Studies were carried out to investigate the contamination of Piracicaba and Mogi-Guaç…

• Abstract
• Introduction
• Materials and methods
• Results
• Discussion
• Conclusions
• References
• Tables
• Figures

There are many factors which may affect the bioavailability and intake of heavy metals by the organisms, such as variations in the physical-chemical parameters in the surrounding water, such as pH, Ca²⁺, total suspended solids (TSS), dissolved organic carbon (DOC) among others (Van Hattum et al., 1996); variations in water flow (high and low), which may cause dilution of the concentrations of heavy metals in water (Camusso et al., 1994); and variations in the physiology of organisms (Kraak et al., 1991; Naimo et al., 1992). These factors remain in constant interaction in the environment and these interactions could cause of different intake patterns of heavy metals by organisms.

Bivalves were chosen for this study because they meet many of the requirements of a good biological monitor (Phillips, 1980). They are somewhat sedentary, regionally abundant, long lived and have adequate tissue mass for analysis. They readily accumulate many metals and their body burden seems to reflect mean exposure levels over time (Naimo, 1995). Consequently, such organisms have been largely used in programmes of biological monitoring in either salt water (Farrington, 1983; De Gregori et al., 1994; McConnell and Harrel, 1995; Avelar et al., 2000) or in freshwater (Manly and George, 1977; Foster and Bates, 1978; Millington and Walker, 1983; Abaychi and Mustafa, 1988; Hameed and Raj, 1990; Kraak et al., 1991; Camusso et al., 1994; Valdovinos et al., 1998; Villar et al., 1999; Rutzke et al., 2000). The species adopted as a biological monitor in this study (A. trapesialis) has been successfully used by different investigators (Avelar et al., 1991; Roma and Longo, 1991; Lopes et al., 1992) in monitoring programmes on some rivers in São Paulo State.

Differences in metal concentrations between specimens may be also due to differences in body weight (Kraak et al., 1991). In order to rule out this possibility, we correlated concentrations of Cd and Pb with body weight of A. trapesialis. Similar to other studies (Hammeed and Raj, 1990; Secor et al., 1993), no significant correlations were found. Therefore, we excluded the possibility of variations of metal concentrations due to differences in body weight. Differences between sexes (male and female) are not the case in A. trapesialis, once this species is hermaphrodite (Hebling, 1976). However, other physiological parameters have to be taken into consideration, such as the reproductive cycle, stress caused by manipulation and adaptation to another environment, which is impossible to control and which might affect the intake of Cd and Pb by the mussels. The changes in the physiology may result in changes of rate of intake, storage and excretion of metals by organisms, resulting in changes in concentrations during the year (Naimo et al., 1992). Bivalves, in particular, exhibit growth and reproductive cycles that result in seasonal changes in the metabolism of the animals, which may affect the rate of absorption and/or excretion of some metals (Kraak et al., 1991; Naimo et al., 1992). Therefore, the different trends observed for the seasonal variations on the Cd and Pb concentrations in A. trapesialis may be due to changes in animal physiology. In other studies, variations of metal concentrations during the year were attributed to variations of metal concentrations in the surrounding water. Camusso et al. (1994) observed that the dissolved metal concentrations in the Po River (Italy) showed a general trend to decrease as flow increases, which was attributed to a dilution effect. Abaychi and Mustafa (1988) observed the same in the Shatt Al-Arab River in Iraq. In Brazil, Boldrini et al. (1983) reported that metal concentrations in sediment were higher during the period of low water for the Pardo and Mogi-Guaçu Rivers. Therefore, due to the smaller water volume during the low water period and, consequently, a lesser dilution effect, it was expected to find higher metal concentrations in water during this period, which would reflect higher metal concentrations in the animals. This trend was observed only for Cd in animals collected in the Piracicaba River basin, whereas for Pb an inverse trend was observed, i.e., higher concentrations occurred during the high water period. On the other hand, no seasonal variation was observed in the Mogi-Guaçu River. Thus, the mussel A. trapesialis did not show a seasonal trend of accumulation of Cd and Pb. In addition, other factors, such as variations in the environmental conditions or in the physiology of the animal, might have affected the intake of metals by the organisms during the year.

Cadmium did not accumulate in the test animals in any season, and concentrations were not different among sampling sites in both basins. It has been suggested that the freshwater mussel Dreissena polymorpha can not regulate the Cd concentration in its tissues (Camusso et al, 1994). To support this hypothesis there is the fact that a strong linear relation between Cd accumulation was found in zebra mussels (Del Castilho et al., 1984). The amount of dialysable "free" cadmium in the Rhine River water, indicated that Cd uptake involves selective binding by organic ligands and that absorption is related to the free available Cd species (Del Castilho et al., 1984). Kraak et al. (1993) suggested that every increase in dissolved Cd concentration in water resulted in a significant increase in its concentration in mussels. The same was observed by Graney et al. (1983) and Timmermans (1993). Consequently, if Cd were available in the environment, mussels would probably accumulate it. Therefore, the fact that Cd concentrations in test animals of the Piracicaba and Mogi-Guaçu basins were not higher than in the control group could mean that Cd is in low concentration in the environment or it is not available for absorption by these animals.

Lead in both basins were always higher in test animals than in the control group. Although the number of studies of Pb accumulation in aquatic organisms is smaller than studies of Cd, the few studies indicate that Pb is also a non-essential element. Thus, like Cd, it appears that mussels can not regulate internal Pb concentrations (Timmermans, 1993). For instance, after 60 days of exposure to high Pb concentrations, the internal concentration in Dreissena polymorpha increased from 3.7 µg g^-1 to approximately 6.0 µg g^-1, suggesting a cumulative uptake of Pb (Camusso et al., 1994). Therefore, the higher concentrations of Pb found in mussels of the present study may indicate that this element was available in the river water. Furthermore, Pb concentrations in the mussels showed spatial variations among the sampling sites, reflecting the different levels of exposure in the environment (Table 2).

