- Email: [email protected]
A biomonitoring study: trace metals in algae and molluscs from Tyrrhenian coastal areas
ARTICLE IN PRESS
Environmental Research 93 (2003) 99–112
A biomonitoring study: trace metals in algae and molluscs from Tyrrhenian coastal areas$ Marcelo Enrique Contia, and Gaetano Cecchettib a
Dipartimento di Controllo e Gestione delle Merci e del loro Impatto sull’Ambiente, Universita` di Roma ‘‘La Sapienza’’, Via del Castro Laurenziano 9, 00161, Rome, Italy b Centro per le Valutazioni Ambientali delle Attivita` Industriali, Facolta` di Scienze Ambientali, Universita` degli Studi di Urbino, Campus Scientifico Sogesta, 61029 Urbino (Pu), Italy Received 30 July 2002; received in revised form 16 December 2002; accepted 2 January 2003
Abstract Marine organisms were evaluated as possible biomonitors of heavy metal contamination in marine coastal areas. Concentrations of Cd, Cr, Cu, Pb, and Zn were measured in the green algae Ulva lactuca L., the brown algae Padina pavonica (L.) Thivy, the bivalve mollusc Mytilus galloprovincialis Lamarck, and the two gastropod molluscs Monodonta turbinata Born and Patella cerulea L. collected at six coastal stations in the area of the Gulf of Gaeta (Tyrrhenian Sea, central Italy). The coastal area of the Regional Park of Gianola and Monte di Scauri (a ‘‘Protected Sea Park’’ area) was chosen as a control site. Seawater samples were also collected in each site to assess soluble and total metal concentrations and to gain additional information on both the environmental conditions of the area and possible bioaccumulation patterns. Metal concentrations detected in algae and molluscs did not show significant differences among all stations studied. Moreover, statistical analyses (ANOVA, multiple comparison tests, cluster analysis) showed that the Sea Park station was not significantly different from the others. The hypothesis that the Protected Sea Park would be cleaner than the others must therefore be reconsidered. Data from this study were also compared with those previously obtained from uncontaminated sites in the Sicilian Sea, Italy. The results show clearly differences between these two marine ecosystems. The species examined showed great accumulations of metals, with concentration factors (CFs) higher than 10,000 with respect to the concentrations (soluble fractions) in marine waters. Metal concentrations recorded in this area may be used for background levels for intraspecific comparison within the Tyrrhenian area, a body of water about which information is still very scarce. r 2003 Elsevier Science (USA). All rights reserved. Keywords: Biomonitoring; Trace metals; Marine ecosystems; Tyrrhenian Sea
1. Introduction The use of marine organisms as bioindicators for trace metal pollution is very common these days. Algae and molluscs are among the organisms most used for this purpose (Rainbow, 1995). Macroalgae are able to accumulate trace metals, reaching concentration values that are thousands of times higher than the corresponding concentrations in sea water (Bryan and Langston, 1992; Fo¨ster, 1976; Rai et al., 1981). Algae bind only free metal ions, the concentrations of which depend on $
This work was supported by a research grant from the Ministry of Instruction, University, and Scientific Research (MURST), year 2000– 2001. Corresponding author. Fax: +39-06-445-1566. E-mail address: [email protected] (M.E. Conti).
the nature of suspended particulate matter (Luoma, 1983; Seeliger and Edwards, 1977; Volterra and Conti, 2000), which, in turn, is formed by both organic and inorganic complexes. Moreover, algae satisfy all the basic requirements of bioindicators: they are sedentary, their dimensions are suitable, they are easy to identify and to collect, they are widely distributed, and they accumulate metals to a satisfactory degree (Conti, 2002). The most used algae in biomonitoring studies in the Mediterranean Sea are those of the genus Ulva. Ulva rigida C. Ag. is the species of Ulva most widespread along the Mediterranean coasts (Haritonidis and Malea, 1995, 1999). The species Ulva lactuca L. is especially present in polluted locations near the surface, to a depth of 4–5 m (Favero et al., 1996; Muse et al., 1999; Volterra and Conti, 2000). Despite its distribution along the Mediterranean coasts, the brown algae Padina pavonica
0013-9351/03/$ - see front matter r 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S0013-9351(03)00012-4
ARTICLE IN PRESS 100
M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
has not been used frequently in biomonitoring studies; its suitability as a bioindicator has been proposed only recently (Campanella et al., 2001). Bivalve molluscs, along with the Ulva genus, play an important role as bioindicators for trace metal pollution and appear more and more often in global monitoring programs (Rainbow, 1995; Sericano, 2000). Mussels can accumulate and integrate concentrations of several metals in seawater for relatively long intervals. They also assimilate trace metals from their food and from the ingestion of inorganic particulate material (Phillips, 1977; Struck et al., 1997). In recent years, researchers have focused their attention on the identification of other possible bioindicators for trace metal pollution, such as the gastropod molluscs Monodonta turbinata and Patella cerulea. This is because it is necessary to identify a wider range of bioindicators and thus expand current understanding of different bioaccumulation strategies for trace metals. The objective of this work was to gather more information on the use of five species selected as cosmopolitan biomonitors for the Tyrrhenian area: the green algae U. lactuca L., the brown algae P. pavonica (L.) Thivy, the mussel Mytilus galloprovincialis Lam., and the two gastropod molluscs M. turbinata Born and P. cerulea L. For this purpose we assessed the possible effect of anthropogenic activities on a coastal area of the Gulf of Gaeta, Tyrrhenian Sea (central Italy). This area, for which the literature is quite scant, is actually not completely free from industrial activities and therefore is not completely uncontaminated. Moreover, it is affected by the presence of two fairly big towns: Formia and its harbor (37,500 inhabitants) and the town of Gaeta (24,000 inhabitants). The coastal area of the Regional Park of Gianola and Monte di Scauri was also sampled as a possible monitoring site. The park, with an area of 285 ha was established in 1987. Furthermore, the stretch of coast just off the shoreline of the Regional Park is protected by the establishment of an ‘‘Oasi Blu’’ (Blue Oasis, i.e., a protected sea area) through a public concession to WWF Italy. In this area, covering 50,000 m2, motor boats and fishing are not allowed. Generally, when studying sea pollution, some clearly contaminated stations are singled out and compared with a ‘‘clean’’ monitoring site. In this study we examined supposedly similar sites in order to identify useful ‘‘background levels’’ as a reference for intraspecific comparisons within the Tyrrhenian area, on which information has been very scarce. For a broader picture of both the environmental conditions of the area under investigation and possible bioaccumulation patterns, seawater samples were collected to assess soluble and total metal concentrations. Although occasional analysis of seawater samples cannot represent reliable measurements of the mean
contaminant concentrations in each site (Phillips, 1977), indicative information can be drawn, provided that the data taken from all sampling stations are averaged.
2. Materials and methods 2.1. Sampling and sample pretreatment Samples were collected in May 2000, with a total of six stations distributed in the coastal area of the Gulf of Gaeta, central Italy. The sampling stations were chosen according to their exposure to sea currents (Fig. 1). According to criteria of homogeneity (Conti, 2002; Conti et al., 2002), at each sampling station there were at least two sampling sites. The sampling was performed homogeneously; that is, the selected individuals belonged to the same life cycle, had similar dimensions (and weights, for molluscs), and were always collected at the same depth and distance from the shoreline. Of the six sampling stations, one (protected sea area, WWF Oasis) did not appear to be affected greatly by anthropic activities. Samples of U. lactuca and P. pavonica were handpicked in the subtidal zone at a depth of about 2–3 m. Care was taken to choose thalli all at a similar stage of development. At each station, 10 samples of algae were sampled, and those abundantly covered with epiphyta were rejected. The samples were washed in seawater at the sampling site and transferred to the laboratory in precleaned polyethylene bags under refrigerated conditions. Upon arrival at the laboratory, they were thoroughly cleaned and any epiphyta and sediments were carefully removed with nylon brushes under tap water for a few seconds. Algal material was rapidly rinsed in deionized water (Milli-Q, Millipore Corp., Bedford, MA, USA) to minimize any possible metal loss during the procedure and then pulverized. This was then frozen (201C) until analysis. Algal samples for each station were pooled, and five subsamples of about 0.7 g [dry wt (d.w.)] were digested in a microwave oven (MDS 81D CEM, Cologno al Serio, Italy) after 12 h of cold premineralization with ultrapure concentrated HNO3 Merck Suprapur (Darmstadt, Germany). Individuals of M. galloprovincialis, P. cerulea, and M. turbinata were collected in the tidal zone from each station according to their availability. Then, for a 24-h period, molluscs were placed in filtered seawater collected at the corresponding station to allow the depuration of the particulate matter residues present in the mantle cavity and digestive tract. The soft parts of the molluscs, carefully extracted with a plastic spatula, were individually mineralized with a mixture of concentrated HNO3/H2O2 (5+2 mL). All chemicals used in sample treatments were of ultrapure grade (HNO3, H2O2 30%, Merck Suprapur).
ARTICLE IN PRESS M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
101
Fig. 1. The study area (Gaeta Gulf, Tyrrhenian Sea, Italy) and the sampling stations: (1) Formia Vindicio beach; (2) Porto Romano (Roman harbour); (3) Sant’Agostino beach; (4) Sassolini beach (Protected Sea Park); (5) Santa Croce river; (6) Serapo beach.
Ultrapure water (Milli-Q System, Millipore) was used for all solutions. All glassware was cleaned prior to use by soaking in 10% v/v HNO3 for 24 h and rinsed with Milli-Q water. The standard solutions of metals were prepared from stock standard solutions of ultrapure grade supplied by Merck. The calculation of d.w. on the examined species (10 replicates for each location) was carried out by oven drying at 1051C until constant weight.
Water samples were collected with sampling bottles (precleaned and rinsed twice with seawater) at a depth of 2 m. Temperature and salinity were recorded. Five samples for each location of 1 L (precleaned polyethylene bottles) were filtered through acid-cleaned 0.45-mm membrane filters, acidified, and stored at 41C for soluble metal analysis. The second five samples, previously acidified, were stored at 41C for subsequent mineralization in a microwave oven and analyzed for total metal
ARTICLE IN PRESS M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
102
concentrations. Water samples were preconcentrated before analysis with the ammonium 1-pyrrolidinedithiocarbamate (APDC) complexation method (Bruland et al., 1979; IRSA, 1984). The procedure is described elsewhere (Campanella et al., 2001).
tissue), and 403 (seawater). One sample of reference material and blanks were included in each analytical batch. Results were in agreement with certified values, and the standard deviations were low, proving good repeatability of the method (Table 2). Mean metal concentrations were calculated with standard deviations. The standard deviations of pooled samples (algae) refer to the variability within different replicates. For mollusc analysis, the standard deviations represent the metal content variation among individuals. All molluscs that were collected and mineralized belong to a narrow and homogeneous range of weight and dimension. The one-way analysis of variance (ANOVA) (log-transformed data) was applied to test the differences between mollusc metal concentrations in different sites and species. Multiple comparison tests were carried out to assess the most significant impact levels in the examined stations. Finally, cluster analysis allowed the assessment of homogeneous groups for the impact of the molluscs observed in this study.
2.2. Determination of heavy metals Heavy metal concentrations were determined by graphite furnace atomic absorption spectrometry (Perkin–Elmer Model 1100B equipped with an HGA-700 graphite furnace and a Perkin–Elmer Model AS-70 autosampler, Perkin–Elmer Italia, Monza, Italy). NH4PO4 was used as matrix modifier for Cd and Pb, Mg (NO3)2 was used for Cr, and the method of standard addition was used (Pb, Cd, Cr). Whenever high levels occurred, Cu and Zn in the biota were determined by flame atomic absorption spectrometry (Varian Techtron Model AA-475, Varian Italia, Milan, Italy). Detection limits and coefficients of variation for the different matrices are reported in Table 1. 2.3. Analytical quality control and statistical analysis
3. Results
The accuracy of measurements was tested with Community Bureau of Reference (BCR) certified reference materials CRM 279 (sea lettuce), 278 (mussel
Metal concentration data in the biota are shown in Tables 3 and 4. Metal concentrations detected in P. pavonica and U. lactuca were quite homogeneous and
Table 1 Limits of detection (LOD)a and coefficients of variation (CV)b for the analysis performed by graphite furnace atomic absorption spectrometrya Macrophytes 1
LOD (mg g Cd Cr Cu Pb Zn
Molluscs d.w.)
0.04 0.04 0.10c 0.06 0.12c
Seawater 1
CV
LOD (mg g
1.9 4.0 3.3 2.9 4.8
0.05 0.05 0.15c 0.07 0.17c
d.w.)
CV
LOD (ng L1 d.w.)
CV
2.5 1.8 4.0 3.7 5.4
20 17 42 22 54
4.5 2.7 3.0 2.8 3.6
a
Calculated on the basis of 20 determinations of the blanks as 3 times the standard deviation of the blank. %, relative to X10 determinations performed on the same sample. c LOD for the analysis performed by flame atomic absorption spectrometry (mg g1 d.w.): Cu, 0.95 (macrophytes) and 1.15 (molluscs); Zn, 1.43 (macrophytes) and 1.62 (molluscs). b
Table 2 Analysis of certified reference materials: certified values and found values (means7SD)a Metal
Cd Cr Cu Pb Zn a
CRM 279 (sea lettuce) (mg g1 d.w.)
