Multivariate extraction of dominant geochemical ...

Multivariate extraction of dominant geochemical markers for deposition of 69 elements in the Bregalnica River basin, Republic of Macedonia (moss biomonitoring) Biljana Balabanova, Trajče Stafilov, Robert Šajn & Claudiu Tănăselia

Environmental Science and Pollution Research ISSN 0944-1344 Environ Sci Pollut Res DOI 10.1007/s11356-016-7502-7

1 23

Your article is protected by copyright and all rights are held exclusively by SpringerVerlag Berlin Heidelberg. This e-offprint is for personal use only and shall not be selfarchived in electronic repositories. If you wish to self-archive your article, please use the accepted manuscript version for posting on your own website. You may further deposit the accepted manuscript version in any repository, provided it is only made publicly available 12 months after official publication or later and provided acknowledgement is given to the original source of publication and a link is inserted to the published article on Springer's website. The link must be accompanied by the following text: "The final publication is available at link.springer.com”.

1 23

Author's personal copy Environ Sci Pollut Res DOI 10.1007/s11356-016-7502-7

RESEARCH ARTICLE

Multivariate extraction of dominant geochemical markers for deposition of 69 elements in the Bregalnica River basin, Republic of Macedonia (moss biomonitoring) Biljana Balabanova 1 & Trajče Stafilov 2 & Robert Šajn 3 & Claudiu Tănăselia 4

Received: 9 February 2016 / Accepted: 22 August 2016 # Springer-Verlag Berlin Heidelberg 2016

Abstract Atmospheric deposition was investigated using the terrestrial moss species Hypnum cupressiforme (Hedw.) and Homolothecium lutescens (Hedw.) in the Bregalnica River basin, Republic of Macedonia. Long-term emission occurs in this area due to the hydrothermal exploitation of Pb–Zn deposits (Sasa and Zletovo mines) and copper ore exploitation and floatation (Bučim mine). Determination of the chemical elements was conducted using atomic emission spectrometry with inductively coupled plasma (ICP-AES) and mass spectrometry with inductively coupled plasma (ICP-MS). A combination of multivariate techniques (PCA, FA and CA) was applied for data processing and identification of element association with lithogenic/anthropogenic origin. Seven dominant factors were extracted from the total of 69 analysed elements. Spatial distribution maps were constructed for the determination and localisation of smaller areas with higher contents of certain anthropogenic elements. In this way, the influences of Responsible editor: Céline Guéguen Electronic supplementary material The online version of this article (doi:10.1007/s11356-016-7502-7) contains supplementary material, which is available to authorized users. * Trajče Stafilov [email protected]

1

Faculty of Agriculture, University “Goce Delčev”, Krste Misirkov bb, Štip, Republic of Macedonia

2

Institute of Chemistry, Faculty of Natural Sciences and Mathematics, Ss. Cyril and Methodius University, POB 162, 1000 Skopje, Republic of Macedonia

3

Geological Survey of Slovenia, Dimičeva ulica 14, 1000 Ljubljana, Slovenia

4

INCDO-INOE 2000 Research Institute for Analytical Instrumentation (ICIA), Cluj-Napoca, Romania

selected human activities on local air pollution can be determined. The summarised data show quantification of the element distributions. This not only allows the determination of the distribution of hazardous elements but also presents complete characterisation of element deposition in the environs of mines. Keywords Air pollution . Moss . Biomonitoring . Atomic emission spectrometry with inductively coupled plasma . Mass spectrometry with inductively coupled plasma . Multivariate assessment . Spatial distribution . Bregalnica River basin

Introduction Environmental pollution at hazardous levels for living organisms presents a global problem and a challenge for a macrocase monitoring. Some sub-disciplines have been developed over time in order to consider realistic environmental conditions. Arguably, the understanding of atmospheric pollution is one of the most emergent areas of environmental science (Fernández et al., 2015). Atmospheric pollution represents solutions or suspensions of minute amounts of harmful compounds in the air (Vallero 2014). The degree and extent of environmental changes over the last decades have given a new urgency and relevance to the detection and understanding of environmental changes due to human activities, which have altered global biogeochemical cycling of heavy metals and other pollutants (Athar and Vohora 1995; Acton 2013). Monitoring of toxic air pollutant is needed in order to understand their spatial and temporal distribution and ultimately to minimise their harmful effects. In addition, to direct physical and chemical methods of air pollution monitoring, bioindication has also been used to evaluate air pollution risk (Aboal et al. 2010; Ares et al. 2012; Vallero, 2014).

