Journal of the Chilean Chemical Society
On-line version ISSN 0717-9707
J. Chil. Chem. Soc. vol.50 no.4 Concepción Dec. 2005
J. Chil. Chem. Soc., 50, N° 4 (2005), págs: 697-710
CHEMICAL PROFILES IN LAKE SEDIMENTS IN LAGUNA CHICA DE SAN PEDRO (BIO-BIO REGION, CHILE).
LUIS R. CHIRINOS1,*, a, ROBERTO URRUTIA1, NATHALIE FAGEL2, SEBASTIEN BERTRAND2, NADIA GAMBOA3, ALBERTO ARANEDA1 AND CLAUDIO ZAROR4
1 Centro de Ciencias Ambientales EULA-Chile, Universidad de Concepción, P.O. Box 160-C, Concepción-Chile. a Departamento de Ingeniería, Pontificia Universidad Católica del Perú, Av. Universitaria Cuadra 18 s/n, San Miguel, Lima 32 - Apartado 1761, Perú. *Fax: 0056-41-20-7076; E-mail: email@example.com
Metal profiles in lake sediments could provide historical environmental information on impacts caused by human activities and natural events, with view to the formulation of effective environmental policies. This paper presents data on sedimentary metal profiles in Laguna Chica de San Pedro (LCSP) lake, located in the Bio Bio Region in Southern Chile, where important industrial activities are concentrated. Sediment properties (organic and inorganic matter, grain size, particle distribution, biogenic silica, Mn/Fe ratio), major (Al, Ca, K, Mg, Ti, Na), trace (V, As, Co, Cu, Zn, Ni, Cr, Pb, Sr), and mobile (Fe, Mn, P, S) elements, as well as mineral profiles (plagioclase, quartz, clays, amorphous material) are presented, up to 65 cm depth. In general, relatively constant concentration profiles are observed in the sedimentary core mid section, for most elements. Most changes in composition are seen at the top (recent industrial period) and bottom (before human intervention) sections. Variable redox conditions, generated by biological activity at the sediment-water interface are likely to account for composition profiles at the sediment-water interface. On the other hand, physical processes seem to be mostly responsible for concentration changes in Pre-industrial sediments. Mineral content profiles, such as plagioclase, clay and quartz, as well as total clay content remain fairly constant in most of the core, showing significant changes at its bottom part. High excess V, As, and S values, especially at the upper sediment, arise as a consequence of redox conditions in the lake. Certainly, such metal enrichment is mainly associated with natural sedimentary matter supply from the watershed.
Lake sediments are considered as an archive of physical, chemical and biological conditions in lakes and their drainage basin. Under relative steady sedimentation process and in the absence of bio-turbation, depositional sequence of lake sediments provide a temporal composition pattern, therefore the onset, rate and variations in present and past environmental conditions could be assessed. In this sense, environmental impacts caused by human activities, mainly those related to the release of persistent pollutants, can be identified by appropriate analyses of sediment archives.
In particular, industrial fossil fuel (coal and oil) combustion is an important source of airborne persistent pollutants. Combustion by-products are rapidly transported long distances from their source to the final sink in fresh water bodies, by dry and wet deposition processes. Among various contaminants associated with industrial fossil fuel combustion, heavy metals are considered to be the most harmful due to toxicity, persistence and bio-accumulation, which are capable to affect the bio-geo-chemical cycles in water bodies.
Metal sedimentary profiles have been used to assess inputs from anthropogenic activities such as, industrial and domestic sewage discharges and atmospheric emissions from high temperature industrial activities (1), including pollution in remote places like Arctic lakes (2). In order to define an effective environmental policy to control and mitigate environmental impacts, the source of key pollutants as well as baseline conditions should be identified. As occurred in other developing countries, Chilean industrial activities were carried out with no consideration of environmental implications associated with the release of gaseous, liquid and solid industrial wastes. Indeed, environmental evidences of air, water and soil pollution have been identified in the last decade, particularly in the Bio Bio Region in southern Chile (3,4,5,6). Diverse industrial activities such as steel making, electricity production, cellulose plants, sawmills, fisheries, textiles, oil refineries, petrochemical plants and, until recently, coal mining, among others, are present in that Region.
Recently, environmental pollution related to fossil fuel combustion was detected in lake sediments from Laguna Chica de San Pedro (LCSP) located in the Bio Bio Region (7). Spheroidal carbonaceous particles (SPCs) present in sediments are an index of air-borne pollution from industrial fossil fuel combustion.
