Archives of Environmental Contamination and Toxicology

, 56:416

Identification of a Potential Toxic Hot Spot Associated with AVS Spatial and Seasonal Variation

Authors

    • Instituto de Ciencias Marinas de Andalucía
  • A. Rodríguez
    • Instituto de Ciencias Marinas de Andalucía
  • J. Blasco
    • Instituto de Ciencias Marinas de Andalucía
Article

DOI: 10.1007/s00244-008-9206-6

Cite this article as:
Campana, O., Rodríguez, A. & Blasco, J. Arch Environ Contam Toxicol (2009) 56: 416. doi:10.1007/s00244-008-9206-6

Abstract

In risk assessment of aquatic sediments, much attention is paid to the difference between acid-volatile sulfide (AVS) and simultaneously extracted metals (SEMs) as indicators of metal availability. Ten representative sampling sites were selected along the estuary of the Guadalete River. Surficial sediments were sampled in winter and summer to better understand SEM and AVS spatial and seasonal distributions and to establish priority risk areas. Total SEM concentration (ΣSEM) ranged from 0.3 to 4.7 μmol g−1. It was not significantly different between seasons, however, it showed a significant difference between sampling stations. AVS concentrations were much more variable, showing significant spatial and temporal variations. The values ranged from 0.8 to 22.4 μmol g−1. The SEM/AVS ratio was found to be <1 at all except one station located near the mouth of the estuary. The results provided information on a potential pollution source near the mouth of the estuary, probably associated with vessel-related activities carried out in a local harbor area located near the station.

The toxicity of metals in sediments depends to a large extent on their type of binding forms. The simultaneously extracted metals/acid volatile sulfide ratio (SEM-AVS) is a mechanically determined method to predict sediment toxicity that accounts for metal bioavailability through the normalization to sediment sulfides that can react with cationic metals to create insoluble metal sulfides, thus reducing environmental availability and mobility of trace metals. AVS is extracted from anaerobic sediment with cold 1 M HCl, and divalent metals (usually Zn, Cu, Cd, Ni, and Pb are considered) released during this treatment are referred to as SEMs (Allen et al. 1993; Ankley et al. 1996; Morse et al. 1987).

The amount of AVS in sediments serves as a critical parameter in determining metal bioavailability and toxicity in sediments. The model predicts that when the molar concentration of AVS is higher than the molar concentration of SEM (i.e., when SEM minus AVS is <0, or the SEM/AVS ratio is <1), pore water metal concentrations should be very low and no metal toxicity would be expected. Conversely, metals toxicity may be observed if the measured AVS concentration is lower than the concentration of SEM because, potentially, some fraction of the metal may exist as bioavailable metal and cause toxicity (Di Toro et al. 1990; Campbell et al. 2006). The reason for stating that some fraction of the metals could potentially exist as bioavailable metal is that other constituents in the sediment pore water, such as amorphous iron oxyhydroxides and particulate organic matter, also bind the free metal, thereby reducing its bioavailability to benthic organisms and, ultimately, its toxicity (Burton et al. 2005).

Due to biological activity and variations in organic matter content, oxygen, and temperature, AVS concentrations vary temporally and spatially (Leonard et al. 1993; Di Toro et al. 1996), and the question is how these variations affect the SEM-AVS relationship, introducing a complicating factor in AVS-model applicability in the setting of environmental quality criteria (Morse and Rickard 2004). Actually, within the framework of setting environmental quality criteria for certain metals in sediments (Cu, Pb, Zn, Cd, Ni), the SEM-AVS ratio has been proposed as the primary standardization method to predict the bioavailability of these metals (Allen et al. 1993; Di Toro et al. 1990; Ankley et al. 1994; USEPA 1994). Therefore, the aims of this study were to investigate the spatial and seasonal distribution of both AVS and SEM in anoxic sediments of the estuary of the Guadalete River (southwestern Spain) and the relationship between them and, in a preliminary way, to establish priority risk areas.