Author's personal copy Environ Sci Pollut Res

Mosses have frequently been used to monitor timeintegrated bulk deposition of metals as a combination of wet, cloud and dry deposition, thus eliminating some of the complications of precipitation analysis caused by the heterogeneity of precipitation (Harmens et al. 2004, 2010, 2015). Ectohydric mosses in particular draw negligible amounts of water and minerals from the soil, and almost entirely depend on the atmospheric nutrient inputs (Rühiling and Tyler, 1968). Because mosses have a high cation exchange capacity (CEC), they act as hyper-accumulators of metals and metal complexes. The metals are bound to the tissue with minimal translocation within the plant due to a lack of vascular tissue (Tyler 1990). This results in biological tissue that can be analysed to reveal time-integrated deposition (Zechmeister 1995). Additional advantages of using mosses as heavy metal biomonitors include their stationary nature, widespread geographical distribution and low genetic variability between populations. It has been shown that there is some experimental error due to heterogeneity in morphological characteristics and microenvironments among different populations (Zechmeister et al. 2003). There is also an incomplete understanding of the degree of mineral uptake by ectohydric mosses in direct contact with substrate (Gjengedal and Steinnes 1990). Despite the accuracy and precision of precipitation analysis techniques, however, mosses offer an efficient, lowcost complement for determining metal concentrations at a large number of locations and offer analyses of biologically (Bourennane et al, 2010) relevant fluxes at multiple scales. Harmens and his European colleagues have found that mosses are reliable indicators of air pollution risks to ecosystems, because they get most of their nutrients direct from the air and rain, rather than the soil (Harmens et al, 2008). Since 2000, the European moss survey has been conducted by a special international programme (ICP Vegetation). Moss data provides better geographical coverage than measured deposition data and reveals more about actual atmospheric pollution at a local level (http://icpvegetation.ceh.ac.uk/). The latest data reported by Harmens et al. (2015) and Barandovski et al. (2015) indicates the significant enrichment of some toxic elements. Because of the dominance of volcanic geological units and Pb–Zn mineral deposits, the region of the Bregalnica River basin in the Republic of Macedonia has specific characteristics related to the monitor of the atmospheric distribution of various chemical elements (Balabanova et al. 2010, 2012, 2015; Barandovski et al. 2008, 2015; Stafilov 2014). This region is characterised by several significant pollution sources that emit potentially toxic metals and other chemical elements into the environment: the “Bučim” copper mine and floatation plant near the town of Radoviš and lead and zinc mines “Sasa” near the town of Makedonska Kamenica and “Zletovo” near the town of Probištip (Stafilov 2014). The excavation of copper minerals is carried out from an open ore pit, while in the lead–zinc mines, the exploitation is underground, and the ore

tailings are stored outdoors. The ore produced in the mines is processed in the floatation plants, and in the process of floatation of the relevant minerals, flotation tailings are separated and disposed on a dump site in the open. The study presents multivariate assessment and spatial analysis of dominant element associations in the Bregalnica River basin and is the first such analysis in this unique lithological area. The appearance of dominant Oligocene volcanism in the area of the Sasa mine (Pb–Zn hydrothermal exploitation), the Kratovo-Zletovo district and the Bučim mine area (Cu–Au hydrothermal exploitation) creates specific environmental conditions leading to natural and anthropogenic poly-metallic enrichment. Therefore, the main objective of this study was to assess the lithologenic and anthropogenic distribution of a total of 69 elements along the Bregalnica River basin. A multivariate statistical approach was used for identification of the dominant geochemical associations. Moss species were used to record the lithological and anthropological impact on the atmospheric deposition of potentially toxic and nontoxic elements. The dominant lithogenic and anthropogenic markers were extracted. The spatial distribution patterns of each geochemical marker were also generated.

Investigated area The investigated area includes the basin of the river Bregalnica, which is found in the area of the eastern part of the Republic of Macedonia. The area of the Bregalnica River basin covers ∼200 km (W–E) × 200 km (S–N), within the following geographic coordinates N 41° 27′ 23 –42° 09′ 21 and E 21° 53′ 04 –23° 02′ 21 (Fig. 1). The investigated region of the investigated area is geographically composed of several sub-regions. These sub-regions are characterised by their different geology and land use. As a consequence of industrialisation, urbanisation and lack of treatment of wastewaters from the industry, mines and city sewerage, the waters of this important hydrographic region are exposed to a high level of pollution, especially from the aspect of introduction of higher amounts of certain toxic metals. The pollution most frequently reaches particularly high levels at low river flows, as in the case of the river Bregalnica in its course through the Kočani valley (in the central part of the investigated area). The eastern and northern parts of the investigated area are mainly mountainous while the central and western parts are cultivated agricultural land (Fig. 2). Precipitation in the areas is mostly related to and conditioned by the Mediterranean cyclones (Lazeravski 1993). During the summer period, the region is most often found in the centre of the subtropical anticyclone, which causes warm and dry summers. The central area of the region is driest and the average annual precipitation increases in all directions outwards, because of increases either in the influence of the

Author's personal copy Environ Sci Pollut Res Fig. 1 The investigated area on the territory of the Republic of Macedonia