CRM 278 (mussel tissue) (mg g1 d.w.)
CRM 403 (seawater) (ng L1)b
Certified
Found
Certified
Found
Certified
Found
0.27470.022 (10.7)c 13.1470.37 13.4870.36 51.371.2
0.2870.03 9.170.5 13.2370.41 14.0570.62 51.472.0
0.3470.02 0.8070.08 9.6070.16 1.9170.04 7672
0.3770.04 0.7270.04 9.2370.26 2.0770.08 7873
20.3572.10 (108.6)c 256.4724.3 25.0875.36 17387197
22.4070.81d 104.472.2 264.6711.3 26.7171.31d 17817112
Number of replicates: CRM 279=22, CRM 278=35, CRM 403=5. Original values given on a w/w basis were recalculated on a w/v basis. c Not certified values. d Values close to the detection limits. b
ARTICLE IN PRESS M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
103
Table 3 Concentrations of metals in the thalli of P. pavonica and U. lactuca (means7SD) (mg g1 d.w.)
Sampling station
Cd
Cr
Cu
Pb
Zn
P. pavonica 1 2 3 4 6
0.5270.09 0.4570.06 0.4770.11 0.3970.12 0.6670.10
3.9670.52 3.2070.64 2.8870.65 3.3470.73 3.8670.63
13.271.0 11.872.0 12.471.4 11.972.2 12.171.7
4.8270.92 3.6770.79 3.7370.53 3.0470.46 4.6471.16
5477 4578 4977 5078 5676
U. lactucab 1 3 4 5 6
0.2070.05 0.1870.05 0.1770.03 0.137 0.04 0.217 0.05
1.5870.25 2.0670.39 1.6270.51 1.397 0.26 1.4970.48
6.470.9 5.470.9 4.970.8 5.870.8 6.370.9
1.6770.28 1.7670.19 1.8870.42 2.2870.74 2.1070.50
4275 4376 5078 3776 5475
a
a b
Number of replicates: 5. Wet wt/d.w. ratio: 5.2070.46. Not available at site 5. Number of replicates: 5. Wet wt/d.w. ratio: 6.1270.43. Not available at site 2.
Table 4 Concentrations of metals (mg g1 d.w.) in the soft tissues of M. turbinata, P. caerulea, and M. galloprovincialis (means7SD) Sampling station
Individuals
Cd
Cr
Cu
Pb
Zn
M. turbinata 1 2 3 4 6
8 6 9 10 5
0.8670.20 1.2970.24 1.4170.29 1.0070.44 1.0470.30
0.4870.19 0.3670.09 0.2770.06 0.7270.13 0.2670.08
50.5710.4 51.4710.9 59.5714.5 66.7720.4 83.0719.7
0.6470.10 0.5270.11 0.6270.13 0.4470.10 0.6770.16
77.7711.2 98.0713.3 129.3722.9 92.5712.0 94.1714.2
P. caeruleab 1 2 3 4 5 6
9 8 21 20 10 14
2.8970.57 3.0170.63 3.6470.56 4.0670.84 3.8370.97 3.7971.14
0.9670.33 0.7870.17 0.9270.29 0.7270.18 0.9170.16 0.8270.28
10.271.4 15.975.0 13.073.2 12.873.9 19.273.6 14.773.5
1.5070.29 0.9870.20 0.8070.15 0.5170.15 0.8470.11 1.0570.27
87.4711.8 95.27 7.1 107.9717.2 100.0720.6 97.1710.8 117.1725.0
M. galloprovincialisc 1 2 3 4 5 6
18 6 22 20 12 9
0.3370.09 0.4970.10 0.3470.08 0.3270.09 0.3870.08 0.4570.08
0.8770.25 0.8070.08 0.4670.12 0.9170.28 1.1170.27 1.3170.29
9.5572.19 9.9571.28 5.5171.47 6.4571.50 11.5072.59 9.4072.05
2.2270.69 2.3870.64 2.4970.61 1.6770.41 1.9670.36 1.7370.24
123726 180738 151740 157736 155732 178734
a
a
Shell size (cm): 1.5–2.1 (height); Wet wt/d.w.: 1.7070.11. 1.8–2.4 (length); Wet wt/d.w.: 2.6670.14. c 4.1–4.8 (length); Wet wt/d.w.: 6.5670.21. b
did not indicate meaningful differences among the sites examined. Contrary to what we expected, the protected park station (station 4, Sassolini) did not exhibit meaningful differences ðP ¼ 0:05Þ in comparison to the other sites (see Table 3). In P. pavonica the observed levels of Cd and Pb were higher than those previously reported by Schlacher-Hoelinger and Schlacher (1998) and lower than our precedent work (Campanella et al., 2001). Data for Cr and Cu were higher than those
previously reported (Campanella et al., 2001; SchlacherHoelinger and Schlacher, 1998). Data for zinc were very close to our previous data on uncontaminated areas (Campanella et al., 2001) and relatively higher than those reported by other authors (Bei et al., 1990; Schlacher-Hoelinger and Schlacher, 1998). Metals concentrated in Padina decreased in the order Zn4Cu4PbXCr4Cd; in Ulva the sequence was Zn4Cu4Pb4Cr4Cd. As for molluscs, the following
ARTICLE IN PRESS M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
104
sequences were observed: Zn4Cu4Cd4PbXCr in M. turbinata; Zn4Cu4Cd4PbXCr in P. cerulea; and Zn4Cu4Pb4Cr4Cd in M. galloprovincialis. Despite a few differences, these sequences were comparable to those already identified by us and by other authors (Campanella et al., 2001; Volterra and Conti, 2000). Metal concentrations detected in algae (pooled samples, see Table 3) did not show significant differences ðP ¼ 0:05Þ on the bioaccumulation trend for the examined sites. The Cd, Cr, and Pb values detected in mussels were in the range of 0.33–0.49, 1.67–2.49, and 0.46–1.31 mg g1 d.w., respectively; while Cu and Zn levels were in the range of 5.51–11.50 and 123–180 mg g1 d.w., respectively. As for Patella, the concentration intervals detected for Cd, Cr, and Pb were 2.89–4.06, 0.72–0.96, and 0.51–1.50, respectively; for Cu and Zn they were 10.2–19.2 and 87.4–117.1 mg g1 d.w. respectively. For Monodonta, the Cd, Cr, and Pb concentration ranges were 0.86–1.41, 0.26–0.72, and 0.44–0.67 mg g1 d.w., respectively; those for Cu and Zn were 50.5–83.0 and 77.7–129.3 mg g1 d.w., respectively. To assess the existence of significant differences in the metal concentrations in molluscs, use of the two-way ANOVA was taken into account. For this purpose, normality and homogeneity of variance tests were carried out. The results led to our discarding that the hypothesis of variance homogeneity on both untransformed and log-transformed data. The one-way ANOVA was therefore performed for each species and for each metal. In the 15 tests the hypotheses of variance normality and homogeneity [tested with the Shapiro–Wilk, Lilliefors (Kolmogorov–Smirnov), and Levene tests] were successful. Thirteen of the 15 tests turned out to be significant on a level of 5%. The other two tests turned out to be significant at a level of 1% (Cd on Monodonta and Patella). Table 5 shows the ANOVA results. After assessing the significance of the averages, multiple comparison tests were carried out to asses which averages differed from the others and to accurately assess which sites accumulated metals differently from the others, with reference to the species that were examined. The multiple comparison tests performed were Bonferroni–Dunn, Tukey, Scheffe`, Sidak, Gabriel, and Hochberg.
3.1. Monodonta From the results of the six multiple comparison tests performed for each species, we observed that Cd bioaccumulation in the S. Agostino station was higher than in Vindicio. Contrary to our assumptions from the mean concentrations at each site (see Table 4), the other sites revealed similar cadmium bioaccumulation. Regarding cadmium in Monodonta, the variance homogeneity hypothesis could be accepted at a 1% level; therefore, other multiple comparison tests were performed (Tamhane, T3 Dunnet, Games-Howell, C Dunnet) that do not presuppose homoschedasticity of distribution performed. These tests also showed significant bioaccumulation differences, which were higher in Porto Romano (station 2) than in Vindicio (station 1). Cr accumulation in the Sassolini station (station 4, the Protected Sea Park) was higher than in the other sites. Cr was higher in the Vindicio site than in Serapo and S. Agostino. There were meaningful differences in Cu only between Vindicio–Serapo and P. Romano–Serapo. Cu bioaccumulation turned out to be higher in the Serapo station than in the other stations. Pb bioaccumulation was the same in the Sassolini and P. Romano stations, and in the Sassolini station it was lower than in the other sites. S. Agostino revealed a higher ðP ¼ 0:05Þ Zn than the other stations.
3.2. Patella cerulea Cd was significantly lower ðP ¼ 0:05Þ in Vindicio and P. Romano than in the other stations. According to the F test (see Table 5), the Cr level was similar in all stations. The Cu concentration was quite different among Vindicio–P. Romano, Vindicio–S.Croce, and Vindicio– Serapo, bioaccumulation being lower ðP ¼ 0:05Þ in Vindicio. On the other hand, in S. Croce Cu was higher ðP ¼ 0:05Þ than in Vindicio, S. Agostino, and Sassolini. Pb was higher in Vindicio ðP ¼ 0:05Þ than in the other stations, and in Sassolini it was significantly lower ðP ¼ 0:05Þ than in the other sites. Zn bioaccumulation turned out to be similar in P. Romano, S. Agostino, S. Croce, and Sassolini.
Table 5 Results of the one-way analysis of variance for metal concentrations Factor of variation
M. turbinata P. caerulea M. galloprovincialis
Cd
Cr
Cu
Pb
Zn
F
P
F
P
F
P
F
P
F
P
3.438 3.735 4.016
0.019 0.004 0.001
21.806 1.801 19.848
0 0.123 0
4.029 7.760 21.919
0.009 0 0
5.851 35.608 5.657
0.001 0 0
12.990 2.744 4.236
0 0.025 0.001
ARTICLE IN PRESS M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
3.3. Mussels Cd was higher in the P. Romano site than in Vindicio, S. Agostino, and Sassolini. For Monodonta, the variance homogeneity hypothesis could be accepted at a 1% level. Therefore, other multiple comparison tests were performed (see above). This additional study also showed significant differences between Cd accumulation for Serapo–Vindicio and Serapo–S.Agostino, bioaccumulation being higher in Serapo. Cr was significantly lower in S. Agostino than in the other stations, while Serapo had the highest metal levels. In Sassolini, the Cu was significantly lower than in almost all other sites. Pb was significantly lower in Sassolini–Vindicio than in Sassolini–S.Agostino. Zn was significantly lower in the Vindicio station than in the P. Romano, Sassolini and Serapo stations. Thus, the tests we performed showed that the trend of metal bioaccumulation in molluscs was not clearly univocal; that is to say, no one station showed an accumulation higher than that of any other, at least for the metals we studied. To obtain homogeneous groups for metal bioaccumulation, we performed cluster analyses in our sites. The cluster analysis showed two homogeneous groups for Monodonta (see Fig. 2). The first comprised P. Romano, S. Agostino, and Serapo, and the second comprised Vindicio and Sassolini. The analysis clearly showed that the former group presented metal bioaccumulation higher than that of the latter.
105
On the other hand, the cluster analysis with Patella revealed the presence of three groups. The first included Vindicio, P. Romano, Sassolini, and Serapo, the second comprised the S. Agostino site, and the third comprised the Santa Croce site (Fig. 3). The latter also turned out to be the most dissimilar, with generally higher metal bioaccumulation levels. The cluster analysis revealed three groups of mussels (Fig. 4). The first included Vindicio, S. Croce, P. Romano, and Serapo, the second the Sassolini station, and the third the S. Agostino station. The second and third groups evidenced lower contamination levels. Cluster analysis results (Figs. 2–4) gave further evidence that the Sea Park Station (Sassolini) was not noticeably different from the others, as opposed to what we expected. The hypothesis that Protected Sea Park would be cleaner than the other sites must therefore be reconsidered. The data for the common species were compared, through cluster analysis, with the data we derived from uncontaminated sites in the Sicilian Sea (Favignana Island) (Campanella et al., 2001). This research confirmed the diversity between these two marine ecosystems and allowed us two identify two different groups: the stations in the Gulf of Gaeta and the Sicilian stations. The two species we considered, Monodonta and Patella, revealed that the Gulf of Gaeta had higher metal contamination than the uncontaminated sites in the Sicilian Sea (Figs. 5 and 6).
Fig. 2. Classification of the sampling stations (cluster analysis) for Monodonta turbinata.
Fig. 3. Classification of the sampling stations (cluster analysis) for Patella cerulea.
ARTICLE IN PRESS 106
M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
Fig. 4. Classification of the sampling stations (cluster analysis) for Mytilus galloprovincialis.
Fig. 5. Classification of the sampling stations (cluster analysis) for Monodonta turbinata of Gaeta Gulf and Favignana island (Sicily). Punta Sottile, Preveto, and Cala Rossa are the clean sites of Favignana.