Materials and Methods

Study Area

The Guadalete estuary flows into the Bay of Cádiz, located in the southwestern Iberian Peninsula (Fig. 1). Considered one of the most contaminated rivers in Andalusia, being subjected to intense urban, industrial, and agricultural contamination, the recovery of the Guadalete River began in 1988 with a plan that included the construction of different wastewater treatment plants. However, mismanagement has led to the partial or total paralysis of several installations during these years. For this study, 10 sampling sites, G1, G2, G3, S1, S2, S3, S4, S5, S6, and S7, were selected along the estuary (Fig. 2). Sampling sites G1–G3 were located approximately 1 km apart. G1 was located opposite the port of El Puerto de Santa Maria, in an area with commercial maritime traffic; G2 and G3 were situated farther inland, in areas surrounded by crops and salt marshes. Sites S1–S7 were situated at the mouths of drain channels or waste dumps. The sediments were situated within the tidal range. Salinity values ranged from eurihaline near the mouth of the estuary to mesohaline in the most inland site.
https://static-content.springer.com/image/art%3A10.1007%2Fs00244-008-9206-6/MediaObjects/244_2008_9206_Fig1_HTML.gif
Fig. 1

Map of the study area

https://static-content.springer.com/image/art%3A10.1007%2Fs00244-008-9206-6/MediaObjects/244_2008_9206_Fig2_HTML.gif
Fig. 2

Localization of sampling sites along the estuary of the Guadalete River

Sediment Collection and Pretreatment

Sediments were collected in summer (August 2002) and late winter (March 2003). Each of the 10 sampling sites (G1–G3, S1–S7) was sampled at three points: the riverbanks and the bed of the river, identified as left riverbank, right riverbank, and center. For example, point “G2 r” indicates the right riverbank of site G2. Grab samples were collected at each point in duplicate. Immediately after collection, three surficial (first 6 cm) sediment subsamples were obtained from each grab sample using a clean syringe (7 cm long × 2 cm diam.). Care was taken to choose undisturbed sediment. The three subsamples from each grab were extruded and sealed in a plastic ziplock bag, trying to preserve the bags air-free in order to promote anaerobic conditions, homogenized, and frozen at −20ºC within 3 h of collection, until analysis.

AVS and SEM Analysis

AVS concentrations in the sediment were determined according to the method proposed by Simpson (2001). SEM was extracted adding 15 mL of 10% HCl to ~1.5 g of wet sediment for 2 h, with continuous agitation. After extraction, centrifugation was performed at 3000 g for 30 min to separate hydrochloric acid extracts from solids (Langston and Spence 1994). SEM concentration is expressed as micromoles per gram of sediment, dry weight, after correction by the corresponding humidity value. SEM concentrations for Cd and Pb were determined using a GFAAS graphite furnace (PE 4100ZL) with Zeeman background correction and appropriate matrix modifiers. Zn, Cu, and Ni concentrations were determined on an ICP-OES (PE 2100DV). Blanks were run with each batch of analyses. Samples were analyzed in duplicate and averaged. To determine the analytical accuracy, reference estuarine sediment (BCR n.277) was analyzed simultaneously in combination with all metal analyses. Recovery efficiencies were within 95 to 116% compared with certified values. All solutions were prepared using Suprapur reagents and deionized water (18 mΩ cm). All glassware was soaked in 10% HNO3 (w/v) for 24 h and rinsed several times with deionized water before use.

Analysis of Organic Matter, Grain Size, and Porosity

The percentage organic matter content in sediments was determined by means of the loss on ignition (LOI) method (Dean 1974). After oven-drying of the sediments to a constant weight (24 h at 105°C), organic matter content is combusted at a temperature of 550°C for 16 h. The LOI is then calculated as the difference between the dry weight of the sample before combustion and that after heating to 550°C. The percentage of sediments <63 μm was determined by wet sieving through a 63-μm nylon sieve. Sediment porosity, Φ, was determined using density values according to the following equation (Gonzalez 1996):
$$\Upphi=\rho_{s}\,(1\, - f_{s})/\rho_{w}$$

where ρs represents the sediment density (bulk sediment weight/bulk sediment volume), fs represents the solid fraction of the sediment (dry sediment weight/bulk sediment weight), and ρw is water density, supposed to be equivalent to 1.

