SciELO - Scientific Electronic Library Online

 
vol.61 issue1HYDROXYCINNAMIC ACID DERIVATIVES AND FLAVONOL PROFILES OF MAQUI (Aristotelia chilensis) FRUITSCHEMICAL CHARACTERIZATION OF SUB-BITUMINOUS COAL FROM THE ARAUCO PROVINCE - CHILE author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

  • On index processCited by Google
  • Have no similar articlesSimilars in SciELO
  • On index processSimilars in Google

Share


Journal of the Chilean Chemical Society

On-line version ISSN 0717-9707

J. Chil. Chem. Soc. vol.61 no.1 Concepción Mar. 2016

http://dx.doi.org/10.4067/S0717-97072016000100011 

HEAVY METAL CONCENTRATIONS IN WATER AND SEDIMENTS FROM AFFLUENTS AND EFFLUENTS OF MEDITERRANEAN CHILEAN RESERVOIRS

SYLVIA V. COPAJA1*, VESNA R. NUÑEZ.1, GIGLIOLA S. MUÑOZ1, GEISSY L. GONZÁLEZ 1, IRMA VILA2 AND DAVID VÉLIZ.2 3

1Departamento de Química, 2Departamento de Ciencias Ecológicas, 3Instituto de Ecología y Biodiversidad, Nucleo Milenio en Ecología y manejo sustentable de Islas Oceánicas, Facultad de Ciencias. Universidad de Chile. Casilla 653 Ñuñoa, Santiago, Chile.

e-mail:scopaja@uchile.cl


ABSTRACT

RIVER flows have constant interaction between water and bed sediments; for this reason knowledge of the characteristics of the sediments is fundamental to understand water chemistry. This study determined the concentrations of heavy metals in water and sediments in the affluents and the effluents of the MediterraneanChilean reservoirs Cogotí, Corrales, La Paloma, and Recoleta. We explore possible ecological risk and toxicity using the enrichment factor (EF), risk assessmentcode (RAC), threshold effect concentrations (TEC) and probable effect concentrations (PEC). The results showed that five metals: Al, Fe, Cu, Mn and Zn out of the ten measured metals were detected in both surface water and the sediments. The risk assessment code (RAC) suggested that Fe represents a medium risk in the affluent of Cogotí Reservoir: Cu, Zn and Mn represent a medium to high risk in all the dams and in both zones (affluents and effluents). The enrichment factor(EF) determined that the five metals were lithogenic. Fe, Cu, and Mn are the elements that present the greatest toxicity to microorganisms in these aquatic systems.

Key words: Heavy metals, dams, enrichment factor, sediment, risk assessment code.


1. INTRODUCTION

RIVER flows have constant interaction between water and bed sediments; for this reason knowledge of the characteristics of the sediments is fundamentalto understand water chemistry. It is known that this interaction may solubilizeor capture compounds that may be bio-available. One effect of this interactionis the possible pollution into the rivers by the presence of compounds orelements coming from human activities. It is estimated that currently over onemillion different substances are introduced in natural water discharges fromanthropogenic use1. Many of them are not considered toxic, but they can alteor the organoleptic characteristics of water or severely disrupt the ecosystem2.Some of the chemicals are potentially toxic, such as heavy metals including:Al, As, Cd, Cu, Cr, Hg, Mn, Ni, Pb and Zn, even though some of these do notcorrespond to the exact definition of what is considered a heavy metal3.

Because heavy metals are not biodegradable, they usually are not removed from aquatic ecosystems by natural processes4'6, the refore heavy metals are of great significance as indicators of the ecological quality of all water flow, due to their toxicity, persistence and bio-accumulative behavior7, 8 The contribution of these metals to the hydrological cycle comes from different sources, the mostimportant being of lithogenic or geochemical origin. However, at present the irgreatest concentration is due to human activity, namely mining, agriculture,industrial processes and household waste9. The refore, an important part of the heavy metals in the water may be related to human activities10, n. This humanactivity has often led to the transformation of the water of rivers, lakes and coasts into waste deposits; natural balance is severely disturbed and in manycases totally lost4.

Once heavy metals enter the aquatic environment they generally show affinities to bind to suspended matter and thus to accumulate in sedimentsthrough sedimentation, mainly in rivers, lakes and seas4, 12. These elements caneasily move from the solid to liquid phase of water and vice versa following changes in both the biotic and abiotic components. The metals in the sedimentsmay re-solubilize in different chemical forms due to changes in environmentalconditions such as pH, redox potential (Eh), dissolved oxygen and presence of organic carbon13'16. The analysis of heavy metals in sediments allows usto detect contamination and also provides information on the critical areas of aquatic systems1, 7 9 17, 19, 20.

Bioavailability of metals in sediments has a direct impact on some aquatic species, many of which can accumulate high concentrations of metals that cancause chronic effects on their populations21. For this reason these pollutantsare among the most frequently monitored using standard analytical techniques for extraction and quantification. Since the 1980s, many efforts have beenundertaken around the world to measure and characterize the behavior and distribution of heavy metals in sediments22'24.