Fig. 6. Classification of the sampling stations (cluster analysis) for Patella cerulea of Gaeta Gulf and Favignana island (Sicily). Preveto, Cala Rossa, Cala Azzurra, and Punta Sottile are the clean sites of Favignana.
In order to assess the concentration factors (CFs) for observation of the bioaccumulation ability of every single species, analyses of the waters were performed on both metals in solution and total metals. The CF may be used to evaluate the state of conservation of an ecosystem or to monitor its state (Conti and Cecchetti, 2001). Table 6 shows the results of the water analysis and Table 7 shows the mean metal concentrations detected in both biota and seawater. The data on seawater metals must be interpreted with some caution, due to the high variability of metal concentrations in
seawater, which depends on several factors (physicochemical, environmental, etc.). The determination of trace metals in marine waters is a critical part of biomonitoring studies because of its many important analytical aspects. The levels of trace metals are often low in coastal waters. It is therefore important to prevent the use of inadequate sampling and sample treatment techniques. [For instance, sample contamination may arise from components of the sampling gear (Achterberg, 2000)].
ARTICLE IN PRESS M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
107
Table 6 Total and dissolved metal concentrations in coastal seawater samples (means7SD) Sampling station
Cd (ng L1)
Cr (ng L1)
Cu (mg L1)
Pb (mg L1)
Zn (mg L1)
Total metals 1 2 3 4 5 6
30175 25277 29178 24776 23878 21477
39172 42173 38877 51171 509710 43072
2.1270.07 3.0570.03 2.6370.04 2.5170.08 3.2170.12 2.8170.09
2.2170.02 3.1070.06 2.0070.11 3.1570.03 2.2370.09 2.3670.06
10.4070.12 9.6370.35 7.5070.22 10.1570.18 9.6470.38 8.9170.47
Dissolved metals 1 2 3 4 5 6
13972 11576 13275 12071 14274 13373
27974 341711 27772 33879 37275 29371
1.3470.02 0.8270.01 1.4370.03 1.6770.03 1.3270.01 0.7970.03
1.6570.05 2.1270.03 1.3270.03 1.0570.02 2.8370.01 1.1170.02
5.9670.16 6.3770.21 4.7570.11 5.0970.10 7.2170.24 5.1570.09
Number of replicates: 5. Mean values of salinity and temperature during sampling: S ¼ 3671; t ¼ 18711C: Table 7 Mean metal concentrations in biota and seawater from stations 1 to 6 (means7SD) Matrix
(mg g1 d.w.) Cd
Cr
Cu
Pb
Zn
P. pavonica U. lactuca M. turbinata P. caerulea M. galloprovincialis
0.5070.07 0.1870.06 1.1270.48 3.5470.87 0.3870.11
3.4570.70 1.6370.60 0.4270.15 0.8570.24 0.9070.21
12.371.8 5.871.1 61.1720.1 14.374.9 8.3871.62
3.9870.67 1.9470.38 0.5870.16 0.9570.19 2.0770.49
51711 45712 98723 100.8718 152737
Total metals (mg L1)a Dissolved metals (mg L1)
0.2670.06 0.1370.04
0.4470.03 0.3270.05
2.7270.18 1.2370.12
2.5070.11 1.6870.15
a
9.371.58 5.772.11
Soluble+particulate.
Table 8 CFsa 103 calculated with reference to soluble and total metal concentrations in seawater Matrix
Cd
Cr
Cu
Pb
Zn
Soluble P. pavonica U. lactuca M. turbinata P. caerulea M. galloprovincialis
3.99 1.43 8.93 28.22 3.03
11.17 5.28 1.36 2.75 2.91
10.36 4.88 52.41 12.05 7.06
2.45 1.19 0.36 0.59 1.27
9.27 8.18 17.82 18.33 27.64
Total P. pavonica U. lactuca M. turbinata P. caerulea M. galloprovincialis
1.99 0.72 4.46 14.11 1.51
8.13 3.84 0.98 2.00 2.12
4.69 2.21 23.70 5.44 3.19
1.65 0.80 0.24 0.39 0.85
5.68 5.01 10.92 11.23 16.94
a
CF ¼ CO =CSW ; where CO =mean concentration in the organism (mg g1 d.w.) and CSW =mean concentration in seawater (mg L1).
4. Discussion Our results support the hypothesis that the coastal area of the Gulf of Gaeta has significant basal contamination levels, which, however, do not reach those of clearly contaminated areas. In any case, the
expression ‘‘background levels’’ must be used with some caution, as its meaning differs according to the local geochemical or hydrodynamic conditions of the body of water under examination (Conti, 2002). Our work shows the importance of using several possible biomonitors when studying the quality of ecosystems, such as algae,
108
Table 9 Selected references of metal concentrations (mg/g d.w.) of molluscs from different geographical areas (mean values and ranges) Species
Sites
Cd
M. galloprovincialis
Four sites, Gulf of Trieste (NE Italy)
41 sites along Italian coastal areas Bakar Bay, (ex-Yugoslavia)
Venice Lagoon, Gulf of Trieste (NE Italy)
Grado e Marano Lagoon Split, Kastela Bay (Croazia)
Pb
Zn
(1) 0.21a
1.26a
0.84a
16.5a
(2) 0.29a (3) 0.29a (4) 0.23a 0.4–6 Site 1
1.41a 1.48a 1.41a 2.4–15.5 Site 1
3.29a 1.85a 0.64a 2.7–117 Site 1
27.4a 25.4a 14.6a 97–644
0.325a 0.120–0.380 Site 2 0.192a 0.130–0.270 Site 3 0.066a 0.018–0.160
1.229a 0.830–2.230
0.137a
1.57a
0.925a 0.480–1.790 Site 2 1.660a 1.100–2.200 Site 3 0.405a 0.160–0.670 Site 4 0.841a 0.400–1.300 0.19a
20.8a
Site 1
Site 1
Site 1
Site 1
0.12–0.38a Site 2 0.18–0.42a 0.062–0.48a
0.77–2.23a Site 2 0.50–3.26a
0.48–1.79a Site 2 1.21–5.74a 0.15–11.8a
6.90–29.60a Site 2 13.70–42.20a
0.5–29
0.16a
0.44a
0.31a
1.41a
0.32a
0.13–0.78 0.23a 0.17–0.35 0.22a
0.18–6.52 1.28a 0.38–3.64 0.94a
0.12–1.09 1.17a 0.58–3.66 0.17a
0.13–0.35
0.11–0.26 0.735a
0.273a
12.23a
0.009–2.53 5.08–20.89
0.014–1.18 6.18–80.26
1.48–37.32 82–185
Venice Lagoon, 44 sites
0.134a
0.28–2.78 0.32–0.75a 0.265a
Venice Lagoon, six sites
n.d–1.64 0.05–4.54
n.d–1.20 0.37–20.38
Notes
References Majori et al. (1978)
Eisler (1981) Gabrielli Favretto and Favretto (1984)
Pool of five individuals
Martincˇic´ et al. (1984) Favretto et al. (1987)
Pools of 40–50 individuals Five pools of three individuals
Giordano et al. (1989, 1991) Ozretic et al. (1990)
36 stations
Majori et al. (1991)
22 stations
Majori et al. (1991)
36 stations
Majori et al. (1991)
Six stations 10 individuals for each site
Vukadin and Odzak (1991) Zatta et al. (1992)
Pools of 10–15 individuals
Widdows et al. (1997)
ARTICLE IN PRESS
Nine stations, Lim Fjord (N Adriatic Sea) Two sites, Muggia Bay, Trieste (NE Italy)
Cu
M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
NW Mediterranean Sea Four sites, Gulf of Trieste (NE Italy)
Cr
Chile, four sites
Bakar Bay (ex-Yugoslavia)
4.13 2.58–5.99 0.38 0.33–0.49 700a
Bakar Bay (ex-Yugoslavia)
417–1140 0.50a
This work P. caerulea
Site 2: marine reseve of S. Maria di Castellabate
Monodonta turbinata
Bakar Bay, (ex-Yugoslavia)
0.70a
0.30
1.7
Monodonta articulata
Larymna (Grecia) (contaminated area)
Monodonta mutabilis
Larymna (Grecia) (contaminated area)
0.20
Site 1
Site 1
0.62a 0.39–0.91 Site 2
1.42a 0.92–2.31 Site 2
0.40a 0.27–0.53 3.54 2.89–4.06 0.20a
1.62a 0.91–4.04 0.95 0.51–1.50 0.61a
0.85 0.72–0.96
14.3 10.2–19.2
0.19–0.22 Larymna (Grecia) (contaminated area) Favignana Island, Sicily (Italy), (uncontaminated area) This work
152 123–180
This work Pools 40 individuals
Taramelli et al. (1991)
Pools of 3–4 individuals
Ozretic et al. (1990)
0.61–0.79 5
Campanella et al. (2001)
Camoni et al. (1980)
100.8 87.4–117.1
This work Pools of 9–12 individuals
Ozretic et al. (1990)
Pools of 10 individuals
Nicolaidou (1994)
0.59–0.63 5.6
109.1
65.7
1.44
0.31
10.9
0.24
31
Campanella et al. (2001)
1.12 0.86–1.41
0.42 0.26–0.72 11.2
62.2 50.5–83.0 125.4
0.58 0.44–0.67
98 77.7–129.3 102.0
This work
4.6
136.6
98.8
Pools of 10 individuals
Nicolaidou (1994)
Pools of 10 individuals
Nicolaidou (1994)
N.d.=not detectable. a Wet weight.
ARTICLE IN PRESS
This work
0.48–0.52 4.41
8.38 5.51–11.50
Bruhn et al. (1999)
M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
Favignana Island, Sicily (Italy), uncontaminated area Tyrrhenian Sea, Naples (Italy) Site 1: Naples Bay
0.90 0.46–1.31
0.32 0.18–0.40 2.07 1.67–2.49
109
ARTICLE IN PRESS 110
M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
bivalve molluscs (filtering organisms), and herbivores (gastropod molluscs). Each biomonitor responds to a particular metal fraction of the water body: mussels to the fraction present in particulates, algae to metals in solution. Moreover, metal concentrations in algae could affect the concentrations of metals in gastropods (the phenomenon of biomagnification) (Conti, 2002). The data available on metal concentrations in the water of this coastal area are quite scarce. The Cd and Zn levels detected in the present study (Table 6) are close to those we had found previously in the Sea of Sicily (Campanella et al., 2001); on the other hand, the Cr, Cu, and Pb levels are 1.5–2.0 times higher. Generally, the data on waters (Table 6) are comparable with those concerning the coastal areas of the Tyrrhenian Sea (Alpha et al., 1982). Table 8 reports the calculation of CFs for the species examined. As mentioned above, the data in Table 8 must be viewed cautiously, as CFs could be influenced by several factors. The passage of contaminant through the trophic chain can affect CF values. The data shown in Table 8 confirm the high aptitude of the species we examined as bioaccumulators. It is actually possible to observe that very high CFs can be calculated with reference to soluble metal concentrations. Regarding net accumulation, we can see that Patella caerulea was the strongest accumulator of Cd, P. pavonica that of Cr and Pb, M. turbinata that of Cu. These results back up our previously reported findings for these species for the Sea of Sicily (Campanella et al., 2001). M. galloprovincialis turned out to be the highest Zn accumulator. Some authors have reported a reasonable ability of molluscs for regulating Cu and Zn (Amiard-Triquet et al., 1986). As is well known, only a few marine organisms are able to regulate metal concentration levels in tissues. This ability, however, exists for only a limited range of environmental concentrations for some essential elements such as Cu and Zn (Phillips, 1995). The total metal contents in molluscs found in the present work are comparable with those found in the literature (Table 9). In particular, the levels detected in M. galloprovincialis rank generally quite low compared to some highly contaminated sites (Venice Lagoon). For P. caerulea and M. turbinata, it is clear that the mean metal concentrations of Cr, Cu, Pb, and Zn are higher than those in uncontaminated sites (Favignana Island, Sicily; Table 9), while the average Cd levels in the Gulf of Gaeta have turned out to be similar to those in the Sicilian Sea. When trace metal levels from different areas are compared, one must be quite wary, as data are often reported in fresh weight (Table 9). While the content of water does not change too much for fish, it is very variable for the molluscs, being affected by such factors as habitat, vital condition, and pretreatment and sample conservation procedures after sampling. For instance,
Crisetig et al. (1984) observed in M. galloprovincialis variable fresh weight/d.w. ratios in the range of 4.32– 8.50 for mussels as long as 4–5 cm; Martincˇic´ et al. (1992) found variable ratios in the interval of 4.27–12.99 in mussels transplanted into an estuary environment (Krka River, eastern Adriatic coast). We think that data concerning d.w. are susceptible to less variability than those of fresh weight.