Statistics and Graphics

Distribution of variables was checked to determine normality by Shapiro-Wilks test by examining skewness and kurtosis values, and variance hypotheses by Levene test. Since the variables were not normally distributed, nonparametric statistical tests were performed. Krustal-Wallis (K) was applied to test the difference between sites into each season, while Friedman ANOVA (k) was used to compare mean values of each site between different seasons. Spearman nonparametric correlation analysis was performed to determine association between different variables. Statistical significance was set at p < 0.05, if not otherwise noted. Statistical tests were performed using Statgrafics Plus 5.0 and Statistica 5.0. A spatial analysis was performed on the variables. The spatial variability was assessed using the kriging gridding method of the Golden Software Surfer 8.0, which is a frequently used geostatistical method in soil science and occasionally applied to aquatic analysis (Van Griethuysen et al. 2003; Stein and Staritsky 1995).

Results

Sediment Characteristics

All sediments were anoxic, with a mean value of −215 ± 69 mV, and pH values relatively constant around 7.1 ± 0.2. Organic matter content, porosity, and grain size are listed in Table 1. For the most part sediments were characterized by a high content of fine-grained fraction, which ranged between 59 and 99%, in contrast with sandier sediments near the mouth. In fact, at site G1 the values decreased to 14 and 3%, respectively, for the left and right riverbank. Similarly, the lowest organic matter content and porosity values were observed in the left and right riverbanks of site G1. In general, organic matter content was in the range from 2 to 11% and porosity values were generally near 0.70 in most sediments.
Table 1

Sedimentary variables in surficial sediments of Guadalete estuary: means ± SD of summer and winter data

Site

Porosity

OM (%)

Grain size (%), fraction < 63 μm

G1

l

0.47 ± 0.02

2.6 ± 0.1

14 ± 8

c

0.79 ± 0.02

11.2 ± 1.0

89 ± 7

r

0.33 ± 0.05

1.7 ± 0.1

3 ± 0

S1

l

0.69 ± 0.01

5.6 ± 0.9

59 ± 13

c

nd

nd

nd

r

0.75 ± 0.03

8.9 ± 1.2

88 ± 7

S2

l

0.76 ± 0.02

8.5 ± 3.5

92 ± 7

c

0.76 ± 0.01

8.8 ± 2.1

83 ± 4

r

0.71 ± 0.05

9.0 ± 0.4

83 ± 11

G2

l

0.74 ± 0.04

8.9 ± 0.1

79 ± 10

c

0.77 ± 0.06

9.0 ± 0.7

92 ± 5

r

0.71 ± 0.10

7.6 ± 0.5

85 ± 8

S3

l

0.75 ± 0.02

8.9 ± 1.1

91 ± 4

c

0.64 ± 0.19

6.5 ± 2.1

64 ± 18

r

0.71 ± 0.10

8.2 ± 2.6

93 ± 1

S4

l

0.71 ± 0.06

7.2 ± 2.5

71 ± 29

c

0.67 ± 0.14

4.0 ± 3.7

66 ± 45

r

0.73 ± 0.04

6.0 ± 1.8

99 ± 1

S5

l

0.68 ± 0.02

5.4 ± 2.4

94 ± 4

c

0.70 ± 0.02

5.5 ± 1.3

93 ± 9

r

0.70 ± 0.00

5.9 ± 0.9

86 ± 1

S6

l

0.70 ± 0.05

4.8 ± 1.5

74 ± 17

c

0.77 ± 0.06

6.0 ± 0.9

83 ± 18

r

0.65 ± 0.03

5.6 ± 1.2

83 ± 11

S7

l

0.73 ± 0.01

5.8 ± 0.1

92 ± 2

c

0.74 ± 0.06

6.2 ± 2.6

99 ± 1

r

0.70 ± 0.04

4.9 ± 0.9

94 ± 7

G3

l

0.80 ± 0.11

5.9 ± 1.0

94 ± 2

c

0.72 ± 0.01

6.5 ± 0.9

96 ± 3

r

0.71 ± 0.02

5.5 ± 0.5

97 ± 2

Note: OM, organic matter; l, left riverbank; r, right riverbank; c, center of the river; nd, not determined