In order to obtain a complete picture of sedimentary register in LCSP, metal composition profiles are needed. This paper describes the sedimentary metal and mineral profiles at this site, with view to establish historical background levels and current anthropic impacts. Data are assessed using principal component analysis to determine the relationship among chemical and minerals species present in LCSP sediments.
2.- MATERIALS AND METHODS:
2.1.- Study area:
Laguna Chica de San Pedro (LCSP) is a coastal lake, located at 36° 51'S, 73° 05'W, on the mountain range of Nahuelbuta, near the Pacific Ocean (Figure 1). The LCSP is surrounded by mountains of metamorphic basement geology and layers of fluvial basaltic sediments on its eastern and western side respectively (8). LCSP presents a mean depth around 17 m, with a surface area of 0.87 km2 and a drainage basin of 4.5 km2 (9). LCSP was originated by an obstruction of ravine caused by sediments of Bio Bio river soon after the last glaciation. Since that time, the resulting lake is capturing sediments yield from watershed carried out by affluents in winter time (10).
2.2.- Sediment sampling:
Duplicate sediment cores were taken 1 m apart from the central basin of LCSP (at depth 17 m) in may 2003. Sampling site was selected with the aid of a Lowerance 16 echo sounder and sediment cores were obtained by divers using a device equipped with 1 m long Plexiglass tube with 5.8 cm internal diameter. Both sediment cores were capped, sealed and stored at 4 °C at the Laboratories of the Environmental Science Centre until analysis.
2.3.- Sediment analysis:
Grain size and mineralogical (bulk and clay fractions) analyses were conducted in sediment samples at the University of Liege, Belgium, (Department of Chemistry and Department of Geology, respectively). Sedimentary core was sliced at 1 cm thick using a Plexiglass spatula and samples were oven dried at 60 °C until constant weight and stored in clean plastic bags prior to sediment analyses.
Total Organic Carbon (TOC), Biogenic Silica (Bio- Si), Total Organic Matter (LOI-550), and Inorganic Carbonate (LOI-1000) analyses were carried out at Laboratorios de Biología Ambiental del Centro de Estudios Ambientales, Universidad de Concepción, while total element analysis at Laboratorios del Instituto de Corrosión y Protección, Pontificia Universidad Católica del Perú.
2.3.1.- Mean grain size:
Grain-size analysis was carried out by a laser diffraction particle analyser Malvern Mastersizer 2000 (0.02 to 2000 mm). Wet samples were segregated by ultrasound and mixed at 2000 rpm into 100 ml de-ionised water. Sample quantity was adjusted to obtain a laser beam obscuration ranging between 10 and 20%. Mean grain size parameters were calculated over 10 000 scans and their textural distribution was identified (11). In addition, clay, silt and sand fractions were determined and expressed as a percentage of total content.
Bulk and clay mineral analyses were carried out by X-ray diffraction (XRD) on a Bruker D8-Advance diffractometer with CuKa radiation. Dry bulk sediment samples were crushed thoroughly in an agate mortar to less than 100 mm, then an aliquot was mounted on a plastic holder as disoriented powder according to the back-side method (12) and scanned between 2° and 45° (2q). Mainly, plagioclase, quartz, clay, olivine and amorphous material (i.e. volcanic glass, diatoms, non-crystalline minerals) were identified. Semi-quantitative estimation of each mineral was based on the intensity of the principal diffraction peak corrected by a multiplication factor (13).
For clay mineral analysis, about 1 g of dry bulk sediment was mixed with distilled water and after 50 minutes of settling an aliquot of supernatant was obtained to prepare oriented aggregates on glass slides (14) and scanned between 2° and 30° (2q). X-ray intensity measurements were obtained after three routine clay analysis treatments, an air-dried or natural dried (N), solvation with ethylene glycol for 24 hours (EG) and oven heated to 500 °C for 4 hours (500). Main clay species, such as illite, chlorite, smectite and kaolinite, were estimated under semi-quantitative procedure (15). Mineral and clay contents were expressed as a percentage of total mineral and clay abundance respectively.