5. Conclusions The use of bioindicators turned out to be very valuable for the study of a coastal area with fairly significant basal contamination levels. The green algae U. lactuca, the brown algae P. pavonica, the bivalve mollusc M. galloprovincialis, and the two gastropod molluscs M. turbinata and P. cerulea possess high potential as cosmopolitan biomonitors for trace metals in the Mediterranean. These species have the necessary prerequisites for use as bioindicators: they are easy to identify and to sample, are available all year round, and are present in almost all coastal areas of the Mediterranean Sea. Samples of the above-mentioned species were collected at six stations along the Gulf of Gaeta (Tyrrhenian Sea, central Italy) The picture of bioavailable Cd, Cr, Cu, Pb, and Zn was not homogeneous. Statistical analyses (one-way ANOVA, multiple comparison tests, cluster analysis) showed that no one site was more contaminated than the others. Therefore, the supposition of the protected Marine Park (station 4) as a control, ‘‘clean’’ monitoring site had to be rejected; this is a useful indication for to further elaboration and betterment of the environmental protection policies for marine sites. The cluster analysis for molluscs also showed the presence of homogeneous groups within the examined sites and a clear differentiation as opposed to the impact levels of unpolluted sites in other coastal areas of Southern Italy. The use of molluscs in biomonitoring studies demands great accuracy, especially with regard to the dimensions and weight of individuals. This aspect is something of a critical issue. It is necessary to pick individuals belonging strictly to very narrow weight and size ranges. If this were not possible, other methods of statistical analysis could be used [for instance, covariance (ANCOVA), which can examine samples differing in weight and size] (Conti, 2002; Cubadda et al., 2001). The species examined showed a great ability to accumulate concentrations of metals several thousands times those detected in marine water (soluble fraction). It is worthwhile to remark that P. caerulea and M. galloprovincialis are commonly consumed seafood in many Mediterranean countries. The passage of trace
ARTICLE IN PRESS M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
metals from the abiotic section to the tissues of these species is highly interesting for food safety. Although the species we studied in the present work and suggested as biomonitors present numerous advantages, more information and more studies are necessary to clarify accumulation patterns. It is important, as well, not to overlook the possible existence of regulating and/ or competition mechanisms of metals inside tissues. Also, the most recent developments in the field of analytical techniques can significantly contribute to our knowledge in this field (Conti et al., 2002). It is quite important to take into account the knowledge derived from the study of metal speciation in seawater, along with traditional biomonitoring investigations, for use in environmental protection policies of the future.
Acknowledgments The authors thank Prof. Francesco Battaglia and Dr. Domenico Cucina for they valuable support with the statistical analysis. We also thank Dr. Luca Ciprotti, Dr. Carlo Sucapane, and Fabio Spadoni for their valuable assistance. Grateful thanks are due to Maria Brighi, Petra Tomassini, and Pierluigi Tomassini, for their support during our stay in Formia.
References Achterberg, E.P., 2000. Automated techniques for real-time shipboard determination of dissolved trace metals in marine surface waters. In: Conti, M.E., Botre`, F. (Eds.), The Control of Marine Pollution: Current Status and Future Trends; Int. J. Environ. Pollut. 13, 249–261, Inderscience Enterprises Ltd., Milton Keynes, UK. Alpha, M., Caristi, C., Cimino, G., Ziino, M., 1982. Heavy metals in Ionian and Tyrrhenian waters from a section of the Sicilian coast. Mar. Pollut. Bull. 13 (11), 399–400. Amiard-Triquet, C., Berthet, B., Metayer, C., Amiard, J.C., 1986. Contribution to the ecotoxicological study of cadmium, copper and zinc in the mussel Mytilus edulis. II. Experimantal study. Mar. Biol. 92, 7–13. Bei, F., Papathanassiou, E., Catsiki, V.A., 1990. Heavy metals concentrations in selected marine species from Milos Island (Aegean Sea). Rapp. P.-V. Re´un. CIESM 32 (1), 126. Bruhn, C.G., San Francisco, N.A., Neira, J.Y., 1999. Determination of cadmium and lead in mussels by tungsten coil electrothermal atomic absorption spectrometry. Talanta 50, 967–975. Bruland, K.W., Franks, R.P., Knauer, G.A., Martin, J.H., 1979. Sampling and analytical methods for the determination of copper, cadmium, zinc, and nickel at the nanogram per liter level in sea water. Anal. Chim. Acta 105, 233–245. Bryan, G.W., Langston, W.J., 1992. Bioavailability, accumulation and effects of heavy metals in sediments with special reference to United Kingdom estuaries: a review. Environ. Pollut. 76, 89–131. Camoni, I., Cotta-Ramusino, F., Funari, E., Giordano, R., Ovidi, P., Dohrn, P., Ballio, A., 1980. Heavy metals and organochlorine residues in marine animals and coastal sediments of Tyrrhenian Sea around Naples. Ann. Ist. Super. Sanita´ 16, 645–656.
111
Campanella, L., Conti, M.E., Cubadda, F., Sucapane, C, 2001. Trace metals in seagrass, algae and molluscs from an uncontaminated area in the Mediterranean. Environ. Pollut. 111, 117–126. Conti, M.E., 2002. SEAM Ed., Rome, Il monitoraggio biologico della qualita` ambientale. Rome, 181pp. Conti, M.E., Cecchetti, G., 2001. Biological monitoring: lichens as bioindicators of air pollution assessment—a review. Environ. Pollut. 114, 471–492. Conti, M.E., Tudino, M.B., Muse, J.O., Cecchetti, G., 2002. Biomonitoring of heavy metals and their species in the marine environment: the contribution of atomic absorption spectroscopy and inductively coupled plasma spectroscopy. Res. Trends Appl. Spectrosc., in press. Crisetig, G., Cattani, O., Viviani, R., 1984. Metalli tossici nei mitili dell’alto e medio adriatico. Arch. Vet. Ital. 35, 10–16. Cubadda, F., Conti, M.E., Campanella, L, 2001. Size-dependent concentrations of trace metals in four mediterranean gastropods. Chemosphere 45, 561–569. Eisler, R., 1981. Trace Metal Concentrations in Marine Organisms. Pergamon Press, Oxford, 685pp. Favero, N., Cattalini, F., Bertaggia, D., Albergoni, V., 1996. Metal accumulation in a biological indicator (Ulva rigida) from the Lagoon of Venice (Italy). Arch. Environ. Contam. Toxicol. 31, 9–18. Favretto, L., Gabrielli Favretto, L., Felician, L., 1987. Principal component analysis for the identification of pollution sources in mussel survey by trace metals. Lebensm. Unters. Forsch. 184, 101–109. Fo¨ster, P., 1976. Concentrations and concentration factors of heavy metals in brown algae. Environ. Pollut. 10, 45–53. Gabrielli Favretto, L., Favretto, L., 1984. Heavy metals at trace level in edible mussels (Mytilus galloprovincialis Lamarck) from the Gulf of Trieste. Lebensm. Unters. Forsch. 179, 197–200. Giordano, R., Arata, P., Ciaralli, L., Rinaldi, S., Giani, M., Cicero, A.M., Costantini, S., 1991. Heavy metals in mussels and fish from Italian coastal waters. Mar. Pollut. Bull. 22, 10–14. Giordano, R., Arata, P., Rinaldi, S., Ciaralli, L., Giani, M., Rubbiani, M., Costantini, S., 1989. Mercury, cadmium and lead levels in marine organisms (Mytilus galloprovincialis LMK) collected along the Italian coasts. Ann. Ist. Super. Sanita´ 25, 511–516. Haritonidis, S., Malea, P., 1995. Seasonal and local variation of Cr, Ni and Co concentrations in Ulva rigida C. Agardh and Enteromorpha linza (Linnaeus) from Thermaikos Gulf, Greece. Environ. Pollut. 89, 319–327. Haritonidis, S., Malea, P., 1999. Bioaccumulation of metals by the green alga in Ulva rigida from Thermaikos Gulf, Greece. Environ. Pollut. 104, 365–372. IRSA, 1984. Metodi di analisi per acque di mare: Cadmio (225.2), Rame (260.1). Quad. Ist. Ric. Acque 59. Luoma, S.N., 1983. Bioavailability of trace metals to aquatic organisms—a review. Sci. Total Environ. 28, 1–22. Majori, L., Nedoclan, G., Daris, F., Modonutti, G.B., 1991. Metal distribution (Cd, Cr, Cu, Fe, Mn, Pb, Zn) in Mytilus galloprovincialis LMK in coast and lagoon areas in northern Adriatic. Rev. Int. Oceanogr. Med. 101–104, 225–228. Majori, L., Nedoclan, G., Modonutti, G.B., Daris, F., 1978. Study of the seasonal variations of some trace elements in the tissues of Mytilus galloprovincialis taken in the Gulf of Trieste. Rev. Int. Oce´anogr. Med. XLIX, 37–40. Martincˇic´, D., Kwokal, Z., Peharec, Z., Margus, D., Branica, M., 1992. Distribution of Zn, Pb, Cd and Cu between seawater and transplanted mussels (Mytilus galloprovincialis). Sci. Total Environ. 119, 211–230. Martincˇic´, D., Nurnberg, H.W., Stoeppler, M., Branica, M., 1984. Bioaccumulation of heavy metals by bivalves from Lim Fjord (North Adriatic Sea). Mar. Biol. 81, 177–188.
ARTICLE IN PRESS 112
M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
Muse, J.O., Stripeikis, J.D., Ferna`ndez, F.M., d’Huicque, L., Tudino, M.B., Carducci, C.N., Troccoli, O.E., 1999. Seaweeds in the assessment of heavy metal pollution in the Gulf San Jorge, Argentina. Environ. Pollut. 104, 315–322. Nicolaidou, A., 1994. Identification of biological indicators of heavy metals at the site of a smelting factory. UNEP/FAO, Final Reports on Research Projects Dealing with the Effect of Pollutants on Marine Organisms and Communities. MAP Technical Reports Series No. 80, Athens, pp. 25–38. Ozretic, B., Krajnovic-Ozretic, M., Santin, J., Medjugorac, B., Kras, M., 1990. As, Cd, Pb, Hg in benthic animals from the Kvarner– Rijeka region, Yugoslavia. Mar. Pollut. Bull. 21, 595–598. Phillips, D.J.H., 1977. The use of biological indicator organisms to monitor trace metal pollution in marine and estuarine environment—a review. Environ. Pollut. 13, 281–317. Phillips, D.J.H., 1995. The chemistries and environmental fates of trace metals and organochlorines in aquatic ecosystems. Mar. Pollut. Bull. 31, 193–200. Rai, L.C., Gaur, J.P., Kumar, H.D., 1981. Phycology and heavy-metal pollution. Biol. Rev. 56, 99–151. Rainbow, P.S., 1995. Biomonitoring of heavy metal availability in the marine environment. Mar. Pollut. Bull. 31 (4–12), 183–192. Schlacher-Hoenlinger, M.A., Schlacher, T.A., 1998. Differential accumulation patterns of heavy metals among the dominant macrophytes of a Mediterranean seagrass meadow. Chemosphere 37, 1511–1519. Seeliger, U., Edwards, P., 1977. Correlation coefficients and concentration factors of copper and lead in seawater and benthic algae. Mar. Pollut. Bull. 8 (1), 16–19. Sericano, J.L., 2000. The Mussel Watch approach and its applicability to global chemical contamination monitoring programmes. In: Conti, M.E., Botre`, F. (Eds.), The Control of Marine Pollution:
Current Status and Future Trends; Int. J. Environ. Pollut. 13 (1–6), 340–350, Inderscience Enterprises Ltd., Milton Keynes, UK. Struck, B.D., Pelzer, R., Ostapczuk, P., Emons, H., Mohl, C., 1997. Statistical evaluation of ecosystem properties influencing the uptake of As, Cd, Co, Cu, Hg, Mn, Ni, Pb and Zn in seaweed (Fucus vesiculosus) and common mussel (Mytilus edulis). Sci. Total Environ. 207, 29–42. Taramelli, E., Costantini, S., Giordano, R., Olivieri, N., Perdicaro, R. (1991). Cadmium in water, sediments and benthic organisms from a stretch of coast facing the thermoelectric power plant at Torvaldiga (Civitavecchia, Rome). UNEP/FAO, Final Reports on Research Projects Dealing with Bioaccumulation and Toxicity of Chemical Pollutants. MAP Technical Reports Series No. 52, Athens, pp. 15–31. Volterra, L., Conti, M.E., 2000. Algae as biomarkers, bioaccumulators and toxin producers. In: Conti, M.E., Botre`, F. (Eds.), The Control of Marine Pollution: Current Status and Future Trends; Int. J. Environ. Pollut. 13 (1–6), 92–125, Inderscience Enterprises Ltd., Milton Keynes, UK. Vukadin, I., Odzak, N., 1991. Fate and distribution of chromium in waters, sediments and mussels of the Kastela Bay. In: Gabrielides, G.P. (Ed.), Proceedings of the UNEP/FAO/IAEA Consultation Meeting on the Accumulation and Transformation of Chemical Contaminants by Biotic and Abiotic Processes in the Marine Environment, La Spezia, Italy, 24–28 September 1990. MAP Technical Reports Series No. 59, Athens, pp. 363–369. Widdows, J., Nasci, C., Fossato, V.U., 1997. Effects of pollution on the scope for growth of mussels (Mytilus galloprovincialis) from the Venice lagoon, Italy. Mar. Environ. Res. 43 (1/2), 69–79. Zatta, P., Gobbo, S., Rocco, P., Perazzolo, M., Favarato, M., 1992. Evaluation of heavy metal pollution in the Venetian lagoon by using Mytilus galloprovincialis as biological indicator. Sci. Total Environ. 119, 29–41.