Mediterranean climate or in altitude. About ten climaticvegetation soil areas in the region are distinguished with considerably heterogeneous climate, soil and vegetation characteristics (Lazarevski 1993). Generalised geology and dominant ore mineralisation of the investigated area SE Europe underwent a complex Alpine tectonic evolution. The Srednogorie zone, Kraishte and the Rhodope Massif in Serbia, Macedonia and Bulgaria were sites of extensive westward/southward younger magmatic activity, starting in Jurassic to Cretaceous times with the closure of the Vardar zone (Barcikowski et al. 2003). The investigated area that covers the basin of the river Bregalnica lies on the two main tectonic units—the Serbian–Macedonian massive and the Vardar zone (Dumurdzanov et al. 2004). The polyphasal Neogene deformations, through the insignificant movements associated with the volcanic activities, had a direct influence on the gradual formation of the reefs and the formation of

deposits in the existing basins. From the middle Miocene to the end of the Pleistocene, there were alternating periods of fast and slow landslide accompanied by variable sedimentation (deposition). The Cenozoic volcanism represents a more recent extension into the Serbian–Macedonian massive and the Vardar zone. The oldest volcanic rocks occur in the areas of Bučim, Damjan, the Borov Dol district and in the zone of Toranica, Sasa, Delčevo and Pehčevo (Dumurdzanov et al. 2004). These older volcanic rocks were formed in the mid Miocene from sedimentary rocks that represent the upper age limit of the rocks. The origin of these oldest volcanic rocks is related to the Oligocene–early Miocene period (Dumurdzanov et al, 2005). As volcanic rocks, they are categorised as follows: andesite, latite, quartz–latite and dacite. Volcanism appears sequentially and in several phases forming sub-volcanic areas. On the other hand, the pyroclastites are most frequently found in the Kratovo-Zletovo volcanic area, where dacites and andesites are the oldest formations. The generalised geology of the area is reproduced in Fig. 3, based on the data provided by

Author's personal copy Environ Sci Pollut Res Fig. 2 The region of the investigated area including the type of land use

Rakićević et al. (1968). The most important Macedonian metal deposits are related to regional magmatic activity that occurred in the southern parts of the Carpato-Balkanides during the Eocene–Pliocene (Serafimovski et al. 2006a, b, 2007). The Zletovo Mine is located near the town of Probištip. As reported by Serafimovski et al. (2006a), the mineralisation of this deposit is related to Tertiary calc-alkaline magmatic rocks (dacites and andesites) and it is found in a dacitic volcanosedimentary suite that has been altered to clays and micas. The Sasa mine is situated near the city of Makedonska Kamenica. The mineralisation is located along the contact between Miocene calc-alkaline igneous bodies (latites and dacites) and graphite–chlorite–sericite shists, gneisses and limestones (Cassard et al. 2012). The ore consists of pyrite, galena and sphalerite with additional magnetite and chalcopyrite, while the mass fraction of Pb and Zn is around 10 % with additional contents of Ag, As, Cd, Mn and Sb. The ore is concentrated at the mine by floatation and tailings are stored in a dam in a valley just below the mine. The Kaminicka River is culverted beneath the tailings dam and flows 12 km until it meets in the Kalimanci Lake (an artificial reservoir on the Bregalnica River). The Bučim mine is situated near the town of Radoviš. The copper–porphyry type of mineralisation occurs in this deposit (Serafimovski et al. 2010; Lehmann et al. 2013). The ore concentrate is characterised by an average content of 0.3 % Cu and 0.3 g/t Au (Alderton et al. 2005).

Drainage waters from milling and flotation are discharged into the Lakavica River.

Materials and methods Moss sampling The carpet-forming moss species Pleurozium schreberi and Hylocomium splendens are usually preferred for moss trace element deposition monitoring (Fernández et al. 2015). However, these species are commonly found only in some parts of the country at an elevation above 1000 m (Cekova, 2005) and the altitude of the investigated area varies between 290 and 1932 m. Therefore, the samples of the pleurocarpous moss species Homalotecium lutescens and Hypnum cupressiforme were also collected in the investigated area in the period of August–September 2012. Researchers setting up a large-scale survey often face the problem that the location of the predicted sampling spot becomes subordinate to the presence/absence of the selected species (Fernández et al. 2015). This problem can be overcome by using more than one moss species within the same survey; however, it is clear that the concentrations of elements may vary considerably between species, thus precluding comparison of the results obtained (Boquete et al. 2013). Carballeira et al. (2008)

Author's personal copy Environ Sci Pollut Res Fig. 3 Generalised geology of the investigated area

suggested that when the regression line slope and elevation do not differ significantly from the line of slope 1 (in covariance analysis), the species could be combined. Considering this, four species were used to conduct the monitoring deposition analysis on a local scale. Depending on the conditions and the accessibility of the locations, species available and typical for the region were collected. Random samples (in the very close vicinity of the pollution source) and 149 samples according to a sampling network (5 × 5 km) were taken, as presented in Fig. 4. A detailed description of the samples (according to officially accepted techniques) is given by Fernández et al. (2015). Sample preparation and spectroscopy analysis For the digestion of moss samples, a microwave digestion system (CEM, model Mars) was applied. Precisely weighed masses (0.5000 g) of moss samples were placed in Teflon vessels and 5 mL concentrated HNO3 (trace pure) and 2 mL H2O2 (30 %, m/v) were added. The Teflon vessels were