SEM and AVS

Seasonal SEM and AVS values in surficial sediment are listed in Tables 2 and 3. The concentration of total SEM (ΣSEM), taken as the sum of Zn, Cu, Cd, Pb, and Ni SEM values, ranged from 0.26 to 2.08 μmol g−1 in summer and from 0.43 to 4.68 μmol g−1 in winter. The greatest variation appeared at site G1 both in summer and in winter. Minimum values were found in the left riverbank (G1 l) and especially high concentrations in the right riverbank (G1r), up to threefold and eightfold higher in summer and in winter, respectively, compared with the mean ΣSEM concentration (~0.6 μmol g−1). Principal contributors to SEM, averaged among summer and winter values, were Cu and Zn, which made up about 40 ± 9 and 40 ± 8%, respectively, followed by Ni, at 13 ± 1%, Pb, at 6 ± 1%, and Cd, at only 0.2 ± 0.01%.
Table 2

SEM and AVS values in surficial sediments of Guadalete estuary in August 2002: means ± SD (μmol g−1)

Site

SEM/AVS

AVS

ΣSEM

SEM-Cd

SEM-Pb

SEM-Cu

SEM-Zn

SEM-Ni

G1

l

0.1

3.49 ± 0.86

0.31 ± 0.00

3.8E-4

0.02

0.07

0.20

0.02

c

0.1

6.95 ± 0.15

0.61 ± 0.07

8.6E-4

0.02

0.22

0.24

0.12

r

1.1

1.82 ± 0.23

2.08 ± 0.13

3.1E-4

0.18

1.05

0.82

0.03

S1

l

0.3

2.48 ± 0.27

0.65 ± 0.02

1.0E-3

0.04

0.16

0.41

0.04

c

nd

nd

nd

nd

nd

nd

nd

nd

r

0.1

6.49 ± 1.63

0.74 ± 0.02

1.2E-3

0.04

0.22

0.40

0.07

S2

l

0.1

9.84 ± 1.80

0.80 ± 0.03

1.1E-3

0.04

0.26

0.42

0.07

c

0.1

4.77 ± 0.16

0.55 ± 0.00

9.6E-4

0.03

0.21

0.23

0.08

r

0.1

7.56 ± 0.33

0.70 ± 0.05

8.0E-4

0.05

0.21

0.37

0.06

G2

l

0.2

3.84 ± 0.47

0.72 ± 0.06

1.1E-3

0.04

0.24

0.37

0.06

c

0.1

3.56 ± 0.64

0.47 ± 0.00

9.0E-4

0.02

0.18

0.19

0.07

r

0.2

4.29 ± 0.04

0.71 ± 0.01

1.2E-3

0.04

0.24

0.36

0.07

S3

l

0.1

5.69 ± 0.70

0.69 ± 0.03

8.9E-4

0.04

0.23

0.34

0.07

c

0.1

2.06 ± 0.08

0.29 ± 0.00

5.0E-4

0.02

0.10

0.13

0.04

r

0.2

3.35 ± 0.21

0.69 ± 0.05

1.1E-3

0.04

0.26

0.31

0.08

S4

l

0.1

6.12 ± 0.65

0.62 ± 0.01

1.2E-3

0.03

0.21

0.30

0.07

c

0.1

3.76 ± 0.59

0.26 ± 0.01

6.5E-4

0.01

0.09

0.12

0.03

r

0.1

7.16 ± 0.26

0.67 ± 0.03

1.3E-3

0.03

0.24

0.31

0.07

S5

l

0.1

4.37 ± 0.36

0.53 ± 0.01

1.0E-3

0.02

0.22

0.22

0.06

c

0.1

5.56 ± 0.25

0.44 ± 0.02

1.1E-3

0.02

0.18

0.15

0.08

r

0.1

4.57 ± 2.36

0.62 ± 0.02

1.2E-3

0.03

0.23

0.28

0.07

S6

l

0.2

2.82 ± 0.12

0.46 ± 0.00

8.2E-4

0.03

0.16

0.21

0.06

c

0.1

5.30 ± 1.31

0.45 ± 0.02

9.3E-4

0.02

0.17

0.17

0.08

r

0.1

4.62 ± 0.03

0.48 ± 0.04

6.5E-4

0.03

0.15

0.22

0.08

S7

l

<0.1

17.66 ± 4.23

0.63 ± 0.02

1.3E-3

0.03

0.22

0.28

0.09

c

<0.1

11.24 ± 1.39

0.58 ± 0.05

1.0E-3

0.03

0.21

0.27

0.07

r

<0.1

11.84 ± 2.43

0.55 ± 0.01

1.2E-3

0.03

0.22

0.20

0.09

G3

l

<0.1

22.45 ± 2.81

0.64 ± 0.00

1.3E-3

0.04

0.17

0.33

0.10

c

0.1

4.28 ± 0.02

0.56 ± 0.01

1.1E-3

0.03

0.20

0.26

0.07

r

<0.1

15.26 ± 1.44

0.57 ± 0.02

1.0E-3

0.03

0.19

0.26

0.08

Note: AVS, acid-volatile sulfide; SEM, simultaneously extracted metal; l, left riverbank; r, right riverbank; c, center of the river; nd, not determined

Table 3

SEM and AVS values in surficial sediments of Guadalete estuary in March 2003: means ± SD (μmol g-1)

