Effect of sea water intrusion on Hg-mobilization in coastal aquifers: the case of Monte Argentario (Italy)

A. Adorni-Braccesi*, P.F. Sciuto** R. Caboi,*** R. Cidu*** (cidur@unica.it), A. Cristini***, L. Fanfani***, L. Rundeddu***, P. Zuddas***

*Istituto di Geocronologia e Geochimica Isotopica, CNR, Ghezzano, I 56010 Italy

**Dipartimento di Scienze della Terra. , University of Siena, Siena, I 53100, Italy

*** Dipartimento di Scienze della Terra, University of Cagliari, Cagliari, I 09100, Italy

ABSTRACT

Since Monte Argentario, as other coastal areas, is subjected to an increasing demand of water, a hydrogeochemical study has been undertaken to point out possible pollution effects, due to overpumping besides salinization. A survey of springs and wells in the area reveals mercury contamination derived from dispersed cinnabar mineralizations localized in the area. The mercury dissolved content ranges from values lower than 0.5 µg/l in phreatic water to 42 µg/l in deeper waters under overpumping. This geochemical process is strictly related to sea water intrusion and occurs even when the chloride content in fresh waters does not exceed 700 mg/l.

INTRODUCTION

Everywhere in the world, a serious hazard in coastal areas is connected with seawater intrusion. This process, which is induced by indiscriminate pumping of groundwater for various purposes, determines an increase in dissolved solids, and may cause mobilization of chemical elements toxic for human health. How seawater ingression may affect the amount of toxic metals in mining districts has been already described (Cidu et al., 1998). When this happens contamination evolves extremely fast and is reversible only over a long period of time. It follows that the contamination risk of water resources must be carefully assessed with geochemical studies. Consequently modelling of the hydrogeological system is greatly recommended since it meets the increasing demand for protecting the environment and human health against “natural” disasters.

At the Monte Argentario promontory (Tuscany, Italy) a study on water quality was carried out with the financial support of the local administration (Municipality of Monte Argentario). Its aim was to recognize possible contaminated zones, and relate them to the geological and mineralogical features of the territory.

Study Area

Monte Argentario is located in southern Tuscany (in the province of Grosseto). The promontory has an area of 60 Km², with a maximum altitude of 625 m a.s.l., and is geographically isolated from the mainland by a lagoon. Its climate is typically Mediterranean (the yearly mean rainfall is 650 mm), and characterized by short, though at times intensive precipitations from October to April. All superficial and groundwater is therefore the consequence of winter recharge.

The Monte Argentario promontory is made up of four geological units (Thieye et al., 1997) as reported in Fig.1. The dominant lithology is represented by the Calcare Cavernoso (cavernous limestone), which forms the central-eastern part of the promontory. Most of the contacts between different lithologies are tectonic, and characterized by the presence of diffuse sulfide mineralizations in small and discontinuous veins containing cinnabar, pyrite, chalcopyrite, galena, and sphalerite (Tanelli, 1983), related to the intrusion of the Giglio Island granite. In the region, the presence of rising hydrothermal fluids is observed on the southern coast of Tuscany, particularly on Elba (iron ores), in Monte Amiata (mercury ores), and in northeastern Latium (antimony ores). At Monte Argentario, high-temperature Fe-Mn mineralizations are mainly located on the eastern side of the promontory, near the lagoon. Exploitation of this ore terminated in the 70’s.

Fig. 1. A schematic sketch of the area and a simplified geological map of the Monte Argentario promontory (after Thieye et al., 1997) and location of the sampling points.

No perennial streams flow in the promontory. The springs are generally located in limestone or at the contact with the lower tectonic unit; their flow is low and related to shallow and local circuits. Hypothermal water (22-25 °C) associated with deep circulation is located under the lagoon, and is used for fish-breeding.

SAMPLING AND ANALYSYS

In October 1997 (minimum groundwater level) and June 1998 (maximum groundwater level), 25 water samples from springs, lagoon, and wells were collected, assuming that one sample from each 3 Km² is enough for thematic mapping based on a Geographic and Geochemical Information System (Lampio et al., 1994).

The area distribution of the samples was chosen homogeneously on the territory so as to be representative of the different aquifers.

The water samples were filtered in situ using filters with a 0.4 µm pore-size, collected in pre-cleaned bottles, and acidified to 1 % HNO₃ for the analysis of most metals, and 0.2 % HCl for As and Sb; unfiltered samples were acidified to 0.2 % H₂SO₄ (with addition of a KMnO₄ grain) for Hg. At the sampling site water temperature, pH, redox potential, alkalinity by HCl titration, and conductivity were measured. Anions were determined by ion chromatography. Metals were determined by inductively coupled plasma optical emission spectroscopy (ICP-OES) and inductively coupled plasma mass spectrometry (ICP-MS). Mercury, antimony, and arsenic were determined by ICP-MS after Hg-vapour or hydride generation. The detection limit for chemical components was calculated at five times the standard deviation of the mean blank solution (Cidu, 1996).

