DETERMINATION OF HEAVY METAL DEPOSITION IN THE COUNTY OF OBRENOVAC ( SERBIA ) USING MOSSES AS BIOINDICATORS . I : ALUMINUM ( AL ) , ARSENIC ( AS ) , AND BORON ( B )

In the present study, the deposition of three heavy metals (Al, As and B) in the county of Obrenovac (Serbia) is determined using four moss taxa (Bryum argenteum, Bryum capillare, Brachythecium sp., and Hypnum cupressiforme) as bioindicators. Distribution of average heavy metal content in all mosses in the county of Obrenovac is presented in maps, while long-term atmospheric deposition (in the mosses Bryum argenteum and B. capillare) and short-term atmospheric deposition (in the mosses Brachythecium sp. and Hypnum cupressiforme) are discussed and given in tables. Areas of the highest contaminations are highlighted. UDC 582.32(497.11Obrenovac):581.5


INTRODUCTION
Surveillance of heavy metals in mosses was originally established in the Scandinavian countries in the 1980s.However, the idea of using mosses to measure atmospheric heavy metal deposition was developed alredy in the late 1960s (Rhülingand Tyler, 1968;Tyler, 1970).It is based on the fact that mosses, especially the carpet-forming species, obtain most of their nutrients directly from precipitation and dry deposition.In the present investigations, we decided to use two acrocarpous moss species (Bryum argenteum Hedw.and Bryum capillare Hedw.) that can give us an idea of long-term atmospheric deposition, inasmuch as they are attached to the substrate and also accumulate metals deposed during the last few decades in the surface layers of the substrate.
Two pleurocarpous taxa ( Brachythecium sp. and Hypnum cupressiforme Hedw.) were used to scan shortterm atmospheric deposition of heavy metals, considering that these taxa are not strongly attached to the substrate and accumulate mostly from precipitation.Mosses are better than other higher plants in scanning heavy metal deposition because: -they are perennial without deciduous periods; -they have high cation exchange capacity that allows them to accumulate great amounts of heavy metals between the apoplast and symplast compartments without damaging vital functions of the cells; -mosses do not possess thick and strong protective layers like cuticles.
Also, this time-integrated way of measuring patterns of heavy metal deposition from the atmosphere to terrestrial ecosystems, besides being spatially oriented, is easier and cheaper than conventional precipitation analyses, as it avoids the need for deploying large numbers of precipitation collectors.The higher trace element concentration in mosses compared to rain water makes analysis more straightforward and less prone to contamination (Berg et al. 1997).
Use of mosses to investigate heavy metal deposition shows transboundary heavy metal pollution and can indi-cate the paths by which atmospheric pollutants enter from other territories or reveal their sources within the investigated area.
Although 15 heavy metals were analyzed in all, only deposition and distribution of aluminum, arsenic and boron are treated in the present study.