2.3.3.- Total Organic Matter (LOI-550), Inorganic Carbonate (LOI-1000), Total Organic Carbon (TOC) and Biogenic Silica (Bio-Si):
LOI-550, and LOI-1000 content in lake sediments were estimated by loss-on-ignition techniques (LOI). LOI-550 content was estimated by sequential oven heating at 105 °C of about 1 g sample for 24 hours and at 550 °C for four hours, meanwhile LOI-1000 content by further heating at 1000 °C for two hours. Both LOI-550 and LOI-1000 contents were determined as a percentage of dry weight (16).
TOC content was estimated by organic matter oxidation with potassium dichromate (K2Cr2O7), and concentrated sulphuric acid (H2SO4) (17). Briefly, about 0.5 g of dry sample was mixed with K2Cr2O7 (0.17M, 10 ml) and concentrated H2SO4 (20 ml) and digested for 30 minutes. Then, distilled water (200 ml) , phosphoric acid (10 ml) and sodium fluoride (0.2 g) were added to the mixture. The solution was titrated with ferrous sulphate ammonium solution. TOC content was determined as a percentage of dry weight.
Biogenic silica (Bio-Si) content was determined by UV-VIS spectrometer Lambda 20 Perkin Elmer. Dry sediment samples were chemically digested with Na2CO3 2M at 85°C (18). Measurement precision of each solution was generally lower than 5%. Standard reference material (Tritisol-Si) was used to verify the analytical performance.
2.3.4.- Total element analysis:
Total element analyses in sediment were determined by Inductively coupled plasma-optical emission spectrometry (ICP-OES, Perkin-Elmer Optima 3300 DV). Sediment samples were chemically digested with mineral acids (19). Briefly, about 0.1 g of grounded dry sediment was attacked with concentrated H2SO4 (2 ml), HF (48%, 5 ml) and HClO4 (70-72%, 0.5 ml), and the mixture was heated at 200 - 225 °C until dryness. Then, desionised water (2 ml) and HClO4 (2 ml) were added to the cool residue. This mixture was sand bath heated until dryness. Finally, HCl (6M, 5 ml) and desionised water (5 ml) were added to the cool residue before the mixture was heated until complete dissolution.
Analytical performance was calibrated with standard reference material (CertiPur-ICP Multi-elemental standard solution IV) and verified every ten samples with multi-elemental solutions (Quality Control N° 1 and N° 2). Measurement precision of each sample was generally lower than 15%. The data was processed by a ICP Winlab Version 1.2. Element concentration was expressed as mg per kg of dry weight (mg/kg).
2.4.- Anthropogenic estimation:
In order to determine anthropogenic influence on metal concentrations in lake sediments, it is necessary to establish the natural background level to estimate excess value (Exs) (20). Excess value is estimated as ratio between surplus of metal at depth x (Sx) and its background level (Bg). Sx is the difference between element concentration measured at level x (Cx) and background level.
Exsx = Sx / Bg = (Cx - Bg) / Bg
Exs = 1, represents 100% increase.
Considering the accuracy measurements of metals (< 15%), Exs values lower than 0.2 should not be considered as significant. Guidelines to evaluate natural background levels of elements in lake sediments are described in the literature (21, 22, 23).
3.- RESULTS AND DISCUSSION:
In order to assess historical records of geo-chemical changes in lake sediments a reliable geo-chronology is required. In this respect, considering 210Pb dates in LCSP (7), lake sediments can be divided into Pre-industrial period (before 1915) and Industrial period (2003-1915). In addition, based on sediment properties and chemical changes described later, those periods could in turn be divided as follows:
Industrial period: Zone I (0-6 cm depth), and Zone II (6-16 cm depth)
3.1.- Sediment Analysis:
Sediment properties in terms of mean grain size, particle size distribution, LOI-550, LOI-1000, TOC, Bio-Si and Fe/Mn ratio are shown in figure 2. A black section is observed from the top to 60 cm depth, but turns light-gray in the rest of the core. No major changes are observed in grain size composition in the upper 60 cm of the core, being composed mainly by silt mud according to Wentworth size classification (Mean Grain Size ~ 10 um; mean clay ~ 7%, mean silt ~ 88%, mean sand ~ 5%). However, in the 60-65 cm section, mean grain size decreases with depth, reaching the lowest value at the bottom of the core.
