The Role of Sedimentary Organic Carbon and Nitrogen in Mercury Cycling in the Gulf of Trieste (Northern Adriatic Sea)

Nives Ogrinc¹, Milena Horvat¹, Stefano Covelli², Jadran Faganeli³ and Vesna Fajon¹

¹“J. Stefan” Institute, Dept. Environ. Sciences, Jamova 39, 1000 Ljubljana, Slovenia

²Dept. Geol., Environ. & Mar. Sci.,University of Trieste, Via E. Weiss 2, 34127 Trieste, Italy

³ Marine Biological Station, Fornače 41, 6330 Piran, Slovenia

Abstract – Biogeochemical processes involving mercury compounds in aquatic environments are of great importance especially in areas where mercury inputs are particularly high, such as in the case of the Gulf of Trieste. In this study the stable isotope data are used in order to determine the role of sedimentary organic carbon and nitrogen in the transformation processes of the Hg cycle. The results show that the major source of inorganic mercury is the river Isonzo, from which mercury is probably transported into the Gulf adsorbed to particles, while the major source of methylmercury is the bottom sediment of the Gulf itself. Terrigenous nitrogen seems to be an important indicator of transformation processes of methylmercury in the sediments. This observation should be further investigated.

1. Introduction

During the last two decades, the global biogeochemical cycles of many elements have been investigated in considerable detail. Much effort has been invested into the mercury cycle because of the toxicity of methylmercury, the accumulation of methylmercury in biota, and its biomagnification in the aquatic food chain. Mercury toxicity often poses a serious threat to wildlife and the human population. The interest in this cycle and others, like those of carbon, and nitrogen, has been heightened by problems associated with global, regional and local environmental changes. In addition, different element cycles are often interconnected with each other requiring a more holistic research approach to fully understand the underlying mechanistic processes.

It is well known that the Gulf of Trieste is one of the most Hg-contaminated areas in the Mediterranean Sea, and is therefore suitable for studying the processes affecting the Hg cycle in the marine environment. The Gulf is characterized by high Hg inputs from the river Isonzo that carries the waste cinnabar deposits of the Idrija mercury mine. Even ten years after the closure of the mine, the mercury levels in seafood still exceed the value of 0,5 mg/kg, which is set as the maximum permissible according to the WHO. Recent estimates of the Hg balance in the Gulf of Trieste have shown that the annual input through the Isonzo discharge is about one ton and a half (Sirca et al., 1999). This ecosystem is also subject to eutrofication, anoxic conditions at the bottom, and winter and summer temperature stratification (Faganeli et al., 1991). Such conditions generally favor the transformation of inorganic mercury to more toxic methylmercury, leading to elevated Hg values in marine organisms.

In the present study stable isotope data of sedimentary organic carbon and nitrogen are used in order to determine the processes influencing the Hg cycle and transport of sediment moving from the Isonzo river into the central part of the Gulf of Trieste.

2. Methods

Surface sediment samples from the Gulf of Trieste were freeze-dried and ground to a fine powder for analysis. In two locations, the analyses were also performed in different sediment fractions.

Total Hg and Methylmercury (MeHg) analyses – Total Hg in sediment was analyzed by CV AAS after acid digestion (Horvat et al., 1996). MeHg in sediments was determined by solvent extraction, aqueous phase ethylation, gas chromatographic separation, pyrolysis and CV AFS detection (Horvat et al., 1993). All analytical methods were regularly validated and performed within a quality control system.

Stable isotope analysis - The isotopic analysis of particulate organic carbon (POC) in sediment was determined after treated with 3M HCl, while for nitrogen (PON) bulk powdered samples were analysed. d¹³C and d¹⁵N were determined after Dumas combustion of samples at 1800°C using an Europa 20-20 continuous-flow stable isotope analyser with ANCA-SL preparation module. Following the standard procedure, the isotopic ratios are expressed in d-notation in parts per mil (‰):

For carbon, *X/X is ¹³C/¹²C and the standard is the V-PDB carbonate, while for nitrogen *X/X is ¹⁵N/¹⁴N and the standard is atmospheric (air) nitrogen. The measurement uncertainties on the d¹³C values were ±0.2‰, and ±0.3‰ on the d¹⁵N values.

Analyses of total and organic C and total N in sediment samples were performed using a Carlo Erba elemental analyzer (mod. EA 1108) after acidification of the samples with 1N HCl (Hedges and Stern 1984). The precision of measurements was ±3%.