The collected waters can be classified in two different types, calcium bicarbonate and alkaline chloride waters. The dissolved major element content, pH value, and conductivity in the first type of waters, generally reflect the main drained lithology (“calcare cavernoso” limestone), while the second type comes from a mixing of the previous water with seawater due to marine inflow in some wells close to the coast. Since the mixing with marine water occurs in different proportions, the passage from the first to the second type is gradual. Some waters sampled close to the mineralizations reveal a significant sulphate content.

MERCURY CONTAMINATION

Presence of mercury at detectable levels (>0.5 µg/l) was observed only in a few samples during the first campaign, but a municipal well (black dot outlined with a white square in Fig.1) located close to Porto S. Stefano revealed a total Hg dissolved content of 42 µg/l. After this the Monte Argentario Municipality excluded the well from the water supply circuit and started a monitoring programme on the water quality to detect the evolution of the contamination. The simultaneous variation of Hg and Cl abundance vs. time in the municipal well is reported in Fig. 2. When pumping was interrupted, the mercury content decreased sensibly to about 4 µg/l. When reactivated, Hg abundance rapidly returned to the high values observed previously. It appears therefore evident that saline intrusion/regression induced a progressive increase/decrease in dissolved mercury. A close relation between marine intrusion and Hg dissolution in mining areas has been reported (Cidu et al., 2000).

Fig. 2: Chemical evolution of the municipal well during survey after interruption of pumping, and after pumping reactivation.

The relative abundance of inorganic Hg dissolved species was evaluated using the SOLMIN88 package (Kharaka et al., 1989). The results reveal the presence of mercury prevalently as chloride complexes (HgCl₂⁰, HgCl₃^-, HgCl₄^2-) and Hg(OH)₂⁰. At low salinity (up to 100 mg/l in chloride, with a median concentration of Hg of 0.8 mg/l, practically the potability limit according to European Union standards) Hg(OH)₂⁰ represents the main aqueous species in water. In the chloride range of 100 to 200 mg/l, the abundance of Hg(OH)₂⁰ decreases, and HgCl₂⁰ becomes the predominant species; starting from 200 mg/l, most of the mercury is as HgCl₃^-; over a 1000 mg/l chloride abundance, HgCl₄^2- is significantly present. Hg⁺⁺, HgCl⁺, HgSO₄⁰, and Hg(OH)⁺ are negligible in the system, and decrease systematically as the mixing with seawater increases. Calculations are in agreement with data about the mercury-chloride complex formation (Leckie et al., 1974).

In the following months, an anomalous presence of mercury was also observed in other localities situated along the coast to the north of Grosseto (Follonica, Castiglione della Pescaia), in connection with an increased exploitation of the water resources. This is not surprising, since the province of Grosseto was an important historical mine district rich in sulphide ores, and the Monte Amiata mines, which were dismantled in 1982, represented one of the largest cinnabar mining areas in the world for half a century.

CONCLUSIONS

The chemical and physical data show that groundwaters pumped from aquifers are in hydraulic connection with the sea. The hydraulic gradient created by a seasonal overpumping induces an inflow of marine water into fresh water aquifers, as has clearly been observed in one of the municipal wells. Though the chloride content does not exceed 700 mg/l, the partition of elements between solid phases and water is altered by the stability of dissolved chloride complexes, leading to the release of mercury into solution. The process terminates when the original level of the water table is restored. In this context the behavior of mercury appears to be sensible to the increase in chloride content even at the initial stage of saline intrusion.

Acknowledgements. The present investigation has been carried out with the financial support of the Monte Argentario Administration, MURST and CNR. The authors are grateful to the State Forest Corps and to the Italian Air Force in Monte Argentario for logistic support.

REFERENCES

Cidu R. (1996), Atomic Spectroscopy 17: 155-162.

Cidu R., Fanfani L. (1998), In Proc. WRI-9, Balkema, Rotterdam, pp. 969-972.

Cidu R., Biagini C., Fanfani L., La Ruffa G., Marras I. (2000), Appl.Geoch. (in press).

Kharaka Y.K., Gunter W.D., Aggarwal P.K., Perkins E.H., DeBraal J.D. (1989), U.S.G.S. Water-Resources Investigations Report 88-4227, 157 pp.

Lampio E., Gustavsson N., Salminen R., Tarvainen T. (1994), In: Proc.1^st Eur. Congr.Reg. Geol. Cartogr. Inf. Syst., Bologna, pp. 213-217.

Leckie J.O., James R.O. (1974), Aqueous-Environmental Chemistry of metals, Ann Arbor Science Publishers, 127 pp.

Tanelli G. (1983), Mem. Soc. Geol. It. 25: 91-109.

Thieye T., Reinhardt J., Goffé B., Jolivet L., Brunet C. (1997), Eur. J. Mineral. 9: 859-873.