LOI-550 varies with depth from about 15% at the surface sediments to about 2% in the bottom section of the core. Similar behaviour is observed in TOC content, with maximum and minimum values around 4.5% and 0.5% respectively. In addition, LOI-550 co-relates well with TOC (r2 = 0.83), indicating that organic matter decomposition could account for this. LOI-1000 also decreases with depth varying from around 6 - 7% in most of the core to 2% in the deeper sections. Bio-Si concentrations in Zones I, II and most of III, remain fairly constant, with some scattering between 2 and 4.5% at deeper sections. Higher gradients in LOI-550, LOI-1000, Bio-Si and TOC content are observed in zone IV.
The Mn/Fe ratio in lake sediments is commonly used as indicator for paleo-redox conditions of the sediment-water interface at the time of deposition; low Mn/Fe ratios correspond to anoxic conditions (24). The Mn/Fe ratio profile in LCSP indicates that in Zone I oxidising conditions prevail, whereas Zones II and III present anoxic features. Low Mn/Fe ratios are also observed in the upper section of Zone IV. However, a sharp increase in this index occurs at 60-65 cm depth. Such increase in the ratio could be associated to geo-chemical events; moreover, sharp changes in the rest of sediment properties are also observed in this section.
3.2.- Element concentrations in sediments:
Major (Al, Ca, K, Mg, Ti, Na), trace (V, As, Co, Cu, Zn, Ni, Cr, Pb, Sr), and mobile (Fe, Mn, P, S) elements identified in lake sediments from LCSP are shown in figures 3 to 6. As seen in figure 3, similar trends are observed for Mg, Ca, Sr and Na concentration profiles throughout the core. No major changes are observed throughout zones I-III; whereas, element contents drastically increase with depth in Zone IV.
Figure 4 shows that lower K and Cr concentrations are measured near the sediment surface, and remain fairly constant in Zones II and III. In turn, both K and Cr tend to decrease with depth in zone IV. Despite some scattering and a punctual increase at 45 cm depth, Pb concentrations decrease with depth. A similar pattern is shown for Cr and K, indicating that natural sources would account for such behaviour. Moreover, K varies inversely with Mg and Na, indicating that either distinct precipitation process or mineral sources, or both, occurred.
Concentration profiles for Ti, Al, Zn, Cu, Co and Ni are shown in figure 5. Similar trends in Al and Ti concentrations are observed in Zones I - III. However, Ti tends to increase with depth in zone IV, whereas Al concentrations remain fairly constant. Zones I-III present relatively constant Zn and Cu contents, whereas Zone IV shows increments with depth. Co and Ni concentrations follow similar trends, with little fluctuations throughout.
Mobile elements (Fe, Mn, P, and S), and trace elements (As and V) content profiles are shown in figure 6. All these elements sharply increase as they approach the surface sediment (Zone I). Moreover, Fe, Mn, P, As, and V remain fairly constants in the rest of the core. Similar behaviour is shown by S in Zones II, III, and most of Zone IV, with a sharp increase at 55-65 cm depth.
In general, relatively constant concentration profiles are observed throughout Zones II and III, for most elements. Indeed, as seen in Table 1 most coefficient of variation (CV) are smaller than 10%. On the other hand, most changes in composition are seen in Zones I and IV. Variable redox conditions, generated by biological activity in the sediment-water interface are likely to account for composition profiles in Zone I. Nevertheless, physical processes seem to be mostly responsible for concentration changes in Zone IV.
3.3.- Mineral contents in sediments:
Table 2 compares LCSP sedimentary metal results with those obtained from Chilean pristine sites (25, 26). Metamorphic (27) and igneous (28) rocks composition are also shown as a reference values in Table 2. Metal concentrations in LCSP sediments are higher than in the reported pristine lakes. Moreover, sediment composition approaches metamorphic rather than igneous rocks, suggesting that weathered metamorphic rock is the most probable source.
Figure 7 shows relative abundance in mineral composition such as clays, quartz, plagioclase and olivine and amorphous material in LCSP. Bulk minerals are mainly composed of clay (mean ~ 45%), quartz (mean ~ 16%), plagioclase (mean ~ 8%) and amorphous materials (mean ~ 30%).
Relative abundance of plagioclase increases with depth showing the highest gradient in Zone IV. On the other hand, quartz and clay mineral profiles remain stable, except in Zone IV, where relative abundance decrease with depth. Organic matter, Bio-Si (opal) and non-crystalline minerals (i.e. allophanes) would account for high amorphous material content. Minor relative abundance of olivine mineral is identified at the bottom section in Zone IV. As seen in Figure 1, dominant soil in the LCSP basin is Nahuelbuta type (Figure 1). They are mainly composed of clayey soils developed on metamorphic rocks with debris of quartz (29). Olivine mineral in lake sediments would more probably be associated with volcanic soils reworked by tectonic and/or volcanic destabilisation processes within the drainage basin such as earthquakes, volcanic eruptions and tsunamis (30).