Given the great importance of hydrology, alterations in the flow of rivers caused by humans have serious consequences25, increasing recognition thatanthropogenic changes in rivers such as construction of dams, river diversionsor channel modifications have significant long-term consequences for watersupply, water quality, aquatic ecosystems and sediment budgets26, 27. Amongthese interventions dams are one of the most dramatic and widespread impactsof humans on the natural environment28. Dams were primarily built to supplyhumans with fresh water, either for direct consumption or for agriculture. Later,reservoirs were also built to produce hydroelectric power and to regulate riverflow29. It has been estimated that in the world at present there are around 45,000dams with a water column depth of more than 15 meters30. However, there areno systematic studies on the influence of dams on the distribution of heavymetals in sediment, considering that a dam alters the free flow of a river to the sea; only recently has work begun in which the seasonal distribution of heavymetals is studied in a dam31.

Considering that there are dams in a number of rivers in Chile producing disruption of free water flow, and that mineral salts may accumulate in the zone of the dam32, the goal of this study was to determine the concentrations of heavymetals in both water and sediments as a labile fraction (soluble, exchangeableor bonded to carbonate) and pseudo-totals in the affluent and effluent of four reservoirs (Cogotí, Corrales, La Paloma, and Recoleta). Additionally,we estimated the possible ecological risk from sediment by calculating the enrichment factor (EF)33, risk assessment code (RAC)34'35, threshold effectconcentration (TEC) and the probable effect concentration (PEC)36.

2. EXPERIMENTAL

2.1 Sampling area

Mediterranean rivers from 30°S to 34°S in central Chile were studied in order to determine differences in the effluents and affluents of the large reser-voirs. These reservoirs were built at different times; three of them are located inthe Limarí River basin: Recoleta (built in 1934), Cogotí (built in 1938) and LaPaloma (built in 1966); all are used exclusively for agricultural irrigation. The Corrales dam is located in the Choapa River basin; it was constructed in 2001for irrigation of the Choapa valley (Fig. 1). Sampling was conducted in August2010 in the high flow season (winter), from six sites in the affluent and six inthe effluent separated by 0.5 to 2 km.

2.2 Sampling analyses

Measurements of pH, electrical conductivity (EC) and redox potential (Eh) of the water and sediments were made in situ using a portable multimeter(VWR multi 340i).

2.1.1 Water analysis
200 mL of water was sampled to quantify dissolved oxygen (DO) using the Winkler method. A 1 L sample was obtained from the same sites to quantify total phosphorus (TP) following a protocol described before37.


Figure 1. Location of sampling sites in the Mediterranean Rivers of Chile.

2.1.2 Sediment analysis

In each site, three samples of 1kg of sediment were collected in polyethylene flasks according to the protocol described before38, which requires collectingsamples with a plastic shovel from the top of the superficial sediment zone.Samples were brought to the laboratory and stored at 4 °C. The three sampleswere then pooled to obtain the necessary quantity of <0.63 ^m material. In the laboratory, the Walkley-Black method was used to determine organic carboncontent and water soluble phosphorus was determined by the Olsen method39,40

2.1.3 Heavy metals analysis

The heavy metals analyzed in this study were: Al, Cd, Cr, Cu, Fe, Mn, Mo, Ni, Pb and Zn. To determine differences in these metal concentrations, samplesfrom affluents were compared to those obtained in the effluents. Water wascollected in 1L vials, filtered and fixed with 2% nitric acid (suprapur Merck).For sediments, metals were obtained from the 3 kg sampled per site. The labilefraction of the sediments was obtained by stirring 0.5 g of the pellet with 40mL acetic acid (Merck p.a.) 0.11 mol L'1 for 16 h, and then the samples werecentrifuged at 3500 rpm for 30 min. The pseudo-total fraction was obtainedby digesting 1 g of sediment with 10 mL of nitric acid (suprapur Merck) in ahigh resolution microwave oven (MarsX press) using the following conditions:power 800 W; tower 100-5; time 11 min.; temperature 175 °C, maintenance15 min.; cooling 15 min. This was based on EPA method 3051: Microwave-assisted acid digestion of sediments, sludge, soils, and oils. Finally, the sampleswere kept cold (4 °C) for posterior analysis.

Standard solutions for heavy metals were prepared from Titrisol 1000 mgL-1 (Merck); samples were determined using an atomic absorptionspectrophotometer (AAS) (Shimadzu spectrophotometer 6800, ASC-6100 autosampler and graphite furnace GFA-EX7). The following wavelength lines wereused: Cu = 324.7 nm; Al = 309.3 nm; Cr = 357.9 nm; Cd = 228.8 nm; Fe =248.3 nm; Mn = 279.5 nm; Ni = 232.0 nm; Pb = 217.0 nm; Mo = 313.3 nm and Zn = 213.8 nm.