carefully closed and the microwave digestion method was applied. The digestion method was performed in two steps to totally dissolve the moss tissue as previously described by Balabanova et al. (2010) and Bačeva et al. (2012). When the digestion method was finished, the digests were quantitatively transferred into 25-mL volumetric flaks. The prepared digests from moss tissue were then analysed for their total element contents. A SCIEX Perkin Elmer Elan DRC II (Canada) inductively coupled plasma mass spectrometer (with quadrupole as single detector) was used for measurement of the trace element concentrations. All measurements were performed using the semiquantitative method (Total Quant) supplied by the Elan 3.4 software. This method uses a response factor calibration curve which was obtained by calibration in multiple points of low, medium and high concentration, in the optimum setup, using a multielement Merck VI standard solution diluted to mimic real sample consumption. An atomic emission spectrometer with inductively coupled plasma, ICP-AES (model 715ES, Varian, USA), was used to measure the concentration of


Fig. 4 Digital elevation model with defined zones and location of sampling points

several macro-elements (Al, Ca, Fe, K, Mg, Na and P). For each analysed element, prior optimisation of the instrumental conditions was performed. The instrumental and operating conditions for each of the above-mentioned techniques are given in Table 1. Both certified reference materials M2 and M3 (Steinnes et al. 1997) and spiked intra-laboratory samples were analysed at a combined frequency of 20 % of the samples. The recovery for all of the analysed elements ranged from 76.8 % for Tl to 119 % for Sb (for ICP-MS measurements) and from 87.5 % for Na to 112 % for P (for ICP-AES measurements). Data analysis The obtained values for the contents of the investigated elements were statistically processed using basic descriptive statistics. Bivariate statistics were applied to investigate correlation between element contents. For that purpose, the linear coefficient of correlation was used. Two-dimensional scatter plots were used to visualise relations between two data sets. Individual data points were represented by point markers in two-dimensional space, where axes represent the variables. Multivariate statistical methods (cluster and R-mode factor

analyses) were used to reveal the associations between the chemical elements. Factor analysis was performed on variables standardised to zero mean and unit standard deviation (Filzmoser et al. 2005; Žibret and Šajn 2010). As a measure of similarity between variables, the product–moment correlation coefficient (r) was applied. Various rotational strategies have been proposed (Reimann et al. 2002; Žibret and Šajn 2010), the goal of all of them being to obtain a clear pattern of loadings, so that factors are somehow clearly marked by high loadings for some variables and low loadings for others. Elements with low communalities were excluded because of their lack of significant associations. In this study, the varimax method was used for orthogonal rotation. The aim was to find a rotation that maximises the variance on the new axes, put another way, to obtain a pattern of loadings on each factor that is as diverse as possible, lending itself to easier interpretation. The statistical softer Statistica 11.0 was used for data processing of the values obtained for element contents. The universal method kriging with linear variogram interpolation (Davis 1986) was applied for construction of the areal distribution maps of the 31 particular elements and the factor scores (F1–F7) in topsoil (0–5 cm) and subsoil (20–30 cm). The basic grid cell size for interpolation was 20 × 20 m. For

Author's personal copy Environ Sci Pollut Res Table 1

Instrumental conditions for ICP-MS and ICP-AES

RF generator

ICP-AES

Power output of RF generator Power output stability

1500 W Better than 0.1 %

ICP Ar flow gas rate Plasma parameters

15 L min−1

ICP-MS

Nebuliser

V-groove

Spray chamber Peristaltic pump

Double-pass cyclone 0–50 rpm

Cones Plasma configuration

N.A. Radially viewed

Platinum Axially viewed

Spectrometer

Echelle optical design

Quadrupole

Polychromator Polychromator purge Total voltage (V) Integration measurement time (ms) Measurement at 1 point (isotope) (s) Repetition measurement

400 mm focal length 0.5 L min−1 N.A. N.A.

N.A. N.A.

N.A.

300

Conditions for program ICP-AES measurements Element Wavelength (nm) Al 396.152 Ca 370.602 Fe 238.204 Mg K Na

280.270 766.491 589.592

P

213.618

Micromist

0.1 0.1

3 per point ICP-MS measurements Isotopes 107Ag, 75As, 27Al, 197Au, 11B, 137Ba, 9Be, 209Bi, 79Br, 114Cd, 140Ce, 59Co, 53Cr, 133Cs, 63Cu, 163Dy, 166Er, 153Eu, 56/57Fe, 69Ga, 157Gd, 72Ge, 178Hf, 201/202Hg, 165Ho, 127I, 115In, 193Ir, 139La, 7Li, 175Lu, 55Mn, 95Mo, 93Nb, 146Nd, 60Ni, 189Os, 206/207/208Pb, 105Pd, 141Pr, 195Pt, 85Rb, 185Re, 103Rh, 101Ru, 121Sb, 45Sc, 77Se, 147Sm, 120Sn, 88Sr, 181Ta, 159Tb, 125Te, 47Ti, 232Th, 205Tl, 169Tm, 51V, 182W, 89Y, 172Yb, 66Zn, 90Zr

N.A. not applicable

class limits, the percentile values of distribution of the interpolated values were chosen. Seven classes of the following percentile values were selected: 0–10, 10–25, 25–40, 40–60, 60–75, 75–90 and 90–100.