3.4.- Excess values profiles:
Figure 8 presents Excess (Exc) curves for Pb, P, Fe, Mn, S, V and As estimated according to procedure described before (see point 2.4). Natural background levels are assumed to be located at 30-40 cm depth, since no significant anthropogenic influence would have been present at the time of sediment deposition.
All those elements show a significant increase as the sediment-water interface is approached in Zone I. Higher excess values (Exc) are obtained at the surface sediment for As, Mn, V and S. It must be pointed out that V, As, and S are regarded as index elements of atmospheric emissions from fossil fuel combustion (31, 32). On the other hand, Fe and Mn increases would seem to be related to natural apportions from the basin.
Moreover, trace metal increments would also seem to be associated with eutrophic conditions existing in LCSP (33). Metal adsorption by algae or uptake in oxidising water enhances trace metal precipitation (34). In addition, trace metal adsorption at the manganese oxide surface, and further oxidation, would account for such increments.
3.5.- Relationship among metal, mineral and sediment properties:
The relationship among metal, mineral and sediment properties is assessed by Principal Component Analysis (PCA). PCA, a multivariate statistical technique, is generally employed to represent the dimensionality of a data set, preserving the relationships present in the original data. This technique is commonly used to analyse geo-chemical data by creating factors (new variables) which represent clusters of interrelated variables. The physical meaning of each factor is assigned by considering the variable with strongest factor loading (35,36). Principal component analysis extracted four factors (eigenvalues > 1), that account for 83 % of the variance from the 27 variables in the original data set. This data set comprises geo-chemical parameters, mineralogy (quartz, clay, plagioclase) amorphous material, LOI-550, TOC, Bio-Si, and LOI-1000. PCA diagrams are shown in figure 9.
CP 1: Geo-chemical sediment matrix and mobilisation of lead:
About 48 % of the total variance is explained by this component. High positive factor loading is assigned to elements: Ca, Na, Sr, Mg, Cu and Ti, and plagioclase mineral, whereas negative values to LOI-550, LOI-1000, TOC, K, Cr, Pb, and quartz and clay minerals. Based on these associations, this component describes the geo-chemical matrix composition of sediments from the basin represented by Ca, Na, Sr and LOI-550, which are closely related to plagioclase, and quartz - clay minerals, respectively. Otherwise, relatively high Pb content appears to be associated with LOI-550 instead of being adsorbed onto Fe and Mn oxy-hydroxides (37). Interestingly, considering the stratigraphic sequence of sediment samples, allochthonous components from the basin during the Pre-Industrial period would be mainly composed of plagioclase-rich fraction, changing to clay- and quartz-rich fractions in the Industrial period.
CP 2: Metal enrichment processes:
This component accounts for 19 % of the total variance. High positive value for factor loading is observed in mobile elements Fe, Mn, S, P and in trace elements As and V. Changes in Fe and Mn concentrations in lake sediments are mainly controlled by their supply rate and redox conditions present in the water sediments interface at the time of deposition (38). Reduced Fe and Mn can migrate to surface sediments to be re-oxidised and precipitated as oxy-hydroxides, yielding higher concentration of these elements at the surface sediments. Apparently, S, P, As and V cycles are strongly related to Fe and Mn cycles, responsible for increments in their concentrations in the surface sediments. However, part of these element concentrations in the surface sediment could be associated with high temperature industrial activities such as smelting and fossil fuel combustion occurring around the study area.
CP 3: Trace elements in geo-chemical sediments :
This component, accounting for 9 % of the total variance, delivers high positive factor loading to Ni, and middle factor loading to Zn and Co. All these elements are commonly identified as proxies for atmospheric pollution caused by heavy industrial activities (39). However, no local industrial activities responsible for these metals emissions are present in the study area. Therefore, natural processes are likely to be responsible for the presence of these trace element.
CP 4: Autochthonous productivity :
About 7 % of the total variance is explained by this component, which is composed of Bio-Si and amorphous material, with major and middle factor loading respectively. This component could be related to productivity within the lake. Indeed, biogenic silica is an index of diatom productivity, as well as its variability in fresh water bodies.