Environmental Research 93 (2003) 99–112
A biomonitoring study: trace metals in algae and molluscs from Tyrrhenian coastal areas$ Marcelo Enrique Contia, and Gaetano Cecchettib a
Dipartimento di Controllo e Gestione delle Merci e del loro Impatto sull’Ambiente, Universita` di Roma ‘‘La Sapienza’’, Via del Castro Laurenziano 9, 00161, Rome, Italy b Centro per le Valutazioni Ambientali delle Attivita` Industriali, Facolta` di Scienze Ambientali, Universita` degli Studi di Urbino, Campus Scientifico Sogesta, 61029 Urbino (Pu), Italy Received 30 July 2002; received in revised form 16 December 2002; accepted 2 January 2003
Abstract Marine organisms were evaluated as possible biomonitors of heavy metal contamination in marine coastal areas. Concentrations of Cd, Cr, Cu, Pb, and Zn were measured in the green algae Ulva lactuca L., the brown algae Padina pavonica (L.) Thivy, the bivalve mollusc Mytilus galloprovincialis Lamarck, and the two gastropod molluscs Monodonta turbinata Born and Patella cerulea L. collected at six coastal stations in the area of the Gulf of Gaeta (Tyrrhenian Sea, central Italy). The coastal area of the Regional Park of Gianola and Monte di Scauri (a ‘‘Protected Sea Park’’ area) was chosen as a control site. Seawater samples were also collected in each site to assess soluble and total metal concentrations and to gain additional information on both the environmental conditions of the area and possible bioaccumulation patterns. Metal concentrations detected in algae and molluscs did not show significant differences among all stations studied. Moreover, statistical analyses (ANOVA, multiple comparison tests, cluster analysis) showed that the Sea Park station was not significantly different from the others. The hypothesis that the Protected Sea Park would be cleaner than the others must therefore be reconsidered. Data from this study were also compared with those previously obtained from uncontaminated sites in the Sicilian Sea, Italy. The results show clearly differences between these two marine ecosystems. The species examined showed great accumulations of metals, with concentration factors (CFs) higher than 10,000 with respect to the concentrations (soluble fractions) in marine waters. Metal concentrations recorded in this area may be used for background levels for intraspecific comparison within the Tyrrhenian area, a body of water about which information is still very scarce. r 2003 Elsevier Science (USA). All rights reserved. Keywords: Biomonitoring; Trace metals; Marine ecosystems; Tyrrhenian Sea
1. Introduction The use of marine organisms as bioindicators for trace metal pollution is very common these days. Algae and molluscs are among the organisms most used for this purpose (Rainbow, 1995). Macroalgae are able to accumulate trace metals, reaching concentration values that are thousands of times higher than the corresponding concentrations in sea water (Bryan and Langston, 1992; Fo¨ster, 1976; Rai et al., 1981). Algae bind only free metal ions, the concentrations of which depend on $
This work was supported by a research grant from the Ministry of Instruction, University, and Scientific Research (MURST), year 2000– 2001. Corresponding author. Fax: +39-06-445-1566. E-mail address: [email protected] (M.E. Conti).
the nature of suspended particulate matter (Luoma, 1983; Seeliger and Edwards, 1977; Volterra and Conti, 2000), which, in turn, is formed by both organic and inorganic complexes. Moreover, algae satisfy all the basic requirements of bioindicators: they are sedentary, their dimensions are suitable, they are easy to identify and to collect, they are widely distributed, and they accumulate metals to a satisfactory degree (Conti, 2002). The most used algae in biomonitoring studies in the Mediterranean Sea are those of the genus Ulva. Ulva rigida C. Ag. is the species of Ulva most widespread along the Mediterranean coasts (Haritonidis and Malea, 1995, 1999). The species Ulva lactuca L. is especially present in polluted locations near the surface, to a depth of 4–5 m (Favero et al., 1996; Muse et al., 1999; Volterra and Conti, 2000). Despite its distribution along the Mediterranean coasts, the brown algae Padina pavonica
0013-9351/03/$ - see front matter r 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S0013-9351(03)00012-4
ARTICLE IN PRESS 100
M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
has not been used frequently in biomonitoring studies; its suitability as a bioindicator has been proposed only recently (Campanella et al., 2001). Bivalve molluscs, along with the Ulva genus, play an important role as bioindicators for trace metal pollution and appear more and more often in global monitoring programs (Rainbow, 1995; Sericano, 2000). Mussels can accumulate and integrate concentrations of several metals in seawater for relatively long intervals. They also assimilate trace metals from their food and from the ingestion of inorganic particulate material (Phillips, 1977; Struck et al., 1997). In recent years, researchers have focused their attention on the identification of other possible bioindicators for trace metal pollution, such as the gastropod molluscs Monodonta turbinata and Patella cerulea. This is because it is necessary to identify a wider range of bioindicators and thus expand current understanding of different bioaccumulation strategies for trace metals. The objective of this work was to gather more information on the use of five species selected as cosmopolitan biomonitors for the Tyrrhenian area: the green algae U. lactuca L., the brown algae P. pavonica (L.) Thivy, the mussel Mytilus galloprovincialis Lam., and the two gastropod molluscs M. turbinata Born and P. cerulea L. For this purpose we assessed the possible effect of anthropogenic activities on a coastal area of the Gulf of Gaeta, Tyrrhenian Sea (central Italy). This area, for which the literature is quite scant, is actually not completely free from industrial activities and therefore is not completely uncontaminated. Moreover, it is affected by the presence of two fairly big towns: Formia and its harbor (37,500 inhabitants) and the town of Gaeta (24,000 inhabitants). The coastal area of the Regional Park of Gianola and Monte di Scauri was also sampled as a possible monitoring site. The park, with an area of 285 ha was established in 1987. Furthermore, the stretch of coast just off the shoreline of the Regional Park is protected by the establishment of an ‘‘Oasi Blu’’ (Blue Oasis, i.e., a protected sea area) through a public concession to WWF Italy. In this area, covering 50,000 m2, motor boats and fishing are not allowed. Generally, when studying sea pollution, some clearly contaminated stations are singled out and compared with a ‘‘clean’’ monitoring site. In this study we examined supposedly similar sites in order to identify useful ‘‘background levels’’ as a reference for intraspecific comparisons within the Tyrrhenian area, on which information has been very scarce. For a broader picture of both the environmental conditions of the area under investigation and possible bioaccumulation patterns, seawater samples were collected to assess soluble and total metal concentrations. Although occasional analysis of seawater samples cannot represent reliable measurements of the mean
contaminant concentrations in each site (Phillips, 1977), indicative information can be drawn, provided that the data taken from all sampling stations are averaged.
2. Materials and methods 2.1. Sampling and sample pretreatment Samples were collected in May 2000, with a total of six stations distributed in the coastal area of the Gulf of Gaeta, central Italy. The sampling stations were chosen according to their exposure to sea currents (Fig. 1). According to criteria of homogeneity (Conti, 2002; Conti et al., 2002), at each sampling station there were at least two sampling sites. The sampling was performed homogeneously; that is, the selected individuals belonged to the same life cycle, had similar dimensions (and weights, for molluscs), and were always collected at the same depth and distance from the shoreline. Of the six sampling stations, one (protected sea area, WWF Oasis) did not appear to be affected greatly by anthropic activities. Samples of U. lactuca and P. pavonica were handpicked in the subtidal zone at a depth of about 2–3 m. Care was taken to choose thalli all at a similar stage of development. At each station, 10 samples of algae were sampled, and those abundantly covered with epiphyta were rejected. The samples were washed in seawater at the sampling site and transferred to the laboratory in precleaned polyethylene bags under refrigerated conditions. Upon arrival at the laboratory, they were thoroughly cleaned and any epiphyta and sediments were carefully removed with nylon brushes under tap water for a few seconds. Algal material was rapidly rinsed in deionized water (Milli-Q, Millipore Corp., Bedford, MA, USA) to minimize any possible metal loss during the procedure and then pulverized. This was then frozen (201C) until analysis. Algal samples for each station were pooled, and five subsamples of about 0.7 g [dry wt (d.w.)] were digested in a microwave oven (MDS 81D CEM, Cologno al Serio, Italy) after 12 h of cold premineralization with ultrapure concentrated HNO3 Merck Suprapur (Darmstadt, Germany). Individuals of M. galloprovincialis, P. cerulea, and M. turbinata were collected in the tidal zone from each station according to their availability. Then, for a 24-h period, molluscs were placed in filtered seawater collected at the corresponding station to allow the depuration of the particulate matter residues present in the mantle cavity and digestive tract. The soft parts of the molluscs, carefully extracted with a plastic spatula, were individually mineralized with a mixture of concentrated HNO3/H2O2 (5+2 mL). All chemicals used in sample treatments were of ultrapure grade (HNO3, H2O2 30%, Merck Suprapur).
ARTICLE IN PRESS M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
101
Fig. 1. The study area (Gaeta Gulf, Tyrrhenian Sea, Italy) and the sampling stations: (1) Formia Vindicio beach; (2) Porto Romano (Roman harbour); (3) Sant’Agostino beach; (4) Sassolini beach (Protected Sea Park); (5) Santa Croce river; (6) Serapo beach.
Ultrapure water (Milli-Q System, Millipore) was used for all solutions. All glassware was cleaned prior to use by soaking in 10% v/v HNO3 for 24 h and rinsed with Milli-Q water. The standard solutions of metals were prepared from stock standard solutions of ultrapure grade supplied by Merck. The calculation of d.w. on the examined species (10 replicates for each location) was carried out by oven drying at 1051C until constant weight.
Water samples were collected with sampling bottles (precleaned and rinsed twice with seawater) at a depth of 2 m. Temperature and salinity were recorded. Five samples for each location of 1 L (precleaned polyethylene bottles) were filtered through acid-cleaned 0.45-mm membrane filters, acidified, and stored at 41C for soluble metal analysis. The second five samples, previously acidified, were stored at 41C for subsequent mineralization in a microwave oven and analyzed for total metal
ARTICLE IN PRESS M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
102
concentrations. Water samples were preconcentrated before analysis with the ammonium 1-pyrrolidinedithiocarbamate (APDC) complexation method (Bruland et al., 1979; IRSA, 1984). The procedure is described elsewhere (Campanella et al., 2001).
tissue), and 403 (seawater). One sample of reference material and blanks were included in each analytical batch. Results were in agreement with certified values, and the standard deviations were low, proving good repeatability of the method (Table 2). Mean metal concentrations were calculated with standard deviations. The standard deviations of pooled samples (algae) refer to the variability within different replicates. For mollusc analysis, the standard deviations represent the metal content variation among individuals. All molluscs that were collected and mineralized belong to a narrow and homogeneous range of weight and dimension. The one-way analysis of variance (ANOVA) (log-transformed data) was applied to test the differences between mollusc metal concentrations in different sites and species. Multiple comparison tests were carried out to assess the most significant impact levels in the examined stations. Finally, cluster analysis allowed the assessment of homogeneous groups for the impact of the molluscs observed in this study.
2.2. Determination of heavy metals Heavy metal concentrations were determined by graphite furnace atomic absorption spectrometry (Perkin–Elmer Model 1100B equipped with an HGA-700 graphite furnace and a Perkin–Elmer Model AS-70 autosampler, Perkin–Elmer Italia, Monza, Italy). NH4PO4 was used as matrix modifier for Cd and Pb, Mg (NO3)2 was used for Cr, and the method of standard addition was used (Pb, Cd, Cr). Whenever high levels occurred, Cu and Zn in the biota were determined by flame atomic absorption spectrometry (Varian Techtron Model AA-475, Varian Italia, Milan, Italy). Detection limits and coefficients of variation for the different matrices are reported in Table 1. 2.3. Analytical quality control and statistical analysis
3. Results
The accuracy of measurements was tested with Community Bureau of Reference (BCR) certified reference materials CRM 279 (sea lettuce), 278 (mussel
Metal concentration data in the biota are shown in Tables 3 and 4. Metal concentrations detected in P. pavonica and U. lactuca were quite homogeneous and
Table 1 Limits of detection (LOD)a and coefficients of variation (CV)b for the analysis performed by graphite furnace atomic absorption spectrometrya Macrophytes 1
LOD (mg g Cd Cr Cu Pb Zn
Molluscs d.w.)
0.04 0.04 0.10c 0.06 0.12c
Seawater 1
CV
LOD (mg g
1.9 4.0 3.3 2.9 4.8
0.05 0.05 0.15c 0.07 0.17c
d.w.)
CV
LOD (ng L1 d.w.)
CV
2.5 1.8 4.0 3.7 5.4
20 17 42 22 54
4.5 2.7 3.0 2.8 3.6
a
Calculated on the basis of 20 determinations of the blanks as 3 times the standard deviation of the blank. %, relative to X10 determinations performed on the same sample. c LOD for the analysis performed by flame atomic absorption spectrometry (mg g1 d.w.): Cu, 0.95 (macrophytes) and 1.15 (molluscs); Zn, 1.43 (macrophytes) and 1.62 (molluscs). b
Table 2 Analysis of certified reference materials: certified values and found values (means7SD)a Metal
Cd Cr Cu Pb Zn a
CRM 279 (sea lettuce) (mg g1 d.w.)