2.1.4 Analytical method validation and quality control

To assure the accuracy of the data reported, recovery experiments were performed using standard reference material for water and sediments (ERM-CA615 and BCR-320R, respectively). The concentration ranges were basedon the limit of detection (LOD) and the limit of quantification (LOQ) for eachmetal. The experiment was performed in triplicate; a calibration curve wasobtained to determine the linear relationship between absorbance and metalconcentration in the concentration range used. Reagent blanks were prepared and measured in the same way as samples.

2.1.5 Statistical analysis

To determine differences in pH, electrical conductivity and redox potential between dams and sites (affluents and effluents) in water and sediments, two-way ANOVA permutations were performed using R software41. A principalcomponents analysis (PCA) was used to determine the relationships betweenmetals in water, sediment and environmental variables.

3. RESULTS AND DISCUSSION

3.1 Parameters for analytical methods

Tables 1 and 2 show the analytical parameters from the validation of methods for water and sediments.


Table 1. Analytical parameters for determination of total heavy metals in
water (Water Reference Material: ERM-CA615).

* Metals measures by electrothermal atomization

 

The percent of recovery were satisfactory and indicated a good agreement between our data and the reference values. For water the highest detection lim-its and quantification limits were for Al, while the lowest limits correspondedto Cu and Mo, both determined by thermal electro-atomization (graphite furnace). For sediment Al also presented the highest limits of detection and quantification in both fractions, while the lowest limits were for Cd.

3.2 Physicochemical Characterization of water and sediment

Table 3 shows the physical and chemical characteristics of surface water and sediment in affluents and effluents of the reservoirs studied. The pH wasalkaline in all the systems; it was most alkaline (pH = 8.63) in the surface waterfrom the effluent of the La Paloma reservoir and least alkaline (pH = 7.27) inthe effluent of the Corrales reservoir. Differences in pH could be related to the oxidation-reduction potential; these parameters define the existence of solublechemical species or allow metal solubilization from the sediments42.

Significant differences in pH were detected for both sediments and water. In the case of water, a difference was detected in the interaction between damand site (affluent and effluent) (p = 0.001, Fig. 2a) showing that the effluent of Corrales dam presented lower values compared to the other sites (p < 0.005,Fig. 2b).


Table 2. Analytical parameters for determination of soluble and total heavy metals in
sediment (Sediment Reference Material: BCR- 320R).


Table 3. pH, conductivity (EC) and redox potential (Eh) measured in superficial water and sediments



Figure 2. Boxplot pH in sediments (a) and water (b)

Electrical conductivity (EC) was highest in the effluent of the Recoleta dam for both water and sediment, while the lowest value was measured inthe effluent of Cogoti (Table 3). The ANOVA performed for EC showed asignificant effect of the interaction of dams and site for both water and sediment(p < 0.001 and p = 0.002, respectively). In the water column, the effluent of the Recoleta dam showed higher values (p < 0.005, Fig.3a). In the case of the sediment, the effluent from La Paloma had higher EC values than the other sites(p < 0.005, Fig. 3b). Low EC values indicate low saline waters and sediments(EC 0.0-3.0 dS m'1); values >3 dSm'1 indicate salinity problems. In this study the highest values were found in the effluent of Recoleta reservoir (surfacewater and sediment, 1.60 ± 0.90 dS m'1) and the effluent of La Paloma 1.13 ±0.90 dS m'1, indicating that these two sites must be considered to have moderatesalinity (quality C1).


Figure 3. Boxplot of for EC in sediments (a) and water (b).

 

Eh valúes indícate oxidant conditions in water and sediments. The ANOVA performed for Eh showed a significant effect of the interaction in both water and sediments (p < 0.001 and p = 0.008 respectively). In the water column, the effluent of Corrales and Recoleta showed lower values compared to the other sites (p < 0.005, Fig. 4a), in the sediments, the effluent and affluent of Recoletashowed the highest values (p< 0.005, fig4).


Figure 4. Boxplot of for Eh in sediments (a) and water (b).

Table 4 shows the concentrations of dissolved oxygen and total phosphorus in surface water and percentages of organic carbon and soluble phosphorus in the sediment. In the case of the oxygen no significant differences were foundbetween dams or sites (p > 0.05).

There were not large differences in %OC between dams and between af-fluents and effluents except in the effluent of Recoleta reservoirs. This could be explained by the proximity of the city of Ovalle, which may be a source of urban waste; the same explication applies for the high value of soluble phos-phorous. Dissolved oxygen values did not show large differences betweenreservoirs and affluents-effluents, indicating oxygenated systems. The effluent of the Corrales reservoir was the exception, which showed the lowest value(7.34 mgL'1). There were differences in total P between and among reservoir tributaries and effluents. Total phosphorous in water showed the highest valuesin the effluent of Corrales, which may be explained because this sample was collected near a camping site. Water-soluble phosphorous in sediment showed the highest value in the effluent of La Paloma, probably this is due to use offertilizers (Table 4).