In general, relatively constant concentration profiles are observed in the sedimentary core mid section (Zones II and III), for most elements. Most changes in composition are seen at the top (Zone I) and bottom (Zone IV) sections. Variable redox conditions, generated by biological activity in the sediment-water interface, are likely to account for composition profiles at the sediment-water interface. Nevertheless, geo-chemical processes seem to be mostly responsible for concentration changes in Zone IV.
Mineral composition such as plagioclase, clay and quartz, as well as total clay content remain fairly constant throughout zones I-III, however significant changes in both components are observed in zone IV.
High excess V, As, and S values, specially in Zone I, lead to assume that anthropogenic metal inputs are responsible for these increments; however, metal enrichment processes observed at surface sediments arise as a consequence of redox conditions in the lake. Certainly, such metal enrichment is probably associated with natural sedimentary matter supply from the watershed.
Therefore, given such dominant sedimentary metal enrichment under redox conditions, metal anthropogenic inputs in lake sediments are not clearly differentiated from natural contributions. In this sense, further chemical analysis (e.g. sequential extractions) must be performed to adequately assess the extent of human metal contributions.
The authors would like to express their special thanks to Dr. Oscar Figueroa, Mrs. Mónica Montory and Mr. Cristian Espinoza for their contribution. This work was partially supported by DAI-PUCP N° 3001, DIUC N° 230.310.0.35-1.0. and Project 3.3, 2003-2005 (Scientific Co-operation Exchange Program between University of Concepción (Chile) and University of Liege (Belgium)). In particular the first author has received a grant for a two months research stay in Belgium funded by the "Commissariat Général aux relations internationales de la Communauté française de Belgique".
1.- K. Johansson, Water Air Soil Poll., 1989, 47, 441-455. [ Links ]
2.- M. H. Hermanson, Environ. Sci. Technol., 1991, 25, 2059-2064. [ Links ]
3.- R. Barra, M. Cisternas, R. Urrutia, K. Pozo, P. Pacheco, O. Parra and S. Focardi, Chemosphere, 2001, 45, 749-757. [ Links ]
4.- R. Barra, M. Cisternas, C. Suarez, A. Araneda, O. Piñones and P. Popp, Chemosphere, 2004, 55, 965-972. [ Links ]
5.- CONAMA-Chile, Programa de Recuperación de Talcahuano (P.R.A.T.). Diagnóstico, 1994, pp 163. [ Links ]
6.- S. M. Mudge and C. G. Seguel, Mar. Pollut. Bull., 1999, 38, 1011-1021. [ Links ]
7.- L. Chirinos, N. L. Rose, R. Urrutia, P. N. Nuñoz, F. Torrejón, L. Torres, F. Cruces, A. Araneda and C. Zaror, Environmental evidence of fossil fuel pollution in Laguna Chica de San Pedro lake sediments (Central Chile) (submitted). [ Links ]
8.- E. Acencio, Análisis integrado de los sistemas naturales Laguna Grande y Laguna Chica de San Pedro, Tesis de Licenciatura, Facultad de Humanidades y Arte, Universidad de Concepción, 1994. [ Links ]
9.- O. Parra, V. Dellarosa and E. Ugarte, Estudio limnológico de las lagunas Chica de San Pedro, La Posada y Lo Méndez, 1976, 50, 73-86. [ Links ]
10.- M. Cisternas, A. Araneda, O. Retamal and R. Urrutia, Rev. Geo. Norte Grande, 1997, 24, 151-156. [ Links ]
11.- R. L. Folk and W. C. Ward, J. Sediment. Petrol., 1957, 27, 3-26. [ Links ]
12.- G. W. Brindley and G. Brown, Crystal structures of clay minerals and their x-ray identification. Mineral Society Monograph, ed. London, 1980, pp. 495. [ Links ]