Site

SEM/AVS

AVS

ΣSEM

SEM-Cd

SEM-Pb

SEM-Cu

SEM-Zn

SEM-Ni

G1

l

0.3

1.47 ± 0.08

0.43 ± 0.00

4.7E-4

0.04

0.11

0.25

0.03

c

0.5

1.05 ± 0.04

0.53 ± 0.01

1.2E-3

0.04

0.25

0.15

0.09

r

3.8

1.24 ± 0.04

4.68 ± 0.75

5.3E-4

0.40

2.68

1.59

0.03

S1

l

0.5

1.27 ± 0.01

0.59 ± 0.01

1.0E-3

0.04

0.21

0.28

0.06

c

0.3

1.41 ± 0.07

0.51 ± 0.00

9.0E-4

0.03

0.23

0.16

0.08

r

0.1

8.96 ± 0.29

0.65 ± 0.03

9.6E-4

0.04

0.26

0.27

0.08

S2

l

0.4

1.75 ± 0.21

0.65 ± 0.04

8.3E-4

0.04

0.27

0.26

0.08

c

0.5

1.02 ± 0.06

0.56 ± 0.01

8.5E-4

0.03

0.26

0.17

0.09

r

0.7

0.86 ± 0.10

0.57 ± 0.01

9.4E-4

0.04

0.24

0.21

0.07

G2

l

0.6

1.03 ± 0.12

0.61 ± 0.05

8.0E-4

0.04

0.26

0.24

0.07

c

0.6

1.02 ± 0.25

0.58 ± 0.08

1.1E-3

0.06

0.26

0.15

0.09

r

0.3

1.92 ± 0.08

0.50 ± 0.01

7.5E-4

0.03

0.20

0.21

0.05

S3

l

0.4

1.54 ± 0.09

0.58 ± 0.01

9.8E-4

0.04

0.25

0.23

0.07

c

0.4

1.23 ± 0.11

0.51 ± 0.02

9.8E-4

0.03

0.25

0.13

0.09

r

0.6

0.90 ± 0.22

0.58 ± 0.06

8.9E-4

0.03

0.27

0.19

0.08

S4

l

0.8

0.65 ± 0.01

0.50 ± 0.02

1.4E-3

0.03

0.24

0.15

0.08

c

0.3

1.45 ± 0.28

0.48 ± 0.02

1.3E-3

0.03

0.23

0.12

0.09

r

0.4

1.06 ± 0.00

0.47 ± 0.00

1.3E-3

0.03

0.23

0.14

0.07

S5

l

0.4

1.28 ± 0.00

0.49 ± 0.01

1.4E-3

0.03

0.24

0.14

0.07

c

0.4

1.43 ± 0.19

0.52 ± 0.01

1.3E-3

0.03

0.27

0.13

0.09

r

0.1

0.75 ± 0.23

0.48 ± 0.00

1.0E-3

0.03

0.24

0.14

0.07

S6

l

0.1

7.71 ± 0.62

0.61 ± 0.00

8.9E-4

0.04

0.28

0.21

0.07

c

0.1

3.74 ± 1.08

0.55 ± 0.01

9.8E-4

0.02

0.28

0.15

0.09

r

0.5

1.02 ± 0.97

0.51 ± 0.01

9.1E-4

0.03

0.25

0.16

0.08

S7

l

0.3

1.79 ± 0.50

0.58 ± 0.01

1.0E-3

0.03

0.27

0.20

0.07

c

0.4

1.33 ± 0.01

0.54 ± 0.06

1.1E-3

0.03

0.27

0.16

0.09

r

0.3

2.24 ± 0.03

0.67 ± 0.09

1.0E-3

0.03

0.33

0.21

0.10

G3

l

0.5

0.89 ± 0.13

0.48 ± 0.01

1.0E-3

0.03

0.22

0.15

0.08

c

0.2

2.63 ± 0.27

0.54 ± 0.00

1.1E-3

0.03

0.25

0.17

0.08

r

0.4

1.35 ± 0.37

0.49 ± 0.01

8.4E-4

0.03

0.23

0.15

0.08

Note: AVS, acid-volatile sulfide; SEM, simultaneously extracted metal; l, left riverbank; r, right riverbank; c, center of the river