Results and discussion Characterisation of moss accumulation of the selected elements The element content distribution was assessed in the investigated area and characterised using values obtained from basic descriptive statistics (Table 2). The values for the element contents were also characterised for the dominant geological units in the investigated area (Table 3). The distribution of the macro-elements Al, Ca, Fe, K, Mg and P was in the range typical for this kind of sample. High matrix contents originating from plant tissues can sometimes increase the values

compared with the atmospheric distribution of these elements. The determined median values for Al, Ca, Fe, K, Mg and P were 0.27, 0.94, 0.26, 0.22, 0.17 and 0.14 %, respectively (Table 2). These data were not significantly different to those published by Barandovski et al. (2015) and Harmens et al. (2015) for the whole Macedonia and for Europe. The distribution of the rest of the elements varied due to certain natural phenomena (geology of the area, winds, rain, anthropogenic activities, etc.). Silver content was determined in the range of 0.005–1.1 mg kg−1, with a median of 0.038 mg kg−1 (Table 2). Native silver is sometimes associated with sulphide ores as a result of their chemical reduction. In areas containing copper sulphides, where hydrothermal exploitation occurs in the area of Proterosoic shales, the obtained median value for Ag was 0.051 mg kg−1. The dominant geological unit with the highest median value for Ag moss samples was Quaternary terraces (Table 3). Cu was distributed very similarly to Ag also being enriched in the Quaternary terraces (median value 6.7 mg kg−1, Table 3). These metals, collectively known as

mg/kg

La

1.7

0.42

1.0

67

CV mg/kg

0.19 −0.35 Li

2.8

0.16

3.1 0.006

0.036 1.1 3.5 5.2

−0.01 −0.47 V −0.09 −0.40 W −0.14 2.70 Y −0.05 −0.32 Yb

87

0.03 −0.42

37 −0.08 −0.38 Zn 34 0.45 3.45 Zr

72 49 90 69

0.052 130 0.03 −0.49 Sm 1.8 270 0.13 0.21 Sn 0.027 70 −0.02 −0.38 Sr 0.013 70 −0.03 −0.31 Tb 5.9 68 −0.04 −0.60 Ti 0.016 65 0.18 −0.56 Tl 0.060 66 0.01 −0.36 Tm

5.5 18 1.2

58 1.8 4.0 0.14 5.4 5.4 19 1.2

1.9 5.0 0.55

mg/kg mg/kg

mg/kg μg/kg mg/kg mg/kg

16 1.1

5.9 62 1.5 0.16

16 1.0

6.0 63 1.5 0.17

6.6 0.037

240 13

1.8 32 5.0 1300 0.35 7.6 0.033 0.77

2.0 1.8 110 310 570 280 130

47 96 6.5

59 28 2300 1.8 0.44 10 4.1 1.6 45 0.14 0.041 0.29 4.9 2.0 1400

mg/kg 0.37 0.37 0.050 mg/kg 0.20 0.19 0.025 mg/kg 27 26 9.4 μg/kg 66 68 15 mg/kg 120 130 32 μg/kg 39 36 5.0 μg/kg 28 28 5.0

mg/kg μg/kg mg/kg

75 64 91

0.05 −0.49 Rb 0.03 −0.70 Sb 0.04 −0.08 Sc

0.29 0.067 0.44

20

Max

810 4.2

0.70

Min

0.03 −0.20 Mn mg/kg 150 150 32 0.07 1.16 Mo mg/kg 0.22 0.22 0.063 mg/kg mg/kg mg/kg % mg/kg

67 71

6.2 65 −0.04 −0.43 Na 14 270 −0.16 1.68 Nd 0.20 110 0.02 −0.38 Ni 0.026 33 0.41 0.74 P 25 190 0.09 −0.14 Pb

0.82 3.4

26 0.17

2.6

Mеd

120 0.42

25 0.16

2.5

Mean

5.0 0.058

1.51 Lu μg/kg 0.46 Mg %

Unit

E

A

Med median, Min minimum, Max maximum, S standard deviation, CV coefficient of variation, A skewness, E kurtosis

1.7

14

260 75 130 0.66 0.18 0.25

92 91 38 0.22 0.22 0.067

μg/kg %

0.29 0.76 18 34 27 49 47 120

I K

5.9 1.6 6.4

2.6 91 340 340

2.3 0.65 2.5

5.1 0.16 0.49 240 3.7 5.8 1.7 0.23 0.60 0.87 0.12 0.30 390 50 130 1.2 0.15 0.39 5.0 0.62 1.3