Generally, good relationships were found between the proximity of human impacts, such as number of inhabitants or percent of urbanized area in the Piracicaba basin, and variables that characterized the elemental composition of the rivers, such as dissolved oxygen, dissolved inorganic carbon, Cl^-, Ca among others (e.g. Ometto et al., 2000). The Mogi basin has smaller population than Piracicaba basin, consequently a smaller volume of domestic sewage being discharged into rivers. Furthermore, the number of industries in the Mogi basin is also significantly smaller than in the Piracicaba basin. Thus, it would be expected that heavy metal concentrations should be higher in the Piracicaba basin. However, this was not true for heavy metal concentrations in mussel tissues. Firstly because the animals exposed in the sampling sites of the less impacted watershed (Mogi-Guaçu) showed higher concentrations than the animals of the more impacted watershed (Piracicaba). Secondly because high concentrations of Pb were observed in the headwaters of the Piracicaba basin (P1 and P2), and in the lower Mogi region (M3), which are the less disturbed regions of the basins. With regard to that, Bilos et al. (1998) observed that concentrations of Cr, Mn, Ni and Cd in Corbicula fluminea in the La Plata River basin (Argentina) presented decreasing values with proximity of major urban centers (the most industrialized and populated region of the basin). Therefore, it has to be considered the possibility that Cd and Pb are in higher concentrations in the Mogi-Guaçu basin, which may be related to distinct sources in the basin. Alternatively, the availability of these metals may have distinct sources in the two basins. It is very well documented that heavy metals have a strong affinity for organic matter and both the formation of complexes of metals with dissolved organic matter and the adsorption on particulate organic matter may decrease the bioavailability of some metals to organisms (Förstner and Wittmann, 1983; Benjamin and Honeyman, 1992). Several authors reported these characteristics of heavy metals, and observed an increase in metal bioavailability when organic matter concentrations decreased (Fernandes et al., 1994; Salomons et al., 1995; Villar et al., 1999). Bodek et al. (1988) reported diminished absorption rates of metals in fish of the west coast the U.S. affected by sewage discharges, suggesting the effect of organic load on metal speciation. With regard to metals, Pb for example, was strongly bound to humic acids and therefore has less bioavailability for the unionid Elliptio complanata (Campbell and Evans, 1987). Furthermore, other water parameters such as pH, Ca, TSS and conductivity have a strong influence on metal availability (Van Hattum et al., 1996). The content of total organic matter, specially particulate organic matter, was higher in the Piracicaba than in the Mogi-Guaçu River basin (Table 1), but no direct correlations was observed between heavy metal concentrations in the bivalves and organic matter concentrations in water. The same was observed for the others parameters (Table 1). According to Villar et al. (1999), correlations were not observed between environmental parameters and content of metals in Limnoperna fortunei and Corbicula fluminea from the La Plata River basin. The authors attributed the differences among sampling sites to different degrees of bioavailability. In addition to the bioavailability factor, it has also be considered the possibility that metals had higher concentrations in the Mogi basin, from which could be implied that a particular source of these metals might exist in this basin, and broad indicators of human impacts, as those used in this study, were not enough to characterize the sources of metals. Žáková and Koková (1999) reported the importance of wet deposition in the load of Cd and Pb to aquatic ecosystems, and showed that a substantial part of Pb, Hg and Cd contamination in the Thaya River basin (Czech Republic) had its origin in non-point sources of pollution such as atmospheric deposition and application of mineral fertilizer containing trace elements in agriculture. A non-point source of heavy metals in the Mogi-Guaçu River basin could be the hilly headwater regions of this basin, where intensive use of agrochemicals is common. As point sources of heavy metals, paper and plastic industries are among the candidates (Cetesb, 1992).

Although Pb concentrations in animals placed in the Piracicaba and Mogi-Guaçu basins were significantly higher than the control group, it was difficult to define the degree of contamination of these basins. One alternative way to establish this degree was to compare the investigated basins with other areas in the world where similar studies were carried out (Table 3).

Obviously such comparisons could not be perfect because of the variability existing among species and due to differences between sites. However, some intra and interspecific comparisons are possible, particularly when comparisons include populations exposed to unusually high metal bioavailabilities. Such comparisons are the basis of any heavy metal biomonitoring program that would necessarily involve the use of net accumulators (Phillips and Rainbow, 1994). The average Cd concentrations in test of Piracicaba and Mogi-Guaçu basins were comparable to those found in non polluted or background areas (Table 3), which confirmed the conclusion that this metal was not available in the basins of this study. For instance, a concentration of 0.68 µg g^-1 was found in mussels of Oneida Lake, considered as a pristine environment (Secor et al., 1993). In addition, Czarnezki (1987) related background concentrations of about 0.32 µg g^-1 for Lampsilis ventricosa. Conversely, the Pb concentrations were higher in animals of the Piracicaba and Mogi-Guaçu basins than the background concentrations found in Lampsilis ventricosa (0.42 µg g^-1) and Dreissena polymorpha (0.5 µg g^-1) (Czarnezki, 1987; Kraak et al., 1991). However, Pb concentrations found in our basins were not as high as the ones found in more impacted areas such as the Thames River in England (Manly and George, 1977), and some points of the Big River in the U.S. (Czarnezki, 1987) (Table 3).