MATERIAL AND METHODS
The acrocarpous mosses Bryum argenteum and Bryum capillare were used to research long-term atmospheric deposition, while the pleurocarpous Brachythecium sp. and Hypnum cupressiforme were used to scan short-term atmospheric deposition in the county of Obrenovac (Serbia).Hypnum cupressiforme is one of the standard species used in Europe to survey heavy metal deposition (Buse et al. 2003.), while the other three are standard in Europe, but do not grow in this region.In estimating which other species are eligible for use in heavy metal deposition monitoring, we relied on the experience of Thöni (1996), Herpin et al. (1994), Siewers and Hairpin (1998) Zechmeister (1994), andRoss (1990).
As far as possible, moss sampling was conducted according to guidelines, set out in the experimental protocol for a survey carried out in 2000/2001(UNECE, 2001)).Details of the procedure are given in Rühling et al. (1998).
Each sampling site was located at least 300 m from main roads and populated areas and at least 100 m from any road or single house.In forests or plantations, samples were collected in small open spaces to preclude any effect of canopy drip.Sampling and sample handling were carried out using plastic gloves and bags.About three moss samples were collected from each site.Dead material and litter were removed from the samples.Green parts of mosses were used for the analyses.
The county of Obrenovac was chosen for this investigation because of its industry and location.Each sampling site was GPS located with a precision of ±10 m, and GPS dates (Germin) were digitalized on the maps using the following softwares: OziExplorer 3.95.3b,© D&L Software Pty Ltd, and WinDig 2.5 Shareware, © D.Lovy.All material was collected during November of 2002.Not more than one site was chosen in a 50x50 m square.Seventy-five localities were chosen out of 129 for comparisons and further analyses.The selection was based on the presence of all investigated species and yearly biomass.More than 500 samples were analyzed.
After collecting, samples were dried as soon as possible in a drying oven to constant dry weight (dw) at a constant temperature of 35 0 C, then stored at -20 0 C. Following homogenization in porcelain, the samples were treated with 5+1 parts of nitric acid and perchloric acid (HNO 3 :HClO 4 = 5:1) and left for 24 hours.After that, a Kjeldatherm digesting unit was used for digestion at 150-200 0 C for about one hour.Digested samples were filtered on qualitative filter paper to dispose of silicate remains.Sample volume was then normated to 50 ml.Aluminum (Al) was detected by AAS on a Pye Unican SP9 atomic absorbance spectrophotometer from Philips using the flame of acetylene/nitrogen-suboxide.Boron (B) was detected by ICP-ES, while arsenic (As) was determined by AAS using hydride techniques.
For explanation of the results and their presentation on maps, the folloving statistical statistics parameters were used: average values, standard deviation, minimum, maximum, and percent deviation.Map making and interpolation of exact data were made using Agis software (v1.71 32 byte, © Agis Software, 2001).

RESULTS AND DISCUSSION
Aluminum in Obrenovac County probably comes from heavy road traffic, intensive coal burning, and incorrect waste disposal.The most contaminated areas according to deposition in mosses are the town of Obrenovac and the northernmost part of the county by the Sava River (>40 mg/g).A somewhat lower level of Al deposition (20-30 mg/g) occurs in the central part of the county, where Al is probably spread by traffic on the main roads.
Aluminum represents 8% of the Earth's crust.It comes from bauxite (Merian, 1984), is very widespread in all forms of industry (machine construction, electro engineering, civil engineering, the food industry, agriculture), and is also present in aluminum foil, iron plates, and other widely used articles.Aluminum emission into the atmosphere was great during the last century.Only in the 70s it was 7.2 million tons per year (Lantzyand Mackenzie, 1979).It is known that the atmospheric concentration of aluminum is significantly elevated (>10mg/m3) in areas with a cement indu-stry (Merian, 1984).
There are no data on the effects of arsenic on plants (Thöniand Seitler, 2004).For animals and man, arsenic is known to be toxic, carcinogenic, teratogenic and mutagenic.The chronic presence of arsenic pollution can lead to serious hematological and neurological diseases (Merian, 1984).Borderline values of arsenic concentration are as follows: 1 mg/m 3 in the air, 0.1 mg/l in water, and 0.05 mg/l in drinking water (Thöniand Seitler, 2004).Since all of the sampled species could not be found at each locality, average values of all specimens were extrapolated on the maps to get an idea of heavy metal deposition in the county of Obrenovac (Fig. 1).However, if one separates the values of deposition obtained from pleurocarpous mosses (Brachythecium sp. and Hypnum cupressiforme) and ones obtained from acrocarpous mosses (Bryum argenteum and Bryum capillare), it is readily discernible that the first two give us an idea of short-term and the last two of long-term deposition, respectively (Tab.1).
These differences are easily explained in terms of the life forms of these mosses and their uptake of heavy metals.Pleurocarps are not closely attached to the substrate and so get the bulk of deposited heavy metals directly from the atmosphere (during their pauciennial life period), while acrocarps are strictly attached to the substrate and receive the bulk of deposited heavy metals from the substrate solution (metals deposited over a period that is longer than their pauciennial life span).

Fig. 1 .
Fig. 1.Exact locations of moss sampling (left) and extrapolated maps (right) of selected heavy metal deposition in the county of Obrenovac (Serbia).