24 4.0 32

mg/kg 0.48 0.50 0.11 μg/kg 25 25 11 μg/kg 37 36 5.0 μg/kg 73 74 16

0.93 0.31 0.85

19 66

Gd Ge Hf Ho

3.6 1.1 4.3

0.018

S

55 12 200 −0.14 0.86 0.058 100 −0.02

0.42

P75

64 150 20 51 0.72 2.4 0.77 1.1 60 120

8.1 37

24 0.29

0.18

P25

mg/kg 0.27 0.27 0.057 mg/kg 4.5 4.4 2.2 mg/kg 0.37 0.38 0.079 mg/kg 0.18 0.19 0.041 μg/kg 84 85 22 % 0.25 0.26 0.077 mg/kg 0.90 0.93 0.28

3.6 1.0 4.2

530 1900 15 2.1 2200

59 350

1100 4.4

1.5

Max

Cs Cu Dy Er Eu Fe Ga

Ce mg/kg Co mg/kg Cr mg/kg

μg/kg 95 96 30 μg/kg 32 35 5.0 mg/kg 1.3 1.1 0.14 % 0.95 0.94 0.29 μg/kg 82 80 32

1.8 14

Be Bi Br Ca Cd

13 46

mg/kg 13 mg/kg 48

5.0 0.05

B Ba

0.080

0.27

Min

0.27

%

Mean Mеd

Descriptive statistic parameters for element content in moss samples

Ag μg/kg 37 38 As mg/kg 0.48 0.47

Al

Unit

Table 2

41 0.20

3.8

P75 0.20

8.3 27 1.7

13 0.80

20 1.5

2.6 0.11

3.6 8.8 0.43 24 140 17 0.97 2.5 0.11 0.10 0.26 0.01

0.027 0.018 1.6 4.8 8.2 4.0 2.0

0.40 1.3 0.066

E 0.03 −0.22

A

−0.02 −0.11 0.10 −0.05 0.10 −0.07 −0.07

−0.12 3.82 −0.29 −0.45 −0.42 0.33 0.06

140 99

0.08 0.05

0.99 5.36

71 0.06 −0.52 150 −0.03 −0.26 71 0.01 −0.38 69 −0.04 −0.29

71 85 59 70 67 90 70

73 0.06 −0.25 69 −0.08 0.12 56 0.03 −0.22

0.10 0.60 0.03 −0.61 0.06 0.05 0.29 0.25 0.43 −0.35

0.14 0.53

70 −0.08 −0.14 37 0.30 0.41

75

CV

9.4 64 0.12 0.038 140 −0.07

1.7 0.005

S

76 15 230 2.9 0.14 72 5.6 0.51 110 0.17 0.004 34 8.5 17 360

0.21 0.60 0.16 0.25 20 41 40 100 74 190 25 65 18 44

3.7 13 0.93

47 1.0 3.0 0.11 3.6

100 220 0.16 0.34

16 0.13

1.7

P25


% μg/kg mg/kg mg/kg mg/kg μg/kg μg/kg mg/kg % μg/kg mg/kg mg/kg mg/kg mg/kg % mg/kg μg/kg μg/kg μg/kg % mg/kg % mg/kg mg/kg mg/kg mg/kg % mg/kg mg/kg μg/kg mg/kg mg/kg mg/kg mg/kg μg/kg mg/kg μg/kg mg/kg mg/kg mg/kg

Al Ag As B Ba Be Bi Br Ca Cd Co Cr Cs Cu Fe Ga Ge Hf I K Li Mg Mn Mo Na Ni P Pb Rb Sb Sc Sn Sr Ti Tl V W Y Zn Zr

0.35 56 0.73 15 59 130 49 1.4 1.0 84 1.6 6.6 0.36 6.7 0.38 1.3 34 57 100 0.25 3.1 0.20 170 0.36 71 5.2 0.14 6.7 7.6 21 1.7 0.31 25 210 59 9.3 49 2.2 19 1.8

River terraces (Q) 0.25 41 0.43 12 48 98 30 0.83 0.90 86 0.93 4.1 0.28 4.3 0.24 0.86 24 44 79 0.20 2.5 0.15 150 0.20 57 4.5 0.13 5.5 5.9 17 1.2 0.23 27 120 35 5.3 39 1.5 16 1.3

River sediment (Ng) 0.38 46 0.95 19 78 130 54 1.0 1.1 92 1.1 3.6 0.57 4.8 0.30 1.1 28 45 100 0.21 3.0 0.18 190 0.31 54 3.7 0.15 9.8 7.4 23 1.5 0.20 39 170 61 7.6 97 1.9 19 1.3

Pyroclastite (Ng)

Q Quaternary, Ng Neogene, Pg Paleogene, R Rifey, Pt Proterosoic, Pz Paleozoic

Unit 0.28 33 0.49 20 45 100 35 1.7 1.1 76 1.2 5.7 0.30 4.9 0.25 0.89 27 44 93 0.24 3.5 0.20 110 0.24 59 5.0 0.14 5.8 5.9 18 1.3 0.22 42 100 39 6.4 71 1.2 16 1.4