Table 4. Summary of the chemical characteristics measured in water and sediments. Each value
represents the mean (and standard deviation) of six samples. <LD =below detection limit

 

3.4 Heavy metal determination in water

The analysis of heavy metals in surface water, labile fraction and pseudo-total of sediments showed values above the detection limit for five metals (Al, Cu, Fe, Mn and Zn), thus the Figures and Tables show results only forthese elements. We found no differences in most metal concentrations betweenaffluents and effluents, indicating that the reservoirs do not act as a filter43, 44.

Metal concentrations in surface water were higher than in the soluble fraction of the sediments in both tributaries and effluents. The highest concentrations found in surface water were Mn > Fe > Al > Cu > Zn. There are no studies of heavy metals in surface waters in affluent and effluent of reservoirs. Thus wecan only compare with other basins, for instance in thirteen sites of the basin of the river Ebro (Spain ) Cu values ranged from 3.68 - 94.6 ^gL- 1; Mn from 2.41- 1227 ^gL 1; Zn from <0.2 - 120 ^gL -1; Al and Fe were not measured in thisstudy45. A better comparison because of geographical similarity is the ChoapaRiver basin (Chile); Al: 71.1-357; Cu: 2.45-65.1; Fe: 81.5-222; Mn: 10.3-70.6;Zn: 13.5-31.2. Only the high values of Cu, Mn and Zn from the Ebro basin arenear the values of the metals in the reservoirs; the same is true comparing withthose of the Choapa river basin except for Mn, which showed considerablyhigher values. This might be explained by the presence of high concentrations of manganese carbonate, which is solubilized when the pH is decreased forsample preservation46-48.

3.5 Heavy metal determinations in sediment

In all reservoirs, both in affluents and effluents the pseudo-total fraction from sediment was higher, indicating that this fraction has mainly lithogeniccontributions. The highest concentrations of the total fraction were thoseof Al and Fe (7097 to 37769 and 6785 to 42567 mg g-1, respectively). The labile or soluble fraction of sediment was lower in all situations or underdetection limits (Table 6). se results should be considered rememberingthat solubility of metals depends on characteristics of water systemsuch as pH and Eh42. Heavy metals tend to form partnerships with mineralsby ion exchange phenomena, complexes or precipitates (carbonates, sulfates,phosphates, etc.) and with organic substances by adsorption, chelation and others49- 51.


Table 5. Summary of the heavy metal concentrations (^gL'1) in surface water.
Each value represents the mean of six samples. <LD = below the detection limit.

It is difficult to compare these values with those reported in other studies, mainly due to the large geographical and geological differences. Determination of heavy metals in sediments of the Mapocho River52 indicated concentrationsin the total fraction of Cu between 209 ± 12 and 2850 ± 490 and Zn: 607 ± 240 and 1290 ± 370; Al, Fe and Mn were not determined in this study. Analysis of metals in the total fraction of sediments of the Aguanilpa Reservoir (Mexico),reported that Al values were found between 22100 and 7760; Fe: 15900 and 4740; Cu: 60.8 and 0.79; Zn: 51.8 and 14.8 ^gg'1; Mn was not determinedin this study31. A better comparison because of geographical similarity corre-sponds to the determination of heavy metals in the basin of the Choapa River(winter season); the values found for Al ranged from 15649-6200; Fe: 3462321656; Cu: 4814-70; Mn: 1671-356; Zn: 91-3346 51.

Cu and Zn concentrations found in sediment (total fraction) in the Mapocho River are slightly higher than those found in our study, indicatingfurther contamination of these sediments. Values reported in the Aguanilpa

Reservoir were relatively similar to ours for Al and Fe, while for Cu and Zn these values were lower; the authors concluded that this reservoir could beconsidered unpolluted. The comparison with the values of the five metalsdetermined in the Choapa River basin indicates that the values found weregenerally slightly lower than those determined in our study, which could be duemainly to the hydrodynamic sediment conditions, i.e. without much sedimentmovement with little ability to transfer, while in the Choapa basin there is anincreased movement of sediment due to the slope of the river46.

In the soluble faction Al was under the detection limit in the most of sites, except in the affluent of the Recoleta reservoir (230 ugg- 1) and the tributary of Cogotí (359 ugg-1) reservoir. Al concentration was higher even when the pH of the water would not solubilize it. Since the ion Al (III) is present only at pH<443, it is unlikely to find Al in significant concentrations in water at alkalinepH.


Table 6. Summary of the heavy metal concentrations in soluble and total fractions of sediments (^gg-1).
Each value represents the mean of six samples. <LD = below the detection limit.