Greater variability was observed for AVS (Tables 2 and 3). The concentrations ranged from low to moderately high in summer between 1.82 and 22.45 μmol g−1 and were significantly decreased in winter, ~1 μmol g−1, in 84% of sediments analyzed. As occurs for SEM levels and other sedimentary variables, extreme AVS concentrations were observed at site G1r, where the lowest AVS value in the estuary was monitored in summer. However, a similar trend was not detected in winter.

Table 4 presents the results of statistical analysis performed on ΣSEM and AVS. Analysis showed significant temporal variation in AVS concentrations between summer and winter determined by the Freidman ANOVA. Statistical analysis also revealed that temporal variation was not significant for ΣSEM. Kruskal-Wallis tests highlighted a significant spatial variability between different sampling sites for AVS and ΣSEM both in summer and in winter.
Table 4

Results of ΣSEM and AVS statistical analysis using nonparametrical statistical tests: boldface values indicate significant difference at < 0.05

 

AVS

ΣSEM

Friedman ANOVA (k)

0.825

0.017

p-value

0.000

0.137

 

Summer

Winter

Summer

Winter

Kruskal-Wallis (K)

54.24

50.61

55.11

53.77

p-value

0.002

0.008

0.001

0.003

Note: AVS, acid-volatile sulfide; SEM, simultaneously extracted metal

The correlations between variables are presented in Table 5. Significant correlations were observed between porosity and grain size and between porosity and organic matter content. To improve statistical analysis data were checked for the presence of outliers. If statistical analysis was performed considering outlier data from site G1, a significant correlation was also observed between SEM and organic matter content (Spearman R = 0335; p = 0.014; valid N = 53). AVS and SEM were not significantly correlated with any other sedimentary variables.
Table 5

Relationship between variables expressed by nonparametric Spearman correlation coefficient: boldface values indicate significant correlation at p < 0.05

Variables

N

Spearman R

p-value

AVS & OM

59

−0.046

0.728

AVS & porosity

59

−0.004

0.977

AVS & grain size

57

0.137

0.307

SEM & OM

59

0.256

0.050

SEM & porosity

59

0.183

0.166

SEM & grain size

57

0.110

0.417

OM & porosity

59

0.667

0.000

OM & grain size

57

0.082

0.541

Porosity & grain size

57

0.377

0.003

Note: AVS, acid-volatile sulfide; OM, organic matter; SEM, simultaneously extracted metal. N, number of cases

