along the Los Remedios River (RR)-San Lorenzo River (SL)-El Comedero dam (CM), and location of the sampling sites (unfilled circles). TD is the location of the discharge point where the mine tailings dam failure occurred
contamination resulting from this spill is nonexistent. Thus, the first study on the composition of the waters was initiated in June of 2013, after various fishermen from the Las Can˜adas de Jacopa cooperative requested help. The purpose of this study was to evaluate the levels of As, Ag, Cd, Cu, Hg, Pb and Zn in the waters of the tributary river Los Remedios, the upper San Lorenzo River, and the El Comedero dam just 3 months after the spill event.
Materials and Methods The El Comedero dam is located in the Cosala´ municipality in Sinaloa state, Mexico. This dam receives waters from the upper San Lorenzo River, which is formed in the Sierra Madre Occidental at Durango State with water contributions from three tributaries: San Juan Camarones, San Gregorio and, most importantly, Los Remedios River (Fig. 1). The El Comedero dam has a total water capacity of 3,966.2 9 106 m3 and covers an area of 9,000 ha. This dam was constructed to control the flow of the San Lorenzo River for harnessing its waters to irrigate approximately 105,000 ha of land in the San Lorenzo and Culiacan valleys and to generate electricity. The climate in the region is warm sub-humid with summer rains (800–1,000 mm), and the average annual temperature is ´ lvarez et al. 2012). 24–26°C (Beltra´n-A A total of 16 surface water samples were collected on June 18, 2013, from (a) the Los Remedios River, which is the main tributary of the San Lorenzo River (three duplicates were collected from sites RR1, RR2 and RR3);
123
(b) upstream of the San Lorenzo River (two duplicates were collected from sites SL4 and SL5); and (c) from the El Comedero dam (three duplicates were collected from sites CD6, CD7 and CD8) (Fig. 1). The water samples were collected using 3-L polyethylene bottles that had previously been acid-washed (2 M HNO3) and rinsed with Milli-Q purified water. Before sampling, the bottles were rinsed two times with river water, and then two 3-L samples were collected at 10 min intervals as close as possible to the main flow and 10–20 cm below the surface. The samples were maintained on ice during transport to the laboratory, where they were filtered through 0.45 lm acidwashed (0.5 M HCl) filters (HA, Millipore). The filtrates were acidified (HNO3 trace metal analysis, J. T. Baker) and used for analysis of the dissolved fraction. The settled and suspended material was obtained from the water samples following a 72 h standing period. This material was then lyophilized (-49°C and 132 9 10-3 mbar for 72 h) and ground. The suspended sediment samples (0.25 g dry weight) were digested with 5 mL of concentrated acids (3 volumes HNO3: 1 volume HCl: 1 volume HF) in capped Teflon (Savillex) vials (UNEP 1995). The mixture was heated at 140°C for 3 h. The samples were diluted with Milli-Q water to a final volume of 25 mL. Blank samples and the standard reference material BCSS-1 were digested (one set) using the same procedure to determine the accuracy and precision. The digested samples were stored in polyethylene containers for further analysis. Depending on the concentration level, analyses of Cd, Cu and Zn were
Bull Environ Contam Toxicol
performed using flame or graphite-atomic absorption spectrometry; Hg determinations were performed using cold-vapor atomic absorption spectrophotometry. The measurements were performed using a Varian SpectrAA 220 instrument. The recovery values from the analysis of the standard reference material BCSS-1 (UNEP 1995) were: As 94.1, Cd 91.6, Cu 101.1, Hg 85.4, Pb 101.7, and Zn 90.3 % (Ag was not certified). The precision, as variation coefficient, varied from 5.0 to 11.2 %. The mean concentrations determined in each river and the dam were compared by one-way repeated measures ANOVA with a p \ 0.05 followed by Tukey´s multiple comparison test (Zar 1999).