CRM 278 (mussel tissue) (mg g1 d.w.)
CRM 403 (seawater) (ng L1)b
Certified
Found
Certified
Found
Certified
Found
0.27470.022 (10.7)c 13.1470.37 13.4870.36 51.371.2
0.2870.03 9.170.5 13.2370.41 14.0570.62 51.472.0
0.3470.02 0.8070.08 9.6070.16 1.9170.04 7672
0.3770.04 0.7270.04 9.2370.26 2.0770.08 7873
20.3572.10 (108.6)c 256.4724.3 25.0875.36 17387197
22.4070.81d 104.472.2 264.6711.3 26.7171.31d 17817112
Number of replicates: CRM 279=22, CRM 278=35, CRM 403=5. Original values given on a w/w basis were recalculated on a w/v basis. c Not certified values. d Values close to the detection limits. b
ARTICLE IN PRESS M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
103
Table 3 Concentrations of metals in the thalli of P. pavonica and U. lactuca (means7SD) (mg g1 d.w.)
Sampling station
Cd
Cr
Cu
Pb
Zn
P. pavonica 1 2 3 4 6
0.5270.09 0.4570.06 0.4770.11 0.3970.12 0.6670.10
3.9670.52 3.2070.64 2.8870.65 3.3470.73 3.8670.63
13.271.0 11.872.0 12.471.4 11.972.2 12.171.7
4.8270.92 3.6770.79 3.7370.53 3.0470.46 4.6471.16
5477 4578 4977 5078 5676
U. lactucab 1 3 4 5 6
0.2070.05 0.1870.05 0.1770.03 0.137 0.04 0.217 0.05
1.5870.25 2.0670.39 1.6270.51 1.397 0.26 1.4970.48
6.470.9 5.470.9 4.970.8 5.870.8 6.370.9
1.6770.28 1.7670.19 1.8870.42 2.2870.74 2.1070.50
4275 4376 5078 3776 5475
a
a b
Number of replicates: 5. Wet wt/d.w. ratio: 5.2070.46. Not available at site 5. Number of replicates: 5. Wet wt/d.w. ratio: 6.1270.43. Not available at site 2.
Table 4 Concentrations of metals (mg g1 d.w.) in the soft tissues of M. turbinata, P. caerulea, and M. galloprovincialis (means7SD) Sampling station
Individuals
Cd
Cr
Cu
Pb
Zn
M. turbinata 1 2 3 4 6
8 6 9 10 5
0.8670.20 1.2970.24 1.4170.29 1.0070.44 1.0470.30
0.4870.19 0.3670.09 0.2770.06 0.7270.13 0.2670.08
50.5710.4 51.4710.9 59.5714.5 66.7720.4 83.0719.7
0.6470.10 0.5270.11 0.6270.13 0.4470.10 0.6770.16
77.7711.2 98.0713.3 129.3722.9 92.5712.0 94.1714.2
P. caeruleab 1 2 3 4 5 6
9 8 21 20 10 14
2.8970.57 3.0170.63 3.6470.56 4.0670.84 3.8370.97 3.7971.14
0.9670.33 0.7870.17 0.9270.29 0.7270.18 0.9170.16 0.8270.28
10.271.4 15.975.0 13.073.2 12.873.9 19.273.6 14.773.5
1.5070.29 0.9870.20 0.8070.15 0.5170.15 0.8470.11 1.0570.27
87.4711.8 95.27 7.1 107.9717.2 100.0720.6 97.1710.8 117.1725.0
M. galloprovincialisc 1 2 3 4 5 6
18 6 22 20 12 9
0.3370.09 0.4970.10 0.3470.08 0.3270.09 0.3870.08 0.4570.08
0.8770.25 0.8070.08 0.4670.12 0.9170.28 1.1170.27 1.3170.29
9.5572.19 9.9571.28 5.5171.47 6.4571.50 11.5072.59 9.4072.05
2.2270.69 2.3870.64 2.4970.61 1.6770.41 1.9670.36 1.7370.24
123726 180738 151740 157736 155732 178734
a
a
Shell size (cm): 1.5–2.1 (height); Wet wt/d.w.: 1.7070.11. 1.8–2.4 (length); Wet wt/d.w.: 2.6670.14. c 4.1–4.8 (length); Wet wt/d.w.: 6.5670.21. b
did not indicate meaningful differences among the sites examined. Contrary to what we expected, the protected park station (station 4, Sassolini) did not exhibit meaningful differences ðP ¼ 0:05Þ in comparison to the other sites (see Table 3). In P. pavonica the observed levels of Cd and Pb were higher than those previously reported by Schlacher-Hoelinger and Schlacher (1998) and lower than our precedent work (Campanella et al., 2001). Data for Cr and Cu were higher than those
previously reported (Campanella et al., 2001; SchlacherHoelinger and Schlacher, 1998). Data for zinc were very close to our previous data on uncontaminated areas (Campanella et al., 2001) and relatively higher than those reported by other authors (Bei et al., 1990; Schlacher-Hoelinger and Schlacher, 1998). Metals concentrated in Padina decreased in the order Zn4Cu4PbXCr4Cd; in Ulva the sequence was Zn4Cu4Pb4Cr4Cd. As for molluscs, the following
ARTICLE IN PRESS M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
104
sequences were observed: Zn4Cu4Cd4PbXCr in M. turbinata; Zn4Cu4Cd4PbXCr in P. cerulea; and Zn4Cu4Pb4Cr4Cd in M. galloprovincialis. Despite a few differences, these sequences were comparable to those already identified by us and by other authors (Campanella et al., 2001; Volterra and Conti, 2000). Metal concentrations detected in algae (pooled samples, see Table 3) did not show significant differences ðP ¼ 0:05Þ on the bioaccumulation trend for the examined sites. The Cd, Cr, and Pb values detected in mussels were in the range of 0.33–0.49, 1.67–2.49, and 0.46–1.31 mg g1 d.w., respectively; while Cu and Zn levels were in the range of 5.51–11.50 and 123–180 mg g1 d.w., respectively. As for Patella, the concentration intervals detected for Cd, Cr, and Pb were 2.89–4.06, 0.72–0.96, and 0.51–1.50, respectively; for Cu and Zn they were 10.2–19.2 and 87.4–117.1 mg g1 d.w. respectively. For Monodonta, the Cd, Cr, and Pb concentration ranges were 0.86–1.41, 0.26–0.72, and 0.44–0.67 mg g1 d.w., respectively; those for Cu and Zn were 50.5–83.0 and 77.7–129.3 mg g1 d.w., respectively. To assess the existence of significant differences in the metal concentrations in molluscs, use of the two-way ANOVA was taken into account. For this purpose, normality and homogeneity of variance tests were carried out. The results led to our discarding that the hypothesis of variance homogeneity on both untransformed and log-transformed data. The one-way ANOVA was therefore performed for each species and for each metal. In the 15 tests the hypotheses of variance normality and homogeneity [tested with the Shapiro–Wilk, Lilliefors (Kolmogorov–Smirnov), and Levene tests] were successful. Thirteen of the 15 tests turned out to be significant on a level of 5%. The other two tests turned out to be significant at a level of 1% (Cd on Monodonta and Patella). Table 5 shows the ANOVA results. After assessing the significance of the averages, multiple comparison tests were carried out to asses which averages differed from the others and to accurately assess which sites accumulated metals differently from the others, with reference to the species that were examined. The multiple comparison tests performed were Bonferroni–Dunn, Tukey, Scheffe`, Sidak, Gabriel, and Hochberg.
3.1. Monodonta From the results of the six multiple comparison tests performed for each species, we observed that Cd bioaccumulation in the S. Agostino station was higher than in Vindicio. Contrary to our assumptions from the mean concentrations at each site (see Table 4), the other sites revealed similar cadmium bioaccumulation. Regarding cadmium in Monodonta, the variance homogeneity hypothesis could be accepted at a 1% level; therefore, other multiple comparison tests were performed (Tamhane, T3 Dunnet, Games-Howell, C Dunnet) that do not presuppose homoschedasticity of distribution performed. These tests also showed significant bioaccumulation differences, which were higher in Porto Romano (station 2) than in Vindicio (station 1). Cr accumulation in the Sassolini station (station 4, the Protected Sea Park) was higher than in the other sites. Cr was higher in the Vindicio site than in Serapo and S. Agostino. There were meaningful differences in Cu only between Vindicio–Serapo and P. Romano–Serapo. Cu bioaccumulation turned out to be higher in the Serapo station than in the other stations. Pb bioaccumulation was the same in the Sassolini and P. Romano stations, and in the Sassolini station it was lower than in the other sites. S. Agostino revealed a higher ðP ¼ 0:05Þ Zn than the other stations.
3.2. Patella cerulea Cd was significantly lower ðP ¼ 0:05Þ in Vindicio and P. Romano than in the other stations. According to the F test (see Table 5), the Cr level was similar in all stations. The Cu concentration was quite different among Vindicio–P. Romano, Vindicio–S.Croce, and Vindicio– Serapo, bioaccumulation being lower ðP ¼ 0:05Þ in Vindicio. On the other hand, in S. Croce Cu was higher ðP ¼ 0:05Þ than in Vindicio, S. Agostino, and Sassolini. Pb was higher in Vindicio ðP ¼ 0:05Þ than in the other stations, and in Sassolini it was significantly lower ðP ¼ 0:05Þ than in the other sites. Zn bioaccumulation turned out to be similar in P. Romano, S. Agostino, S. Croce, and Sassolini.
Table 5 Results of the one-way analysis of variance for metal concentrations Factor of variation
M. turbinata P. caerulea M. galloprovincialis
Cd
Cr
Cu
Pb
Zn
F
P
F
P
F
P
F
P
F
P
3.438 3.735 4.016
0.019 0.004 0.001
21.806 1.801 19.848
0 0.123 0
4.029 7.760 21.919
0.009 0 0
5.851 35.608 5.657
0.001 0 0
12.990 2.744 4.236
0 0.025 0.001
ARTICLE IN PRESS M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
3.3. Mussels Cd was higher in the P. Romano site than in Vindicio, S. Agostino, and Sassolini. For Monodonta, the variance homogeneity hypothesis could be accepted at a 1% level. Therefore, other multiple comparison tests were performed (see above). This additional study also showed significant differences between Cd accumulation for Serapo–Vindicio and Serapo–S.Agostino, bioaccumulation being higher in Serapo. Cr was significantly lower in S. Agostino than in the other stations, while Serapo had the highest metal levels. In Sassolini, the Cu was significantly lower than in almost all other sites. Pb was significantly lower in Sassolini–Vindicio than in Sassolini–S.Agostino. Zn was significantly lower in the Vindicio station than in the P. Romano, Sassolini and Serapo stations. Thus, the tests we performed showed that the trend of metal bioaccumulation in molluscs was not clearly univocal; that is to say, no one station showed an accumulation higher than that of any other, at least for the metals we studied. To obtain homogeneous groups for metal bioaccumulation, we performed cluster analyses in our sites. The cluster analysis showed two homogeneous groups for Monodonta (see Fig. 2). The first comprised P. Romano, S. Agostino, and Serapo, and the second comprised Vindicio and Sassolini. The analysis clearly showed that the former group presented metal bioaccumulation higher than that of the latter.
105
On the other hand, the cluster analysis with Patella revealed the presence of three groups. The first included Vindicio, P. Romano, Sassolini, and Serapo, the second comprised the S. Agostino site, and the third comprised the Santa Croce site (Fig. 3). The latter also turned out to be the most dissimilar, with generally higher metal bioaccumulation levels. The cluster analysis revealed three groups of mussels (Fig. 4). The first included Vindicio, S. Croce, P. Romano, and Serapo, the second the Sassolini station, and the third the S. Agostino station. The second and third groups evidenced lower contamination levels. Cluster analysis results (Figs. 2–4) gave further evidence that the Sea Park Station (Sassolini) was not noticeably different from the others, as opposed to what we expected. The hypothesis that Protected Sea Park would be cleaner than the other sites must therefore be reconsidered. The data for the common species were compared, through cluster analysis, with the data we derived from uncontaminated sites in the Sicilian Sea (Favignana Island) (Campanella et al., 2001). This research confirmed the diversity between these two marine ecosystems and allowed us two identify two different groups: the stations in the Gulf of Gaeta and the Sicilian stations. The two species we considered, Monodonta and Patella, revealed that the Gulf of Gaeta had higher metal contamination than the uncontaminated sites in the Sicilian Sea (Figs. 5 and 6).
Fig. 2. Classification of the sampling stations (cluster analysis) for Monodonta turbinata.
Fig. 3. Classification of the sampling stations (cluster analysis) for Patella cerulea.
ARTICLE IN PRESS 106
M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
Fig. 4. Classification of the sampling stations (cluster analysis) for Mytilus galloprovincialis.
Fig. 5. Classification of the sampling stations (cluster analysis) for Monodonta turbinata of Gaeta Gulf and Favignana island (Sicily). Punta Sottile, Preveto, and Cala Rossa are the clean sites of Favignana.
Fig. 6. Classification of the sampling stations (cluster analysis) for Patella cerulea of Gaeta Gulf and Favignana island (Sicily). Preveto, Cala Rossa, Cala Azzurra, and Punta Sottile are the clean sites of Favignana.