Spatial and Temporal Distribution

Surficial distribution of ΣSEM is shown in Fig. 3a and b. ΣSEM displays a gradient, with a higher concentration in summer at site G1. This gradient became more abrupt in winter. Throughout the rest of the estuary, the concentration remained stable and time independent, commonly at low levels, with a region of minimum values between station S3 and station S4.
https://static-content.springer.com/image/art%3A10.1007%2Fs00244-008-9206-6/MediaObjects/244_2008_9206_Fig3_HTML.gif
Fig. 3

Geochemical maps of simultaneously extracted metal (SEM) distribution (μmol g−1) along the Guadalete estuary. Distribution in (a) summer and (b) winter

Spatial and temporal variations of AVS concentration are presented in the surficial distribution map in Fig. 4a and b. AVS concentration was generally higher than ΣSEM throughout the estuary with the exception of station G1. In summer, the AVS concentration in this area displayed the inverse of the gradient exhibited for ΣSEM; therefore on the right riverbank of site G1, the AVS concentration was lower than the ΣSEM concentration. This observation is represented by the small striped area in Fig. 4a and in the detail (G1r) obtained by superimposing the corresponding AVS and SEM distribution maps. In winter, AVS concentrations decreased dramatically throughout the estuary. This decline resulted in extension of the striped area near site G1, with the SEM/AVS ratio being >1. At site S1, the AVS distribution showed a similar trend in both summer and winter, although the gradient was sharper in winter as a result of a decrease in AVS concentration on the left riverbank, while an increase was observed for the right riverbank. Due to seasonal variation in AVS concentration, sediments between sites S2 and G2 presented winter SEM/AVS ratios >1, also represented by the striped area. The left riverbank of site S6 was the only site to show a threefold increase in AVS concentration in winter. Minimum AVS concentrations were observed during winter for the area between site S4 and site S5. Maximum AVS concentrations were reported during summer, corresponding to the area near sites S7 and G3 located farther inland, although this maximum was not sustained during winter.
https://static-content.springer.com/image/art%3A10.1007%2Fs00244-008-9206-6/MediaObjects/244_2008_9206_Fig4_HTML.gif
Fig. 4

Geochemical maps of acid-volatile sulfide (AVS) distribution (μmol g−1) along the Guadalete estuary. (a) Distribution in summer and detail of site G1. (b) Distribution in winter. Striped areas represent sediments with simultaneously extracted metal (SEM)/AVS ratio >1

Discussion

Spatial and Temporal Variability

The distinct nature of SEM and AVS may help to explain the range of behavior observed for these variables. With regard to AVS concentration, the increase exhibited during summer and its temporal cycle have been reported in several studies (Boothman et al. 2001; Grabowski et al. 2001; van den Hoop et al. 1997; Mackey and Mackay 1996; Howard and Evans 1993; Boothman and Helmstetter 1992), these being related to warmer sediment and water during summer, which may increase the activity of sulfate-reducing bacteria (SRB) in sediments, hence increasing the reduction rate and, consequently, AVS production. For example, van den Berg et al. (1998) observed that high summer temperatures in combination with larger amounts of freshly deposited and highly degradable organic matter resulted in a higher sulfate reduction rate in freshwater sediments. Extrapolating from these studies, such temperature variation may explain the general AVS increase throughout the Guadalete estuary during summer. Furthermore, the combination of physical and biological factors as well as temperature, oxygen, and organic content, may be responsible for determining the varied distribution of AVS in both summer and in winter. In this study, as previously reported (Mackey and Mackay 1996; Leonard et al. 1996), the data did not support a significant correlation between AVS concentration and organic matter or fine-grained sedimentary fraction. This was due to AVS exhibiting abrupt seasonal fluctuations, whereas the organic matter content, generally more refractory, and the fine-grained fraction did not present seasonal changes. Nevertheless, it is plausible to propose that the low AVS concentration, especially observed in summer for the right riverbank of station G1, may be related to the synergistic effect of a very low organic matter content (~1.7%) and extremely low percentage of fine-grained sediments (3 versus 70% over mean), thus affecting AVS production in the sediments. The high AVS concentration observed during the year on the right riverbank at station S1 may be ascribed to sulfate and organic matter input, presumably due to its being situated in front of a tidal waterway used as an urban discharge point. Meanwhile, the area between station S2 and station S4 is subjected to intense mollusk harvesting; in addition to the effect of bioturbation produced by the burrowing activity of benthonic fauna (Peterson et al. 1996; Zhuang et al. 1994), the mixing of sediments induced by harvesting enhances oxidation. This factor would probably affect AVS production and it could explain the pattern, highlighted by the modeling, where SEM concentrations were higher than AVS in winter, as indicated by the striped area in Fig. 4b. On the whole, AVS concentrations observed along the estuary can be considered moderately low but in line with literature values for surficial estuarine sediments reported as part of the EMAP estuary program (Leonard et al. 1996), which were found to range from 0.05 up to 175 μmol g−1 dry sediment.