Results and Discussion The concentrations of metals measured in water samples from the El Comedero dam and associated rivers are shown in Fig. 2. Data for downstream of the San Lorenzo River (sampled during 2009–2010) were taken from Frı´as-Espericueta et al. (2014). The concentrations of metals in the dissolved fractions from the ecosystems studied here were low likely due to adsorption by suspended sediments, complexation by suspended species and by low solubility of elements in aquatic environments. Dissolved concentrations (in lg L-1) in water ranged from \0.065 to 0.218 for Ag, 6.64 to 35.9 for As, 0.015 to 0.338 for Cd, 4.71 to 10.2 for Cu, 0.019 to 0.242 for Hg, \0.15 to 0.65 for Pb, and 86 to 1,080 for Zn. Figure 2 (upper row) shows the dissolved levels of the analyzed elements as a function of distance; Cd and Ag, although irregularly, tend to increase from the Los Remedios River to downstream of the San Lorenzo River; similarly, Zn exhibited the same tendency, but only until El Comedero dam. The concentrations of As, Hg, Pb and Cu exhibited a tendency to decrease with an irregular pattern. Considering the mean concentrations, the order (highest to lowest) was Zn [ As [ Cu [ Pb [ Cd [ Ag [ Hg. With the exception of As (5 of 8 sites), these concentrations were relatively low and were less than the upper limit value for drinking water established by the World Health Organization (UNEP 2008): As (10 lg L-1), Cd (3.0 lg L-1), Cu (2,000 lg L-1), Hg (1.0 lg L-1), Pb (10 lg L-1), and Zn (3,000 lg L-1). Similarly, these concentrations were less than the maximum contaminant levels established by the Mexican Official Norm (NOM 1994): As (50 lg L-1), Cd (5.0 lg L-1), Cu (2,000 lg L-1), Hg (1.0 lg L-1), Pb (25 lg L-1), and Zn (5,000 lg L-1). Waters from the San Lorenzo River and El Comedero dam are used for drinking, irrigation for agriculture, livestock and fish culture. Considering the maximum safe concentrations proposed for the culture of warm-
water species (Boyd 2009), Ag (0.05–0.10 lg L-1), Cd (25–70 lg L-1), Cu (25–70 lg L-1), Hg (0.1 lg L-1), Pb (200–4,000 lg L-1) and Zn (50–100 lg L-1), the waters from the El Comedero dam are, with respect to Cd, Pb and Cu, suitable for fish farming. However, considering Hg, and particularly Ag and Zn, these waters appear to be unsuitable. In general, the concentrations of dissolved Hg found in this study are comparable to or within the range reported for small streams in Europe (\0.001–1.08 lg L-1) (Protano et al. 2014), Indonesia (0.08–0.22 lg L-1) (Yasuda et al. 2011) and China (\0.01–0.69 lg L-1) (Lin et al. 2012). The Ag levels found in this study were below or close to the mean dissolved Ag level reported in major unpolluted world rivers (0.3 lg L-1) (Martı´n and Meybeck 1979), which indicates that the influence of the spill on the dissolved Ag was not significant. In contrast the As concentrations were within of ranges for affected rivers from polluted European rivers (4.5–45 lg L-1) and for rivers mining influenced (\0.2–7,900 lg L-1) (Plant et al. 2005). Similarly, the dissolved concentrations of Cd, Cu and Pb were comparable to and within the ranges found in rivers from the Danube and Marisa drainage basins (Cd, \0.3–9.4 lg L-1; Cu, \0.8–250 lg L-1; Pb, \0.3–200 lg L-1), with the exception of the Zn levels, which were two times higher than those from Eastern Europe (Zn,\1–460 lg L-1) (Bird et al. 2010). However, when the levels are compared to the maximum levels found in streams from southern Ecuadorian rivers impacted by gold mining (Carling et al. 2013) or miningaffected tributary streams from the Danube basin (Bird et al. 2010), our values are low. The available information on metal concentrations in the rivers of NW Mexico is relatively scarce. In waters from the San Pedro River (Cananea mining district), which is affected by drainage from mine tailings, extremely variable ´ lvarez et al. 2009): 10–63,950 levels were found (Gomez-A for Cd, 60–1,200,000 for Cu, \2–75,000 for Pb, and 10–54,000 lg L-1 for Zn. Frı´as-Espericueta et al. (2014) examined eight rivers in the downstream portion of Sinaloa state (including the San Lorenzo River), in which the levels of the dissolved fraction were higher for Cd (0.32–1.06 lg L-1) and Pb (3.43–8.27 lg L-1) and lower for Cu (1.36–2.10 lg L-1) and Zn (8.0–16.9 lg L-1) than the levels found in this study. This rough comparison reveals that dissolved Cu and Zn are enriched in the waters of the El Comedero dam, upstream of the San Lorenzo River and in the Los Remedios River. This situation is more clearly evidenced by comparing the waters from the downstream versus the upstream region of the San Lorenzo River; although the levels of Cd are comparable in both waters, the levels of Zn and Cu are 77 and four times greater in the upper waters than in the lower waters. It is interesting the contrary behavior of Pb, the waters are around 14 times greater in the lower waters than in the