The soluble fraction of the other metals studied ranged between 5% and 70%, depending on the reservoir and the zone (Figure 5). Sequential extrac-tion studies performed in the Choapa River basin show similar behavior51. Themobility and bioavailability of metals varied significantly with sediment prop-erties, organic carbon, carbonates, pH, redox potential, phosphates and waterflow. The soluble fraction was extracted with 0.11 M acetic acid, which givesa pH = 2,9. This indicates that when the pH of the sediment-water system isdecreased, these metals become highly available for the biota, and this couldliberate them in the water. The high percentage of Mn in weakly-bound fractions was probably due to its special affinity for carbonate, indicating that considerable amounts of Mn may be released into water following a non-exchangeprocess and dissociation of the Mn-carbonate phase if sediment conditions be-came more acid39. Cu and Zn are species, besides forming compounds with carbonate which can be released, can be easily exchanged in the adsorption sites of the sediments. Fe solubilized little because at this pH (pH = 2,9) this elementmay be precipitate as iron hydroxylated species, so the highest concentrationof Fe in the Corrales reservoir released in both zones could be due to an extracontribution in this site.

The following figures show the percentages of the soluble fraction of the total sediment fraction for Cu, Fe, Al, Mn and Zn in the affluent and effluent of the four reservoirs.

3.5 Ecological risk and toxicity

3.5.1 Risk assessment code (RAC)

The risk assessment code (RAC) uses the percentage fraction of metals that are exchangeable and associated with carbonates (soluble or labile fraction). Inthis fraction the metals are weakly bound to the sediment, and imply greaterenvironmental risk since they are more available for the aquatic system. The RAC was determined based on the percentage of the total metal content that waspresent in the first sediment fraction (soluble or labile fraction). Percentages of 1-10% represent low risk, 11-30% medium risk and 31-50 % high risk34, 35.

As shown in Fig. 5, Al was found in the soluble fraction in the affluent of Cogotí and Recoleta reservoirs in concentrations ranging from 0.98 to 0.83%. Fe was found only in the affluent of Cogotí, with a concentration of 20.82%. Cuand Zn were found in most of the reservoirs, both in affluents and effluents. The values of Cu were observed in the following order Corrales > Cogotí > Recoleta> La Paloma. Zn showed the lowest percentage in the Recoleta affluent (5.8%),while the highest values were found in the La Paloma reservoir, 20.8% in the affluent and 15% in the effluent. Mn ranged from 69.83% in the effluent of Corrales to 5.94% in the affluent of La Paloma. These results suggest that Feshould be considered as medium risk only in the affluent of Cogotí, while Cuand Zn show medium risk in all the dams and both zones. Finally, Mn showedhigh risk in Corrales, medium risk in Recoleta and Cogotí and low risk in LaPaloma.


Figure 5. Percent of soluble fraction of metals of the total fraction in sediments in affluents and effluents of each dam studied.

These metals may be transferred preferentially to the water column, since higher values in the soluble fraction of these metals represent a greaterpercentage of the total concentration, which is due to the higher pH rangein which these metals have greater solubility, also complemented by the oxygenation of the water. In addition, the highest concentrations of Al and Fewere found in the water, which is probably explained by the re-suspension of particles from the sediment. These two seemingly contradictory positionsindicate that sediments not only act as a source of metals by solubilizationor desorption, but also by re-suspension, depending on the characteristics of the elements involved and the sediment characteristics. Another studyfound similar behavior; Al and Fe were observed in high concentrations inthe mineralogical sediment fraction, while Cu, Mn and Zn would be bound tocarbonates, organic matter or exchangeable fraction51.

3.5.2 Enrichment Factor (EF)

The total concentration of heavy metals in sediments does not represent the degree of contamination coming from either natural or anthropogenic sources,due the grain-size distribution and mineralogy53. The enrichment factor (EF)is an effective method to estimate the anthropogenic impact on sediments33, 54.

EF is defined as follows:

EF = (M/Al)s /(M/Al)g (eq. 1)

Where (M/Al)s and (M/Al)g are the ratio of metal Al concentration in a sample and in the reference sample, respectively. By convention an EF valueranging from 0.5 to 1.5 implies a predominantly natural origin (e.g. weatheringproduct), while values greater than 1.5 indicate an important proportion of non-crustal materials or non-natural weathering processes (e.g. biota, point andnon-point pollution) 53, 55. EF values lower than 0.5 can reflect mobilization and loss of these elements relative to Al; they could indicate an overestimation of the reference metal contents.

In this study the concentration of Al in the Choapa basin in the rithron site was used as reference sample (eq. 1). The site was selected because it is locatedat high altitude in the Andes and is near the headwaters of the river, thus havinglow human intervention56.