13.- H. E. Cook, P. D. Johnson, J. C. Matti and I. Zemmels, in Methods of sample preparation and x-ray diffraction analysis in x-ray mineralogy laboratory, ed. A. G. Kaneps, Init. Rep. DSDP XXVIII. Printting Office, Washington DC., 1975, pp. 997-1007. [ Links ]
14.- D. M. Moore and R. C. Reynolds, X-ray diffraction and the identification and analysis of clay minerals, ed. Oxford University Press, Oxford, 1989, pp. 332. [ Links ]
15.- P. E. Biscaye, Geol. Soc. Am. Bull., 1965, 76, 803-832. [ Links ]
16.- J. F. Boyle, in Inorganic Geochemical Methods in Palaeolimnology, ed. L. W.M. and S. J.P., Kluwer Academic Publishers, Dordrecht, The Netherlands, 2001, pp. 83-140. [ Links ]
17.- H. E. Gaudette, W. R. Flight, L. Toner and D. W. Folger, J. Sediment. Petrol., 1974, 44, 249-253. [ Links ]
18.- R. Mortlock and P. Froelich, Deep-Sea.Res., 1989, 36, 1415-1426. [ Links ]
19.- M. L. Jackson, Soil Chemical Analysis, ed. Prentice-Hall Inc., Englewood Cliffs, New Jersey, 1958, pp 59-67. [ Links ]
20.- H. Erlenkeuser, E. Suess and H. Willkomm, Geochim. Cosmochim. Ac., 1974, 38, 823-842. [ Links ]
21.- W. Salomons and U. Förstner, Metals in the Hydrocycle, ed. Springer-Verlag, Berlin, Heidelberg, New York, Tokyo, 1984, pp. 349. [ Links ]
22.- D. H. Loring and R. T. T. Rantala, Earth-Sci. Rev., 1992, 32, 235-283. [ Links ]
23.- M. W. Binford, J. Paleolimnol., 1990, 3, 253-267. [ Links ]
24.- P. Wersin, P. Höhener, R. Giovanoli and W. Stumm, Chem. Geol., 1991, 90, 233-252. [ Links ]
25.- R. Urrutia, M. Yevenes and R. Barra, J. Chil. Chem. Soc., 2002, 47, 457-467. [ Links ]
26.- R. Ahumada and A. Rudolph, Estuar. Coast Shelf S., 2004, 59, 231-236. [ Links ]
27.- P. S. Vásquez, Petrología y Geotermobarometría del Basamento Metamórfico de la Cordillera de la Costa entre los 36° 30'S y 38° 00'S, Título de Geólogo, Departamento de Geologia, Universidad de Concepción, 2001. [ Links ]
28.- C. E. Creixell, Petrología y Geotermobarometría del las Rocas Intrusivas de la Cordillera de la Costa entre los 36° 30'S y 38° 00'S, Título de Geólogo, Departamento de Geología, Universidad de Concepción, 2001. [ Links ]
29.- FIA, Proyecto suelos forestales de la VIII Región. Informe Final. Ministerio de Agricultura - Fondo de Investigación Agropecuaria, ed. Chillán-Chile, 1990, pp. 150. [ Links ]
30.- M. Lagos, Revista Geográfica Norte Grande, 2000, 27,93-102. [ Links ]
31.- W. M. Henry and K. T. Knapp, Environ. Sci. Technol., 1980, 14, 450-456. [ Links ]
32.- P. Bacci, M. Del Monte, A. Longhetto, A. Piano, F. Prodi, P. Redaelli, C. Sabbioni and A. Ventura, J. Aerosol Sci., 1983, 14, 557-572. [ Links ]
33.- R. Urrutia, K. Sabbe, F. Cruces, K. Pozo, J. Becerra, A. Araneda, W. Vyverman and O. Parra, Rev. Chil. Hist. Nat., 2000, 73, 717-728. [ Links ]
34.- Kuhn, C. A. Johnson and L. Sigg, Cycles of trace elements in a lake with a seasonally anoxic hypolimnion, ed. The American Chemical Society, Washington D.C., 1994, pp. 627. [ Links ]
35.- T. A. DelValls, J. M. Forja, E. González-Mazo, A. Gómez-Parra and J. Blasco, Trends Anal. Chem., 1998, 17, 181-192. [ Links ]
36.- P. Szefer and B. Skwarzec, Mar. Chem., 1988, 23, 109-129. [ Links ]
37.- X. Li, Z. Shen, O. W. H. Wai and Y-S. Li, Mar. Pollut. Bull., 2001, 42, 215-223. [ Links ]
38.- W. Davison, Earth-Sci. Rev., 1993, 34, 119-163. [ Links ]
39.- M. Ravichandran, M. Baskaran, P. H. Santschi and T. S. Bianchi, Environ. Sci. Technol., 1995, 29, 1495-1503. [ Links ]