123
Bull Environ Contam Toxicol
Fig. 2 Concentrations of Ag, As, Cd, Cu, Hg, Pb and Zn in water samples from El Comedero dam and associated rivers: dissolved (top row) (lg L-1) and suspended fractions (bottom row) (lg g-1). Sampling sites are from Fig. 1; SLD corresponds to a site in the San
123
Lorenzo River 60 km downstream of El Comedero dam that was analyzed by Frı´as-Espericueta et al. (2014). An asterisk indicates a concentration that exceeds the PEL for sediments (CCME 2001) or the WHO for water (UNEP 2008)
Bull Environ Contam Toxicol
upper waters. Results from an isotopic study of Pb in the region (Soto-Jime´nez et al. 2008), revealed that the Pb contained in environmental matrices is mostly derived from gasoline (in use until 1997) and US emissions (aerosols from industrial activity), while natural Pb weathered from mother rock represented a minimal contribution. In contrast to the dissolved fraction, concentrations of metals in suspended sediments were relatively high. The concentrations (in mg kg-1) ranged from 22.1 to 531 for Ag, 91.4 to 130 for As, 3.14 to 6.30 for Cd, 65 to 123 for Cu, 0.47 to 1.09 for Hg, 260 to 818 for Pb and 742 to 1,810 for Zn. Considering the mean concentrations, the order of the decrease was Zn [ Pb [ Ag [ As [ Cu [ Cd [ Hg. Unlike the dissolved fraction, the seven elements in the suspended sediments exhibited a gradient that was enriched from the upper sites (RR1 or RR2) through upstream of the San Lorenzo River, the El Comedero dam to downstream of the San Lorenzo River (Fig. 2, bottom row). The concentrations of suspended Cd, Cu, Pb and Zn were higher in comparison to those from downstream waters of the San Lorenzo River (SLD) (Frı´as-Espericueta et al. 2014). As, Ag and Hg data were not available for SLD. In general, the levels of all elements in the suspended sediments from the rivers and dam were relatively high. Particularly, the concentration of As (100 % of samples), Pb (100 %), Zn (100 %), Cd (87.5 %) and Hg (62.5 %) exceeded the PEL (probable effect level) established by the Canadian Sediment Quality Guidelines for the Protection of Aquatic Life. Although Cu clearly showed a concentration gradient from the upstream waters until the dam and the downstream waters, such levels did not exceed the PEL values. The PEL represents the lower limit of the range of chemical concentrations that are usually or always associated with biological effects (CCME 2001). The PEL value for Ag is not available. However, the Ag levels found were comparable or higher than those registered for unpolluted major world rivers (\0.3–60 mg/kg-1) (Martı´n and Meybeck 1979); 62.5 % of the samples exceeded the upper range value. The metal concentrations in the suspended sediments found here were comparable to or greater than, depending on the metal, those from West Java Indonesia (Yasuda et al. 2011). Levels of Hg were similar, while levels of Cd, Cu and Zn were higher than those reported in the Cikaniki River, which is affected by gold mining. However, our results are low compared to those from watersheds influenced by metal mining activities in regions of Europe or South America. In the Danube, levels as high as 33 mg kg-1 for Cd, 8,088 mg kg-1 for Cu, 1,700 mg kg-1 for Pb, and 2,010 mg kg-1 for Zn have been reported (Bird et al. 2010). In Ecuadorian rivers impacted by gold mining, concentrations as high as 46,049 mg kg-1 for As, 104 mg kg-1 for Cd,
9,134 mg kg-1 for Cu, 34 mg kg-1 for Hg, 6,646 mg kg-1 for Pb, and 9,792 mg kg-1 for Zn have been reported (Carling et al. 2013). It is generally assumed that differences between metal concentrations in the water and sediment samples are associated with long-term exposure in sediments, whereas metal concentrations in water indicate recent contamination (Protano et al. 2014). From this scenario it is possible to argue for this case that a few days after the spill the dissolved metals were most enriched with those metals liberated from the tailings. Subsequently, upstream rivers were affected by medium or long-term metal pollution associated with the suspended material from the tailings. Importantly, the dissolved concentrations represent the environmental health risk from each of the elements at the sampling time (or after few days), whereas the suspended sediment metal concentrations provide information about the risks over an extended period of time, which could be of months or years. Acknowledgments G. Ramı´rez-Rese´ndiz for the preparation of figures. This work was supported by the Universidad Nacional Autono´ma de Me´xico, PAPIIT IN210609 and PAPIIT IN208813-2.