EF values of five elements in all the dams and zones indicate null human influence; nevertheless it is important to consider that some places showedvalues > 0.5, for example the effluent of Recoleta reservoir; Al:0.96; Cu: 1.00;Fe: 0.82 and Zn 0.98 and in the effluent of La Paloma reservoir, Mn: 1.10.The analysis using the enrichment factor in other sites confirms the assumptionthat Cu: 31.9 and Zn: 22.0 are carried into the river by mining activities inthe Choapa River basin. In the Cachapoal basin there was a large enrichmentfactor of 17.6 for Cu, which may be associated with Cu-enriched ore from atreatment plant56

3.5.3 Sediment toxicity

The criteria to consider a sediment toxic are defined in the sediment quality guidelines (SQG), which provides thresholds to determine if the concentra-tion of any of the metals present in the sediments may involve risk for aquaticorganisms and consequently for human health. Threshold effect concentrations(TEC) and their probable effect concentrations (PEC) for sediment levels werereported by MacDonald et al., 2000. Where TEC correspond to the concentra-tion below which no adverse effects are observed on benthic organisms. The PEC intends to identify the contaminant concentrations above which harmfuleffects on benthic organisms were expected to occur frequently36.

Threshold effect concentrations (TEC) were: for Fe = 20000; Cu = 31.6; Mn = 450 and Zn = 121 mg kg'1 and probable effect concentration (PEC), forFe = 40000; Cu = 149; Mn = 1100 and Zn = 459 mg kg'1.

Cu concentration in affluents and effluents in all the reservoirs (Table 6) was higher than the TEC value and PEC except in the affluent of Cogotí andboth zones of La Paloma. Fe concentration was higher than the TEC value inthe entire reservoir and both zones except in the effluent of Recoleta; PEC waslower than the TEC values in all the dams and zones except for the effluent of Corrales. Mn concentration was higher than TEC and PEC in all the dams andin both zones, except the latter for La Paloma reservoir. Zn concentration washigher than TEC in all the dams and in both zones except in the effluent of LaPaloma and the effluent of Recoleta, where PEC showed lower values. Of thefive metals considered, Cu, Mn and Zn were the elements which present the greatest toxicity to microorganisms in these aquatic systems.

3.6 Principal component analysis (PCA)

The results of the principal components analysis is shown in the following figures.


Figure 6. PCA from surface water (A) and PCA from sediment (B): Cogotí affluent (1);
Cogotí effluent (2); La Paloma affluent (3), La Paloma effluent(4); Recoleta affluent (5),
Recoleta effluent (6), Corrales affluent (7); Corraleseffluent (8).

Figure 6A and 6B show that reservoirs are influenced by both metals and physical chemical characteristics of water or sediment.

In water (Figure 6A), Cogotí affluent (1) is influenced by Zn and Eh, Cogotí effluent (2), is influenced by Cu, La Paloma effluent (4), is influenced by DO and pH; Recoleta effluent (6) is influenced by EC, while Corrales effluent (8)is influenced by total P and Mn. In sediment (Figure 6B), La Paloma effluent(3) is influenced by Fe, Recoleta affluent (5) is influenced by Zn and Al, whileRecoleta effluent (6) is influenced by OC, EC and water soluble P.

4. CONCLUSIONS

Five of the ten metals analyzed were detected in water and sediments (soluble and total fractions), namely Al, Fe, Cu, Mn and Zn. The total content of metals in sediment was always greater than those found in the soluble orlabile fraction.

We found no differences in most of metal concentrations between affluents and effluents, indicating that the reservoirs do not act as a filter.

Metal concentrations in surface water were similar to concentrations found in other basins, except for Mn that had higher concentration, which may be dueto the mineralogical characteristics of the reservoirs.

Metal concentrations in sediment of the reservoirs showed relatively high values for the five metals compared to the Choapa basin; probably because of the hydrodynamic conditions of the sediment.

Metal concentrations of the soluble fraction of the sediment were between 5-70% of the total fraction: Mn>Cu>Zn>Fe>Al, showing that the solubilization of the metal depends on the sediment characteristics, especially pH.

Results of the risk assessment code (RAC), determined by percent soluble fraction of metals in relation to total fraction of metals in sediments suggestedthat Fe should be considered of medium risk in the affluents of Cogotí, whileCu and Zn showed medium risk in all the dams and both zones and Mn showedmedium to high risk.

The enrichment factor (EF) has shown to be an effective method to estimate the anthropogenic impact on sediments. In this study we determined the existence of five metal concentrations which have lithogenic origin.

Finally, Cu, Mn and Zn concentrations in both affluents and effluents of most of the reservoirs exceeded TEC and PEC values, thus these elementsshould be considered as potential toxic elements for organisms.

Principal components analysis showed that dams and zones are influenced by heavy metal concentrations and physicochemical characteristics of the water or sediments.

ACKNOWLEDGEMENTS

The authors are grateful to Fondecyt Project 1100341 and Proyecto Enlace Universidad de Chile 2013. DV thanks NC 120030, Basal Grant PFB 023 and ICM P05-002.