Unlike sulfides, the metal concentration does not depend strictly and directly on bacterial activity, and likewise, temperature is not a limiting factor for metal distribution; therefore the concentration is seasonally independent. Significant spatial variation of the SEM concentration revealed by statistical analysis seemed not to have chemical and/or ecological relevance, apparently due to environmental heterogeneity. In effect, natural systems inherently possess spatial and temporal variability, and much care must be taken to identify the difference detected between two locations as an anthropogenic impact. Therefore, one of the most important issues in establishing causal relationships between sediment contaminants and ecological effects is the separation of natural spatial-temporal variation from anthropogenic variation (Day et al. 1997). The only exception was represented by the values measured for the right riverbank of station G1, where SEM concentrations were highest throughout the year. This suggests that the results do not depend on natural variability but are the result of a contamination source. No channels discharge directly to this area, meaning that the potential contamination is due to shipping activities and the sewage discharge of the harbor area adjacent to the station. There is a variety of activities associated with vessels that may produce metal inputs, including sacrificial anodes for the control of vessel corrosion, motor exhaust, and antifouling paint hull coatings. Such coatings are the most significant source of metals, in particular, copper (copper content ranges from 20 to 76%), given their wide use for retarding the growth of algae and other encrusting organisms on the hulls and other submerged parts of vessels (Schiff et al. 2004). Analysis of the contribution of each metal to the total SEM concentration at site G1r revealed that concentrations of Pb and Cu in summer were 0.18 and 1.05 μmol g−1, respectively. These concentrations increased in winter, reaching 0.40 μmol g−1 for Pb and more than doubling for Cu, at 2.69 μmol g−1, probably owing to ship repair activity being more intense in this period.

The contamination identified seems to be restricted to the small area referred to on the right riverbank of site G1. Further sampling and hydrological data would be necessary to confirm the effective pollution range.

Conclusion

In general, the results suggest that sediments are able to trap metals through the production of metal sulfides and that often this capacity is far from being exhausted, given that such sediments are able to sustain a substantial metal removal rate from porewater. For such results, little or no hazard is presented. However, the aforementioned capacity was exceeded for certain areas, especially in winter when the AVS concentration decreased, thereby revealing spatial and time dependence. This condition allows detection of a potential toxic hot spot next to the mouth of the estuary, probably associated with vessel-related activities carried out in a local harbor area located close to the sampling site. Nevertheless, a single approach to quality assessment is insufficient. In this case, the associated risk represented by the sediments in this area has to be determined in a weight of evidence sediment assessment procedure because, in complex sediment systems, a variety of factors can mitigate (or enhance) contaminant toxicity. For this reason, biological testing of sediments was carried out and the results are being analyzed in order to be shortly divulged. However, although these results are not decisive, the SEM/AVS model allows separation of sediments that are not of environmental concern from those that are, to a moderate or significant extent. This strategy allows managers and researchers to address and reduce risk areas, thereby saving time and effort (Solomon et al. 1997; Chapman et al. 1997).

Copyright information

© Springer Science+Business Media, LLC 2008