3. Results AND DISCUSSION

d¹³C and d¹⁵N values were used to study the origin of sedimentary organic matter in the Gulf of Trieste. Using the binary mixing equation for d¹³C and d¹⁵N values of terrigenous (d¹³C_t= -28‰, d¹⁵N_t= +1.7‰) and marine organic matter (d¹³C_m= -21‰, d¹⁵N_m= +7.1‰ - average value determined in phytoplankton) (Thornton and McManus, 1994) the percentage of the terrigenous fraction of POC and SON in surficial sediments was established. Sediments at the mouth of the river Isonzo are calculated to contain about 63% of POC and 84% of terrigenous nitrogen material, while in the central part of the Gulf of Trieste they are assumed to contain less terrigenous material (8% of POC and 28% of PON). The deposition of almost entirely marine (phytoplanktonic and microphytobenthic) organic matter is restricted to a rather narrow strip in the central part of the Gulf located between two areas affected by the increasing contribution of terrigenous organic matter.

Results for total Hg in surface sediments show that the highest concentrations are found at the river inflow, ranging from 23.3 mg/g up to 40 mg/g. The concentration gradient from the river mouth towards the SE near-shore area of the Gulf shows over a hundredfold decrease of Hg, reflecting the decline in riverine particulate matter sedimentation rates with distance from the river mouth. The ratio of MeHg to total Hg in sediments increases with distance from the river mouth, and is in positive correlation with the percentage of the clayey fraction.

Figure 1. a) Correlation between total Hg and calculated terrigenous organic carbon and b) correlation between MeHg and calculated terrigenous nitrogen in the surficial sediment in the Gulf of Trieste. In the bracket is the point at the river Isonzo mouth.

As shown in Figure 1a, total Hg is in good correlation with the percentage of terrigenous organic carbon, confirming a strong association with riverine particulate matter (Horvat et al., 1999). On the other hand, no correlation was found for MeHg vs. terrigenous organic C, however a positive correlation was observed between MeHg and organic carbon of marine origin. This confirms our previous conclusion that the river input is a minor source of MeHg in the Gulf of Trieste (Covelli et al., 1999; Sirca et al., 1999), but rather the MeHg is found in the sediment of the Gulf. Interestingly, the correlation between terrigenous N and MeHg was found to be significant (Figure 1b), while no correlation was observed for total Hg vs. terrigenous N. We are not in a position to explain this observation, however the terrigenous nitrogen seems to influence transformation processes of mercury in marine sediment. Further studies will address this issue in more details.

Analyses of different sediment fractions show that larger particles (>50mm) characterized by low organic C and total N contents and low d¹³C values, are also deposited at the month of the river Isonzo. In these fractions the total Hg concentrations are the highest. Therefore the large-size fractions play an important role in mercury transport probably adsorbed to particles (Vasiliev et al., 1996), and also originating from cinnabar from Idrija. Particles deposited in the central part of the Gulf of Trieste contain higher C and N values and higher d¹³C values. This indicates that a minor part of the riverine organic matter, associated with silt, is transported throughout the Gulf and can be used as a tracer of terrigenous organic deposition. The lowest C and N isotopic compositions in the fraction 16-2mm indicates the content of more terrigenous material. In comparison with our previous studies, a negative correlation between MeHg vs. % of sand and a positive correlation with the % of clay (<2mm) was again found (Horvat et al., 1999; Sirca et al., 1999).

In conclusion, a combination of stable isotope data of sedimentary organic carbon and nitrogen with total Hg data suggests a strong association of total Hg with particulate terrigenous carbon, while MeHg is less strongly associated with particles than the inorganic Hg, and rather originates from the bottom sediment. Further studies are in progress in order to understand the role of terrigenous nitrogen in transformation processes of MeHg in marine sediments.

References

Covelli S, Faganeli J, Horvat M, Brambati A (1999) Esturine, Coastal Shelf Sci 48: 415-428.

Faganeli J, Pezdic J, Ogorelec B, Herndl GJ, Dolenec T (1991) In: Modern and Ancient Continental Shelf Anoxia (RV Tyson, TH Pearson, Editors) The Geological Society, London, pp. 107-117.

Hedges JI, Stern JH (1984) Limnol. Oceanogr. 29: 45-57.

Horvat M. (1996) In: Global and regional mercury cycles: sources, fluxes and mass balance (W Baeyens, R Ebinghaus, O Vasiliev, Editors), Netherlands, Kluwer Academic Publishers, pp. 1-31.

Horvat M, Liang L, Bloom NS (1993) Anal. Chim. Acta 281(1): 135-152.

Horvat M, Covelli S, Faganeli J, Logar M, Mandic V, Rajar R, Sirca A, Zagar D (1999) Sci. Total Environ. 237/238: 43-56.

Širca A., Rajar R, Harris RC, Horvat M (1999) Environ. Modelling and Software 14: 645- 655.

Thornton SF, McManus J (1994) Estuarine, Coastal and Shelf Sci. 38: 219-233.

Vasiliev OF, Papina TS, Eyrikh SS, Sukhenko SA (1996) In: Global and regional mercury cycles: sources, fluxes and mass balance (W Baeyens, R Ebinghaus, O Vasiliev, Editors), Netherlands, Kluwer Academic Publishers, pp.273-285.