REFERENCES

1. U. Forstner, Int J. Environ. Anal. Chem., 51, 5, (1993).         [ Links ]

2. C. J. Vorosmarty, P B. McIntyre, M. O Gessner, D Dudgeeon, A Prusevich,P. Green, S. Glidden, S. E. Bunn, C. A. Sullivan, C. Rendy Liermann, P. M. Davies, Nature, 467, 555, (2010).         [ Links ]

3. J. Duffus, Pure Appl. Chem., 74, 793, (2002).         [ Links ]

4. U. Forstner, G. T. W. Wittmann, Heavy metal pollution in the aquaticenvironment 2nd ed. Springer: Berlin Germany, 1981; pp 71-109        [ Links ]

5. K. S. Murray, Environ. Geol. 27, 54, (1996).         [ Links ]

6. F. Prosi, Heavy metals in aquatic organism. In Metal Pollution in the Aquatic Environment F. Forstner, U. and Wittmann, eds Sprong Verlag,Berlin, 1981; pp 271-323        [ Links ]

7. S. M. Moalla, R. M. Awadallah, M. N. Rashed, M. E. Soltan, Hidrobiol. 364, 31, (1998).         [ Links ]

8. D. Purves, Trace elements contamination of the environment. ElsieverAmsterdan, ; pp 3-38 1985.         [ Links ]

9. W. Chen, S. K.Tan, J. H. Tay, Water, Air and Soil Pollution, 92, 273, 1996        [ Links ]

10. G. G. Geesy, L. Borstad, P. M. Chapman, Water Res. 18, 233, (1984).         [ Links ]

11. R. Vink, H. Behrentd, W. Solomon, Water Sci. Technol 39, 215, (1999).         [ Links ]

12. V. M. Dekov, F. Araújo, R. Van Grieken, V. Subramanian, Sci. TotalEnviron. 212, 89, (1998).         [ Links ]

13. L. Sigg, M. Sturm, D. Kistler, Limnology and Oceanography, 32, 112,(1987).         [ Links ]

14. R. Carignan, A. Tessier, Geochim and Cosmochim. Acta 52, 1179, (1988).         [ Links ]

15. P. Vaithiyanathan, A. Ramanathan, V Subramantan, Water, Air and SoilPoUution, 71, 13, (1993).         [ Links ]

16. A. K. Singh, S. l. Hasnain, D. K. Banerjee, Environ. Geol. 39, 90, (1999).         [ Links ]

17. S. I. Santos, C. A. B. García, E. A. Passos, E. A. Passos, J. P. Alves, J. Braz.Chem. Soc. 24, 246, (2013).         [ Links ]

18. W. Solomons, U. Forstner, Environ Technol. Lett. 1, 506, (1980).         [ Links ]

19. R. Baudo, H. Muntau, Chapter I. Lesser Known in-place pollutants and diffuse source problems. In Baudo, R., Giessey, J. P. and Muntau, H.Sediments Chemistry and toxicity of in place pollutants. Ann Arbor. LewisPublisher, 1990; pp 1-50.

20. H. E. Belkin, H. M. Sparck, USA Environ. Geol., 22, (1993).         [ Links ]

21. X-Y Wu, Y. F. Yang, Aquaculture Reserach, 41, 1377, (2010).         [ Links ]

22. A. J. de Groot, K. H. Zschuppel, W. Salomons, Hydrobiologia, 91, 689,(1982).         [ Links ]

23. C. M. Davidson, R. PThomas,S. E. McVey, R. Perala, D. Littlejohn, A. M.Ure, Analytica Chimica Acta, 291, 277, (1993).         [ Links ]

24. A. M. Ure, P Quevauviller, H. Muntauc, B. Griepink, InternationalJournal of Environmental Analytical Chemistry, 51, 135, (1993).         [ Links ]

25. D. L. Saunders, J. J. Meeuwing, A. C. J. Vincent, Conserv. Biol. 16, 30,(2002).         [ Links ]

26. F. J. Gottgens, J. E. Evans, J. Great Lakes Res., 2, 87, (2007).         [ Links ]

27. W. L. Graf, Water Resources Res., 35, 1305, (1999).         [ Links ]

28. M. Dynesius, C. Nilsson, C. Science, 266, 753, (1994).         [ Links ]

29. P Morais, Intern. J. Limnol. 44, 105, (2008).         [ Links ]

30. C. Nilsson, C. A. Reidy, M. Dynesius, C. Revenga, Science. 308, 405,(2005).         [ Links ]

31. J.G Rangel-Peraza, J. de Anda, _F. A. González-Farías, M. Rode,A. Sanhouse-García, Y. A. Bustos-Terrones, Environmental Monitoring and Assessment. 187, 134 (2015).         [ Links ]

32. K. W. Thorton, B. L. Kimmel, F. E. Payne, eds. Reservoir Limnology:Ecological perspectives. John Wiley & Sons, Inc., Somerset, New Jersey,1990, pp 246-256.         [ Links ]