References ´ lvarez R, Sa´nchez-Palacios J, Farı´as-Sa´nchez A, Ramı´rezBeltra´n-A Lozano JP (2012) Aspectos reproductivos de la lobina negra (Micropterus salmonides) en la presa Jose´ Lo´pez-Portillo, El Comedero, Sinaloa, Me´xico. Ciencia Pesquera 20:65–75 Bird G, Brewer PA, Macklin MG, Nikolova M, Kotsev T, Mollov M, Swain C (2010) Dispersal of contaminant metals in the miningaffected Danube and Maritsa drainage basins, Bulgaria, Eastern Europe. Water Air Soil Pollut 206:105–127 Boyd CE (2009) Trace metal toxic at high concentrations. Glob Aquacult Adv Aug pp 24–26 Cabrera F, Ariza J, Madejo´n P, Murillo JM (2008) Mercury and other trace elements in soils affected by the mine tailing spill in Aznalco´llar (SW Spain). Sci Total Environ 390:311–322 Carling GT, Dı´az X, Ponce M, Perez L, Nasimba L, Pazmino E, Rudd A, Merugu S, Ferna´ndez DP, Gale BK, Johnson WP (2013) Particulate and dissolved trace element concentrations in three Southern Ecuador rivers impacted by artisanal gold mining. Water Air Soil Pollut 224:1–16 CCME (2001) Canadian Council of Ministers of the Environment. Canadian sediment quality guidelines for the protection of aquatic life. http://www.ccme.ca/sourcetotap/wqi.html. Accessed 30 May 2014 EPA (2014) United States Environmental Protection Agency. Water: drinking water contaminants. http://water.epa.gov/drink/contami nants. Accessed 30 May 2014 Frı´as-Espericueta MG, Mejı´a-Cruz R, Osuna-Lo´pez I, Muy-Rangel MD, Rubio-Carrasco W, Aguilar-Jua´rez M, Voltolina D (2014) Metal discharges by Sinaloa rivers to the coastal zone of NW Mexico. Bull Environ Contam Toxicol 92:132–136 ´ lvarez A, Meza-Figueroa D, Villalba-Atondo AI, ValenzuGomez-A ela-Garcı´a JL, Ramı´rez-Herna´ndez J, Almendariz-Tapia J (2009) Estimation of potential pollution from mine tailings in the San Pedro River (1993–2005), Mexico–US border. Environ Geol 57:1469–1479
123
Bull Environ Contam Toxicol Lin Y, Vogt R, Larssen T (2012) Environmental mercury in China: a review. Environ Toxicol Chem 31:2431–2444 Martı´n JM, Meybeck M (1979) Elemental mass-balance of material carried by major world rivers. Mar Chem 7:173–206 NOM (1994) Norma Oficial Mexicana NOM-127-SSAI-1994. www. salud.gob.mx/unidades/edi/nom/127ssa14.html. Accessed 30 May 2014 Plant JA, Kinniburgh DG, Smedley PL, Fordyce FM, Klinck BA (2005) Arsenic and selenium. In: Shewood Lollar B (ed) Environmental geochemistry. Elsevier, San Diego, pp 17–66 Protano CL, Zinna S, Giampaoli V, Romano S, Chiavarini S, Vitali M (2014) Heavy metal pollution and potential ecological risks in rivers: a case study from southern Italy. Bull Environ Contam Toxicol 92:75–80 SGM (2011) Panorama Minero del Estado de Durango. Septiembre 2011. Servicio Geolo´gico Mexicano. www.sgm.gob.mx. Accessed 30 May 2014 Soto-Jime´nez MF, Pa´ez-Osuna F, Scelfo G, Hibdon S, Franks R, Aggarawl J, Flegal AR (2008) Lead pollution in subtropical
123
ecosystems on the SE Gulf of California coast: a study of concentrations and isotopic composition. Mar Environ Res 66:451–458 UNEP (1995) Manual for the geochemical analyses of marine sediments and suspended particulate matter. Reference Methods for Marine Pollution Studies No. 63, UNEP UNEP (2008) Water quality for ecosystem and human health, 2nd edn. United Nations Environment Programme Global Environment Monitoring System/Water Programme, Burlington 120 pp Villarroel LF, Miller JR, Lechler PJ, Germanoski D (2006) Lead, zinc, and antimony contamination of the Rio Chilco-Rio Tupiza drainage system, Southern Bolivia. Environ Geol 51:283–299 Yasuda M, Yustiawati MS, Syawal MT, Sikder T, Hosokawa T, Saito S, Tanaka M, Kurasaki M (2011) Metal concentrations of river water and sediments in west Java, Indonesia. Bull Environ Contam Toxicol 87:669–673 Zar JH (1999) Biostatistical analysis, 4th edn. Prentice Hall, New Jersey