Flysh (Pg) 0.23 36 0.44 14 36 76 19 1.4 1.1 77 1.0 4.3 0.20 3.7 0.23 0.76 21 32 96 0.18 2.3 0.16 120 0.17 55 4.3 0.12 3.7 4.8 15 1.2 0.20 28 110 31 5.8 39 1.1 12 0.88

Schists (Pz)

Element distribution according to the dominant geological formation in the area (median values are given)

Element

Table 3

0.29 29 0.50 12 46 95 31 1.6 1.0 69 1.1 4.5 0.27 4.2 0.29 0.99 26 35 100 0.21 2.7 0.16 150 0.22 60 3.9 0.14 5.1 4.6 18 1.3 0.20 25 150 37 6.7 70 2.0 15 1.0

Schists (R) 0.33 51 0.51 13 48 91 38 1.5 0.92 96 1.3 4.5 0.24 4.6 0.30 0.95 25 26 93 0.20 2.5 0.16 170 0.20 64 4.2 0.14 7.7 5.6 19 1.2 0.18 23 140 39 5.9 55 2.0 19 0.71

Schists (Pt) 0.23 33 0.34 6.9 45 78 30 1.1 0.74 95 0.90 3.2 0.19 4.3 0.21 0.76 21 28 89 0.22 1.8 0.13 200 0.18 60 3.4 0.13 4.7 5.0 17 1.1 0.18 18 120 31 4.6 62 1.5 15 0.84

Gneisses (Pt)

0.23 37 0.46 10 44 91 24 1.3 0.87 77 0.84 3.5 0.26 4.2 0.21 0.82 23 36 90 0.23 2.1 0.15 150 0.24 53 3.4 0.14 4.2 4.9 21 1.1 0.19 25 100 41 5.1 71 1.3 15 1.1

Granite (Pt)

Author's personal copy

Environ Sci Pollut Res


the “coinage metals”, have very similar geochemistry associated with their organometallic chemistry (Greenwood and Earnshaw 2005). For As, Sb and Bi, no significant enrichments were detected for their atmospheric deposition in moss. None of these three elements participate significantly in the composition of the Earth’s crust, although several minerals contain them as major constituents. The median value for arsenic content (0.47 mg kg−1) was lower than the median value obtained for whole of Macedonia reported by Harmens et al. (2015). However, arsenic was significantly enriched compared with

Table 4 Matrix of dominant factor loading (dominant loading values are marked in italics)

Element Al Be

F1 0.76 0.68

other European countries, such as Norway (0.13 mg kg−1), France (0.18 mg kg − 1 ) or Finland (0.11 mg kg − 1 ). Dumurdzanov et al. (2004) showed that natural enrichment of arsenic may occur in areas where Neogene vulcanites are dominant geological units. In the present study, bismuth content ranged from 0.005 to 1.9 mg kg−1, with predominant distribution in areas with Neogene volcanism (median value 0.054 mg kg − 1 ). For antimony, a median value of 0.019 mg kg−1 was found. An investigation conducted on the whole territory of the Republic of Macedonia by Barandovski et al. (2015) showed that Sb enrichments are of