In order to assess the concentration factors (CFs) for observation of the bioaccumulation ability of every single species, analyses of the waters were performed on both metals in solution and total metals. The CF may be used to evaluate the state of conservation of an ecosystem or to monitor its state (Conti and Cecchetti, 2001). Table 6 shows the results of the water analysis and Table 7 shows the mean metal concentrations detected in both biota and seawater. The data on seawater metals must be interpreted with some caution, due to the high variability of metal concentrations in
seawater, which depends on several factors (physicochemical, environmental, etc.). The determination of trace metals in marine waters is a critical part of biomonitoring studies because of its many important analytical aspects. The levels of trace metals are often low in coastal waters. It is therefore important to prevent the use of inadequate sampling and sample treatment techniques. [For instance, sample contamination may arise from components of the sampling gear (Achterberg, 2000)].
ARTICLE IN PRESS M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
107
Table 6 Total and dissolved metal concentrations in coastal seawater samples (means7SD) Sampling station
Cd (ng L1)
Cr (ng L1)
Cu (mg L1)
Pb (mg L1)
Zn (mg L1)
Total metals 1 2 3 4 5 6
30175 25277 29178 24776 23878 21477
39172 42173 38877 51171 509710 43072
2.1270.07 3.0570.03 2.6370.04 2.5170.08 3.2170.12 2.8170.09
2.2170.02 3.1070.06 2.0070.11 3.1570.03 2.2370.09 2.3670.06
10.4070.12 9.6370.35 7.5070.22 10.1570.18 9.6470.38 8.9170.47
Dissolved metals 1 2 3 4 5 6
13972 11576 13275 12071 14274 13373
27974 341711 27772 33879 37275 29371
1.3470.02 0.8270.01 1.4370.03 1.6770.03 1.3270.01 0.7970.03
1.6570.05 2.1270.03 1.3270.03 1.0570.02 2.8370.01 1.1170.02
5.9670.16 6.3770.21 4.7570.11 5.0970.10 7.2170.24 5.1570.09
Number of replicates: 5. Mean values of salinity and temperature during sampling: S ¼ 3671; t ¼ 18711C: Table 7 Mean metal concentrations in biota and seawater from stations 1 to 6 (means7SD) Matrix
(mg g1 d.w.) Cd
Cr
Cu
Pb
Zn
P. pavonica U. lactuca M. turbinata P. caerulea M. galloprovincialis
0.5070.07 0.1870.06 1.1270.48 3.5470.87 0.3870.11
3.4570.70 1.6370.60 0.4270.15 0.8570.24 0.9070.21
12.371.8 5.871.1 61.1720.1 14.374.9 8.3871.62
3.9870.67 1.9470.38 0.5870.16 0.9570.19 2.0770.49
51711 45712 98723 100.8718 152737
Total metals (mg L1)a Dissolved metals (mg L1)
0.2670.06 0.1370.04
0.4470.03 0.3270.05
2.7270.18 1.2370.12
2.5070.11 1.6870.15
a
9.371.58 5.772.11
Soluble+particulate.
Table 8 CFsa 103 calculated with reference to soluble and total metal concentrations in seawater Matrix
Cd
Cr
Cu
Pb
Zn
Soluble P. pavonica U. lactuca M. turbinata P. caerulea M. galloprovincialis
3.99 1.43 8.93 28.22 3.03
11.17 5.28 1.36 2.75 2.91
10.36 4.88 52.41 12.05 7.06
2.45 1.19 0.36 0.59 1.27
9.27 8.18 17.82 18.33 27.64
Total P. pavonica U. lactuca M. turbinata P. caerulea M. galloprovincialis
1.99 0.72 4.46 14.11 1.51
8.13 3.84 0.98 2.00 2.12
4.69 2.21 23.70 5.44 3.19
1.65 0.80 0.24 0.39 0.85
5.68 5.01 10.92 11.23 16.94
a
CF ¼ CO =CSW ; where CO =mean concentration in the organism (mg g1 d.w.) and CSW =mean concentration in seawater (mg L1).
4. Discussion Our results support the hypothesis that the coastal area of the Gulf of Gaeta has significant basal contamination levels, which, however, do not reach those of clearly contaminated areas. In any case, the
expression ‘‘background levels’’ must be used with some caution, as its meaning differs according to the local geochemical or hydrodynamic conditions of the body of water under examination (Conti, 2002). Our work shows the importance of using several possible biomonitors when studying the quality of ecosystems, such as algae,
108
Table 9 Selected references of metal concentrations (mg/g d.w.) of molluscs from different geographical areas (mean values and ranges) Species
Sites
Cd
M. galloprovincialis
Four sites, Gulf of Trieste (NE Italy)
41 sites along Italian coastal areas Bakar Bay, (ex-Yugoslavia)
Venice Lagoon, Gulf of Trieste (NE Italy)
Grado e Marano Lagoon Split, Kastela Bay (Croazia)
Pb
Zn
(1) 0.21a
1.26a
0.84a
16.5a
(2) 0.29a (3) 0.29a (4) 0.23a 0.4–6 Site 1
1.41a 1.48a 1.41a 2.4–15.5 Site 1
3.29a 1.85a 0.64a 2.7–117 Site 1
27.4a 25.4a 14.6a 97–644
0.325a 0.120–0.380 Site 2 0.192a 0.130–0.270 Site 3 0.066a 0.018–0.160
1.229a 0.830–2.230
0.137a
1.57a
0.925a 0.480–1.790 Site 2 1.660a 1.100–2.200 Site 3 0.405a 0.160–0.670 Site 4 0.841a 0.400–1.300 0.19a
20.8a
Site 1
Site 1
Site 1
Site 1
0.12–0.38a Site 2 0.18–0.42a 0.062–0.48a
0.77–2.23a Site 2 0.50–3.26a
0.48–1.79a Site 2 1.21–5.74a 0.15–11.8a
6.90–29.60a Site 2 13.70–42.20a
0.5–29
0.16a
0.44a
0.31a
1.41a
0.32a
0.13–0.78 0.23a 0.17–0.35 0.22a
0.18–6.52 1.28a 0.38–3.64 0.94a
0.12–1.09 1.17a 0.58–3.66 0.17a
0.13–0.35
0.11–0.26 0.735a
0.273a
12.23a
0.009–2.53 5.08–20.89
0.014–1.18 6.18–80.26
1.48–37.32 82–185
Venice Lagoon, 44 sites
0.134a
0.28–2.78 0.32–0.75a 0.265a
Venice Lagoon, six sites
n.d–1.64 0.05–4.54
n.d–1.20 0.37–20.38
Notes
References Majori et al. (1978)
Eisler (1981) Gabrielli Favretto and Favretto (1984)
Pool of five individuals
Martincˇic´ et al. (1984) Favretto et al. (1987)
Pools of 40–50 individuals Five pools of three individuals
Giordano et al. (1989, 1991) Ozretic et al. (1990)
36 stations
Majori et al. (1991)
22 stations
Majori et al. (1991)
36 stations
Majori et al. (1991)
Six stations 10 individuals for each site
Vukadin and Odzak (1991) Zatta et al. (1992)
Pools of 10–15 individuals
Widdows et al. (1997)
ARTICLE IN PRESS
Nine stations, Lim Fjord (N Adriatic Sea) Two sites, Muggia Bay, Trieste (NE Italy)
Cu
M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
NW Mediterranean Sea Four sites, Gulf of Trieste (NE Italy)
Cr
Chile, four sites
Bakar Bay (ex-Yugoslavia)
4.13 2.58–5.99 0.38 0.33–0.49 700a
Bakar Bay (ex-Yugoslavia)
417–1140 0.50a
This work P. caerulea
Site 2: marine reseve of S. Maria di Castellabate
Monodonta turbinata
Bakar Bay, (ex-Yugoslavia)
0.70a
0.30
1.7
Monodonta articulata
Larymna (Grecia) (contaminated area)
Monodonta mutabilis
Larymna (Grecia) (contaminated area)
0.20
Site 1
Site 1
0.62a 0.39–0.91 Site 2
1.42a 0.92–2.31 Site 2
0.40a 0.27–0.53 3.54 2.89–4.06 0.20a
1.62a 0.91–4.04 0.95 0.51–1.50 0.61a
0.85 0.72–0.96
14.3 10.2–19.2
0.19–0.22 Larymna (Grecia) (contaminated area) Favignana Island, Sicily (Italy), (uncontaminated area) This work
152 123–180
This work Pools 40 individuals
Taramelli et al. (1991)
Pools of 3–4 individuals
Ozretic et al. (1990)
0.61–0.79 5
Campanella et al. (2001)
Camoni et al. (1980)
100.8 87.4–117.1
This work Pools of 9–12 individuals
Ozretic et al. (1990)
Pools of 10 individuals
Nicolaidou (1994)
0.59–0.63 5.6
109.1
65.7
1.44
0.31
10.9
0.24
31
Campanella et al. (2001)
1.12 0.86–1.41
0.42 0.26–0.72 11.2
62.2 50.5–83.0 125.4
0.58 0.44–0.67
98 77.7–129.3 102.0
This work
4.6
136.6
98.8
Pools of 10 individuals
Nicolaidou (1994)
Pools of 10 individuals
Nicolaidou (1994)
N.d.=not detectable. a Wet weight.
ARTICLE IN PRESS
This work
0.48–0.52 4.41
8.38 5.51–11.50
Bruhn et al. (1999)
M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
Favignana Island, Sicily (Italy), uncontaminated area Tyrrhenian Sea, Naples (Italy) Site 1: Naples Bay
0.90 0.46–1.31
0.32 0.18–0.40 2.07 1.67–2.49
109
ARTICLE IN PRESS 110
M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
bivalve molluscs (filtering organisms), and herbivores (gastropod molluscs). Each biomonitor responds to a particular metal fraction of the water body: mussels to the fraction present in particulates, algae to metals in solution. Moreover, metal concentrations in algae could affect the concentrations of metals in gastropods (the phenomenon of biomagnification) (Conti, 2002). The data available on metal concentrations in the water of this coastal area are quite scarce. The Cd and Zn levels detected in the present study (Table 6) are close to those we had found previously in the Sea of Sicily (Campanella et al., 2001); on the other hand, the Cr, Cu, and Pb levels are 1.5–2.0 times higher. Generally, the data on waters (Table 6) are comparable with those concerning the coastal areas of the Tyrrhenian Sea (Alpha et al., 1982). Table 8 reports the calculation of CFs for the species examined. As mentioned above, the data in Table 8 must be viewed cautiously, as CFs could be influenced by several factors. The passage of contaminant through the trophic chain can affect CF values. The data shown in Table 8 confirm the high aptitude of the species we examined as bioaccumulators. It is actually possible to observe that very high CFs can be calculated with reference to soluble metal concentrations. Regarding net accumulation, we can see that Patella caerulea was the strongest accumulator of Cd, P. pavonica that of Cr and Pb, M. turbinata that of Cu. These results back up our previously reported findings for these species for the Sea of Sicily (Campanella et al., 2001). M. galloprovincialis turned out to be the highest Zn accumulator. Some authors have reported a reasonable ability of molluscs for regulating Cu and Zn (Amiard-Triquet et al., 1986). As is well known, only a few marine organisms are able to regulate metal concentration levels in tissues. This ability, however, exists for only a limited range of environmental concentrations for some essential elements such as Cu and Zn (Phillips, 1995). The total metal contents in molluscs found in the present work are comparable with those found in the literature (Table 9). In particular, the levels detected in M. galloprovincialis rank generally quite low compared to some highly contaminated sites (Venice Lagoon). For P. caerulea and M. turbinata, it is clear that the mean metal concentrations of Cr, Cu, Pb, and Zn are higher than those in uncontaminated sites (Favignana Island, Sicily; Table 9), while the average Cd levels in the Gulf of Gaeta have turned out to be similar to those in the Sicilian Sea. When trace metal levels from different areas are compared, one must be quite wary, as data are often reported in fresh weight (Table 9). While the content of water does not change too much for fish, it is very variable for the molluscs, being affected by such factors as habitat, vital condition, and pretreatment and sample conservation procedures after sampling. For instance,
Crisetig et al. (1984) observed in M. galloprovincialis variable fresh weight/d.w. ratios in the range of 4.32– 8.50 for mussels as long as 4–5 cm; Martincˇic´ et al. (1992) found variable ratios in the interval of 4.27–12.99 in mussels transplanted into an estuary environment (Krka River, eastern Adriatic coast). We think that data concerning d.w. are susceptible to less variability than those of fresh weight.