33. G. Abrahim, R. Parker, Monit Assess. 136, 227, (2008).         [ Links ]

34. Ed. A. Passon, J. C. Alves, I. S. A. Dos Santos, J. P. H. Alves, C. A. B.Garcia, A. C. Spinola Costa, Microchem. J. 96, 50, (2010).         [ Links ]

35. G. Perin, L. Craboleda, M. Lucchese, R. Cirillo, I. Dotta, I., M. Zanette,Orio, Heavy metals speciation in the sediments of Northen Adriatic sea-anew approach for environmental toxicity determination: in Lekkas TD,editor: Heavy metal in the enviroment, 2010, pp. 45-46.         [ Links ]

36. D. D. MacDonald, C. Ingersoll, T. Berger, Arch. Environ. Contam.Toxicol., 39, 20, (2000).         [ Links ]

37. H. Mülhauser, L. Soto, R. Zahradnik, Int. J. Environ of Anal. Chem., 28,215, (1987).         [ Links ]

38. S. I. Simpson, G. E. Batley, A. A. S. Charlston, J. L. Dtauber, C. K. King,J. C. Chapman, R. V. Hyne, S. A. Gale, A. C. Roach, W. A. Maher, J.Sharyan. J, Handbook for Sediment Quality Assessment, (CSIRO Bangor,NSW), 2005; pp 9-19.         [ Links ]

39. L. C. Blackemore, P. L. Scarle, B. K. Daly, B. K. In Methods for Chemical of Analysis of Soils N. Z. Soil Bureau Scientific Report, 1987, pp 18-34.         [ Links ]

40. A. Sadzawka, M. A. Carrasco, R. Grez, M. L. Mora, H. Flores, A.Neuman, A. Métodos de Análisis de Suelos. Instituto de InvestigacionesAgropecuarias (INIA). Serie Actas INIA N° 34, 2006; pp 59-79.         [ Links ]

41. R Core Team. R (2013). A language and environment for statisticalcomputing. R Foundation for Statistical Computing, Vienna, Austria. URLhttp://www.R-proiect.ors/.         [ Links ] .

42. Broberg, A., McMasters, S. Uptake of copper and cadmium by tubificidsworms (oligochaeta) in different freshwater sediments. En: Heavy Metalsin the Hydrological Cycle. MAstruc. & J. N. Lester (Eds). Selper LtdLndon pp 265-272 (1988).         [ Links ]

43. F. Burriel Marti, F. Lucena-Conde, S. Arribas-Jimeno, J. HernadezMéndes. Química Analítica Cualitativa (Décimo Sexta Edición) (EditorialPARANINFO) Madrid, España. cap. XIX y XX, 2003.         [ Links ]

44. G. Klaver, B. Van Os., Ph. Negrel, E. Petelet-Girand, Environ. Pollut, 148,718, (2007).         [ Links ]

45. S. Wang, W. Rui, L. Xing, Earth Sci. Front. 981, 29, (2002).         [ Links ]

46. N. Roig, J. Sierra, M. Nadal, I. Moreno-Garrido, E. Nieto, M. Hampel,E. Perez Gallego, M. Schuhmacher, J. Blasco. Science of the TotalEnviroment, 503-504, 269, (2015).         [ Links ]

47. DGA (Dirección General de Aguas). Caracterización de SedimentosFluviales y su Relación con la Calidad del agua, 2008, pp 1-80.         [ Links ]

48. Q. Yongmin, Y Yang, G. Jiaguang, Z. Jiangang. Mar. Pollut. Bull. 68, 143(2913).         [ Links ]

49. A. Tessier, P G. C. Cambell, M. Bisson, M. Anal. Chem., 51, 844, (1979).         [ Links ]

50. I. Ahumada, A. Maricán, M. Retamal, C. Pedraza, C., L. Ascar, M. A.Carrasco, P. Richter, P, J. Braz. Chem. Soc. 21, 721, (2010).         [ Links ]

51. S. V. Copaja, M. X. Molina, R. M. Tessada, J. Chil. Chem. Soc., 57, 1986,(2014).         [ Links ]

52. R. Segura, V Arancibia, M. C. Zúñiga, P Pastén. Journal of GeochemicalExploration, 91, 71. (2006).         [ Links ]

53. Gu Y. G., Wang Z-H., Lu S-h., Jiang, S-J., Nu, D-S., Zu, Y-H. Environmenta lPollution 163, 248 (2012).         [ Links ]

54. Prego, R., Cobelo-Garcia, A. Twentieth century overview of heavy metalsin the Galician Rios (NW Iberian Peninsule). Environmental Pollution 121, 1425 (2003).         [ Links ]

55. Zhang J.; Liu C. L. Estuarine, Coastal and Shelf Science, 54, 1051, (2002).         [ Links ]

56. S. V. Copaja, G. Díaz, R. Toro, R. M. Tessada, P. Miranda, J. R. Morales,J. Chil Chem Soc. 57, 1400, (2012)        [ Links ]

 

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License