5. Conclusions The use of bioindicators turned out to be very valuable for the study of a coastal area with fairly significant basal contamination levels. The green algae U. lactuca, the brown algae P. pavonica, the bivalve mollusc M. galloprovincialis, and the two gastropod molluscs M. turbinata and P. cerulea possess high potential as cosmopolitan biomonitors for trace metals in the Mediterranean. These species have the necessary prerequisites for use as bioindicators: they are easy to identify and to sample, are available all year round, and are present in almost all coastal areas of the Mediterranean Sea. Samples of the above-mentioned species were collected at six stations along the Gulf of Gaeta (Tyrrhenian Sea, central Italy) The picture of bioavailable Cd, Cr, Cu, Pb, and Zn was not homogeneous. Statistical analyses (one-way ANOVA, multiple comparison tests, cluster analysis) showed that no one site was more contaminated than the others. Therefore, the supposition of the protected Marine Park (station 4) as a control, ‘‘clean’’ monitoring site had to be rejected; this is a useful indication for to further elaboration and betterment of the environmental protection policies for marine sites. The cluster analysis for molluscs also showed the presence of homogeneous groups within the examined sites and a clear differentiation as opposed to the impact levels of unpolluted sites in other coastal areas of Southern Italy. The use of molluscs in biomonitoring studies demands great accuracy, especially with regard to the dimensions and weight of individuals. This aspect is something of a critical issue. It is necessary to pick individuals belonging strictly to very narrow weight and size ranges. If this were not possible, other methods of statistical analysis could be used [for instance, covariance (ANCOVA), which can examine samples differing in weight and size] (Conti, 2002; Cubadda et al., 2001). The species examined showed a great ability to accumulate concentrations of metals several thousands times those detected in marine water (soluble fraction). It is worthwhile to remark that P. caerulea and M. galloprovincialis are commonly consumed seafood in many Mediterranean countries. The passage of trace
ARTICLE IN PRESS M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
metals from the abiotic section to the tissues of these species is highly interesting for food safety. Although the species we studied in the present work and suggested as biomonitors present numerous advantages, more information and more studies are necessary to clarify accumulation patterns. It is important, as well, not to overlook the possible existence of regulating and/ or competition mechanisms of metals inside tissues. Also, the most recent developments in the field of analytical techniques can significantly contribute to our knowledge in this field (Conti et al., 2002). It is quite important to take into account the knowledge derived from the study of metal speciation in seawater, along with traditional biomonitoring investigations, for use in environmental protection policies of the future.
Acknowledgments The authors thank Prof. Francesco Battaglia and Dr. Domenico Cucina for they valuable support with the statistical analysis. We also thank Dr. Luca Ciprotti, Dr. Carlo Sucapane, and Fabio Spadoni for their valuable assistance. Grateful thanks are due to Maria Brighi, Petra Tomassini, and Pierluigi Tomassini, for their support during our stay in Formia.
References Achterberg, E.P., 2000. Automated techniques for real-time shipboard determination of dissolved trace metals in marine surface waters. In: Conti, M.E., Botre`, F. (Eds.), The Control of Marine Pollution: Current Status and Future Trends; Int. J. Environ. Pollut. 13, 249–261, Inderscience Enterprises Ltd., Milton Keynes, UK. Alpha, M., Caristi, C., Cimino, G., Ziino, M., 1982. Heavy metals in Ionian and Tyrrhenian waters from a section of the Sicilian coast. Mar. Pollut. Bull. 13 (11), 399–400. Amiard-Triquet, C., Berthet, B., Metayer, C., Amiard, J.C., 1986. Contribution to the ecotoxicological study of cadmium, copper and zinc in the mussel Mytilus edulis. II. Experimantal study. Mar. Biol. 92, 7–13. Bei, F., Papathanassiou, E., Catsiki, V.A., 1990. Heavy metals concentrations in selected marine species from Milos Island (Aegean Sea). Rapp. P.-V. Re´un. CIESM 32 (1), 126. Bruhn, C.G., San Francisco, N.A., Neira, J.Y., 1999. Determination of cadmium and lead in mussels by tungsten coil electrothermal atomic absorption spectrometry. Talanta 50, 967–975. Bruland, K.W., Franks, R.P., Knauer, G.A., Martin, J.H., 1979. Sampling and analytical methods for the determination of copper, cadmium, zinc, and nickel at the nanogram per liter level in sea water. Anal. Chim. Acta 105, 233–245. Bryan, G.W., Langston, W.J., 1992. Bioavailability, accumulation and effects of heavy metals in sediments with special reference to United Kingdom estuaries: a review. Environ. Pollut. 76, 89–131. Camoni, I., Cotta-Ramusino, F., Funari, E., Giordano, R., Ovidi, P., Dohrn, P., Ballio, A., 1980. Heavy metals and organochlorine residues in marine animals and coastal sediments of Tyrrhenian Sea around Naples. Ann. Ist. Super. Sanita´ 16, 645–656.
111
Campanella, L., Conti, M.E., Cubadda, F., Sucapane, C, 2001. Trace metals in seagrass, algae and molluscs from an uncontaminated area in the Mediterranean. Environ. Pollut. 111, 117–126. Conti, M.E., 2002. SEAM Ed., Rome, Il monitoraggio biologico della qualita` ambientale. Rome, 181pp. Conti, M.E., Cecchetti, G., 2001. Biological monitoring: lichens as bioindicators of air pollution assessment—a review. Environ. Pollut. 114, 471–492. Conti, M.E., Tudino, M.B., Muse, J.O., Cecchetti, G., 2002. Biomonitoring of heavy metals and their species in the marine environment: the contribution of atomic absorption spectroscopy and inductively coupled plasma spectroscopy. Res. Trends Appl. Spectrosc., in press. Crisetig, G., Cattani, O., Viviani, R., 1984. Metalli tossici nei mitili dell’alto e medio adriatico. Arch. Vet. Ital. 35, 10–16. Cubadda, F., Conti, M.E., Campanella, L, 2001. Size-dependent concentrations of trace metals in four mediterranean gastropods. Chemosphere 45, 561–569. Eisler, R., 1981. Trace Metal Concentrations in Marine Organisms. Pergamon Press, Oxford, 685pp. Favero, N., Cattalini, F., Bertaggia, D., Albergoni, V., 1996. Metal accumulation in a biological indicator (Ulva rigida) from the Lagoon of Venice (Italy). Arch. Environ. Contam. Toxicol. 31, 9–18. Favretto, L., Gabrielli Favretto, L., Felician, L., 1987. Principal component analysis for the identification of pollution sources in mussel survey by trace metals. Lebensm. Unters. Forsch. 184, 101–109. Fo¨ster, P., 1976. Concentrations and concentration factors of heavy metals in brown algae. Environ. Pollut. 10, 45–53. Gabrielli Favretto, L., Favretto, L., 1984. Heavy metals at trace level in edible mussels (Mytilus galloprovincialis Lamarck) from the Gulf of Trieste. Lebensm. Unters. Forsch. 179, 197–200. Giordano, R., Arata, P., Ciaralli, L., Rinaldi, S., Giani, M., Cicero, A.M., Costantini, S., 1991. Heavy metals in mussels and fish from Italian coastal waters. Mar. Pollut. Bull. 22, 10–14. Giordano, R., Arata, P., Rinaldi, S., Ciaralli, L., Giani, M., Rubbiani, M., Costantini, S., 1989. Mercury, cadmium and lead levels in marine organisms (Mytilus galloprovincialis LMK) collected along the Italian coasts. Ann. Ist. Super. Sanita´ 25, 511–516. Haritonidis, S., Malea, P., 1995. Seasonal and local variation of Cr, Ni and Co concentrations in Ulva rigida C. Agardh and Enteromorpha linza (Linnaeus) from Thermaikos Gulf, Greece. Environ. Pollut. 89, 319–327. Haritonidis, S., Malea, P., 1999. Bioaccumulation of metals by the green alga in Ulva rigida from Thermaikos Gulf, Greece. Environ. Pollut. 104, 365–372. IRSA, 1984. Metodi di analisi per acque di mare: Cadmio (225.2), Rame (260.1). Quad. Ist. Ric. Acque 59. Luoma, S.N., 1983. Bioavailability of trace metals to aquatic organisms—a review. Sci. Total Environ. 28, 1–22. Majori, L., Nedoclan, G., Daris, F., Modonutti, G.B., 1991. Metal distribution (Cd, Cr, Cu, Fe, Mn, Pb, Zn) in Mytilus galloprovincialis LMK in coast and lagoon areas in northern Adriatic. Rev. Int. Oceanogr. Med. 101–104, 225–228. Majori, L., Nedoclan, G., Modonutti, G.B., Daris, F., 1978. Study of the seasonal variations of some trace elements in the tissues of Mytilus galloprovincialis taken in the Gulf of Trieste. Rev. Int. Oce´anogr. Med. XLIX, 37–40. Martincˇic´, D., Kwokal, Z., Peharec, Z., Margus, D., Branica, M., 1992. Distribution of Zn, Pb, Cd and Cu between seawater and transplanted mussels (Mytilus galloprovincialis). Sci. Total Environ. 119, 211–230. Martincˇic´, D., Nurnberg, H.W., Stoeppler, M., Branica, M., 1984. Bioaccumulation of heavy metals by bivalves from Lim Fjord (North Adriatic Sea). Mar. Biol. 81, 177–188.
ARTICLE IN PRESS 112
M.E. Conti, G. Cecchetti / Environmental Research 93 (2003) 99–112
Muse, J.O., Stripeikis, J.D., Ferna`ndez, F.M., d’Huicque, L., Tudino, M.B., Carducci, C.N., Troccoli, O.E., 1999. Seaweeds in the assessment of heavy metal pollution in the Gulf San Jorge, Argentina. Environ. Pollut. 104, 315–322. Nicolaidou, A., 1994. Identification of biological indicators of heavy metals at the site of a smelting factory. UNEP/FAO, Final Reports on Research Projects Dealing with the Effect of Pollutants on Marine Organisms and Communities. MAP Technical Reports Series No. 80, Athens, pp. 25–38. Ozretic, B., Krajnovic-Ozretic, M., Santin, J., Medjugorac, B., Kras, M., 1990. As, Cd, Pb, Hg in benthic animals from the Kvarner– Rijeka region, Yugoslavia. Mar. Pollut. Bull. 21, 595–598. Phillips, D.J.H., 1977. The use of biological indicator organisms to monitor trace metal pollution in marine and estuarine environment—a review. Environ. Pollut. 13, 281–317. Phillips, D.J.H., 1995. The chemistries and environmental fates of trace metals and organochlorines in aquatic ecosystems. Mar. Pollut. Bull. 31, 193–200. Rai, L.C., Gaur, J.P., Kumar, H.D., 1981. Phycology and heavy-metal pollution. Biol. Rev. 56, 99–151. Rainbow, P.S., 1995. Biomonitoring of heavy metal availability in the marine environment. Mar. Pollut. Bull. 31 (4–12), 183–192. Schlacher-Hoenlinger, M.A., Schlacher, T.A., 1998. Differential accumulation patterns of heavy metals among the dominant macrophytes of a Mediterranean seagrass meadow. Chemosphere 37, 1511–1519. Seeliger, U., Edwards, P., 1977. Correlation coefficients and concentration factors of copper and lead in seawater and benthic algae. Mar. Pollut. Bull. 8 (1), 16–19. Sericano, J.L., 2000. The Mussel Watch approach and its applicability to global chemical contamination monitoring programmes. In: Conti, M.E., Botre`, F. (Eds.), The Control of Marine Pollution:
Current Status and Future Trends; Int. J. Environ. Pollut. 13 (1–6), 340–350, Inderscience Enterprises Ltd., Milton Keynes, UK. Struck, B.D., Pelzer, R., Ostapczuk, P., Emons, H., Mohl, C., 1997. Statistical evaluation of ecosystem properties influencing the uptake of As, Cd, Co, Cu, Hg, Mn, Ni, Pb and Zn in seaweed (Fucus vesiculosus) and common mussel (Mytilus edulis). Sci. Total Environ. 207, 29–42. Taramelli, E., Costantini, S., Giordano, R., Olivieri, N., Perdicaro, R. (1991). Cadmium in water, sediments and benthic organisms from a stretch of coast facing the thermoelectric power plant at Torvaldiga (Civitavecchia, Rome). UNEP/FAO, Final Reports on Research Projects Dealing with Bioaccumulation and Toxicity of Chemical Pollutants. MAP Technical Reports Series No. 52, Athens, pp. 15–31. Volterra, L., Conti, M.E., 2000. Algae as biomarkers, bioaccumulators and toxin producers. In: Conti, M.E., Botre`, F. (Eds.), The Control of Marine Pollution: Current Status and Future Trends; Int. J. Environ. Pollut. 13 (1–6), 92–125, Inderscience Enterprises Ltd., Milton Keynes, UK. Vukadin, I., Odzak, N., 1991. Fate and distribution of chromium in waters, sediments and mussels of the Kastela Bay. In: Gabrielides, G.P. (Ed.), Proceedings of the UNEP/FAO/IAEA Consultation Meeting on the Accumulation and Transformation of Chemical Contaminants by Biotic and Abiotic Processes in the Marine Environment, La Spezia, Italy, 24–28 September 1990. MAP Technical Reports Series No. 59, Athens, pp. 363–369. Widdows, J., Nasci, C., Fossato, V.U., 1997. Effects of pollution on the scope for growth of mussels (Mytilus galloprovincialis) from the Venice lagoon, Italy. Mar. Environ. Res. 43 (1/2), 69–79. Zatta, P., Gobbo, S., Rocco, P., Perazzolo, M., Favarato, M., 1992. Evaluation of heavy metal pollution in the Venetian lagoon by using Mytilus galloprovincialis as biological indicator. Sci. Total Environ. 119, 29–41.