F2 0.14 0.12

F3 0.38 0.50

F4

F5

F6

F7

Communality

0.19 0.23

0.24 0.27

0.23 0.06

0.07 0.11

89.0 87.0

Co

0.73

0.10

0.19

0.20

0.11

0.50

0.19

92.0

Fe Ga Ge

0.82 0.81 0.70

0.11 0.10 0.05

0.30 0.37 0.14

0.14 0.14 0.28

0.17 0.27 0.43

0.34 0.22 0.23

0.13 0.15 0.13

95.9 96.9 84.6

Li Mg Sc

0.52 0.49 0.79

0.01 0.11 0.10

0.48 0.23 0.22

0.40 0.57 0.15

0.19 0.07 0.25

0.34 0.36 0.30

0.11 0.11 0.05

82.0 78.4 85.5

Ti V

0.78 0.74

0.21 0.12

0.03 0.31

0.01 0.20

0.14 0.28

0.25 0.35

0.07 0.15

75.2 92.9

Y La–Gd Eu–Lu

0.95 0.90 0.95

0.07 0.10 0.09

0.14 0.22 0.15

0.10 0.13 0.08

0.02 0.09 0.05

0.04 0.01 0.03

0.07 0.12 0.06

94.5 91.0 95.4

Ba Bi

0.35 0.37

0.52 0.55

0.23 0.31

0.22 0.01

0.37 0.07

0.26 0.31

0.19 0.25

73.5 70.1

Cd Pb Sb Zn

0.04 0.08 0.51 0.25

0.88 0.81 0.63 0.82

0.10 0.32 0.09 0.15

0.06 0.06 0.10 0.05

0.06 0.16 0.15 0.08

0.09 0.10 0.17 0.22

0.08 0.10 0.09 0.00

81.2 80.9 73.7 81.8

As

0.40

0.34

0.55

0.24

0.09

0.25

0.21

75.3

Cs Rb Tl Ca B Sr Hf Zr Cr Cu

0.33 0.47 0.41 0.21 0.16 0.02 0.34 0.25 0.55 0.24 0.27 0.16 0.15 30.3 17.4

0.21 0.24 0.31 0.06 0.02 0.18 0.09 0.15 0.07 0.54 0.08 0.05 0.24 11.7 3.62

0.77 0.66 0.60 0.01 0.22 0.13 0.19 0.21 0.17 0.22 0.02 0.15 0.12 10.2 2.88

0.27 0.07 0.12 0.85 0.81 0.78 0.13 0.11 0.29 0.16 0.35 0.19 0.06 9.8 2.46

0.14 0.30 0.35 0.01 0.01 0.38 0.87 0.88 0.16 0.09 0.16 0.08 0.05 8.3 1.76

0.03 0.07 0.03 0.18 0.09 0.07 0.11 0.15 0.65 0.55 0.72 0.11 0.13 8.0 1.34

0.14 0.10 0.26 0.13 0.04 0.09 0.03 0.02 0.22

86.4 82.2 82.8 81.7 73.7 82.0 94.9 94.7 91.4

0.09 0.16 0.84 0.81 5.7 1.10

74.0 77.8 80.8 77.7 84.0

Ni Br I Total variability (%) Eigenvalue


Elements

Fig. 5 Dendogram for dominant clusters

Al Fe Ga V Sc Ge Be Li Ti Y Eu-Lu La-Gd Co Cr Ni Mg Ca B Sr Br I As Cs Rb Tl Ba Hf Zr Ag Bi Cu Mo Cd Pb Zn Sb 0

25

50

75

100

Dlink/Dmax (%)

anthropogenic origin, occurring in areas with hydrothermal Pb–Zn mineral exploitation. Boron is an element with very specific chemistry in the environment. The content of B in the moss samples ranged from 1.8 to 59 mg kg −1 with a median of 13 mg kg−1. The atmospheric distribution of boron in the investigated area was concentrated in areas of Paleogene flysch (median 20 mg kg−1). Alkaline earth metals, such as Be, Ba and Sr, were very similar to boron. Barium in moss samples ranged from 14 to 350 mg kg−1. Beryllium was dominantly enriched on the Quaternary terraces and in areas with dominant Ng volcanism (0.13 mg kg−1), while the whole investigated area was characterised by Be deposition in the range of 30–350 μg kg−1, with a median value of 96 μg kg−1 (Table 2). Lithium, caesium and rubidium can also be incorporated into this group. These elements showed very low variation in their deposition according to the different geological units. These alkaline metals show very good stability, indicating their lithogenic distribution. The investigated area was characterised by median values for Li, Cs and Rb of 2.6, 0.27 and 5.4 mg kg−1, respectively (Table 2). Gallium and thallium are usually correlated with the occurrence of vulcanites; this was also demonstrated in the present investigation. Deposition of these elements occurred dominantly in areas of Neogene pyroclastites and vulcanite (Kratovo-Zletovo district). Characterisation of the study area according to dominant geological units, extracted Tl, was found at 61 mg kg−1 in areas of Neogene pyroclastites. River terraces were also enriched in Ga and Tl (1.3 mg kg−1 and 59 μg kg−1, respectively).

The halogens bromine and iodine were deposited with median values of 1.1 mg kg−1 and 91 μg kg−1, respectively. These elements show very high reactivity in the environment, and usually, their deposition is correlated with their geochemical occurrence as potassium and sodium salts. There are usually enriched in river terraces, as in the present case. However, no significant variation occurred among the dominant geological units. Chromium and nickel showed very similar distribution and were found in the range of 0.85 to 32 and 1.6–45 mg kg−1, respectively (Table 2). Their distribution is dominant in Quaternary terraces, probably due to the accumulation of soil dust from soil formed in Paleogene flysch (Table 3). The distribution of the LREEs (La–Ce–Pr–Nd–Sm–Gd) was predominantly in Quaternary terraces and the Paleogene flysch. Anthropogenic activities surrounding Pb–Zn and Cu mineral deposits significantly influenced the lithological distribution of this geochemical association of elements. On the other hand, the long-term deposition of the HREEs (Eu–Tb– Dy–Ho–Er–Tm–Yb–Lu) presented a distribution typical of the geochemical markers of the Bregalnica River basin area (Balabanova et al. 2015). The latest results reported from the UNECE ICP Vegetation—Heavy Metals in European Mosses (Barandovski et al. 2008, 2012, 2013, 2015; Harmens et al. 2015) indicate the areas with the highest health risks due to higher values of Cd, Pb and Zn. Such areas have been identified in the eastern Macedonia, due to the operation conducted in mines for hydrothermal exploitation of Pb–Zn (at the Sasa mine and Zletovo mines). Lead and zinc are specifically distributed around Kamenička and Zletovska River, areas with dominant Neogene and Paleogene volcanism. The intensive deposition of these


elements was revealed in the present study by maximum content in moss of 1400 mg kg−1 for Pb and 240 mg kg−1 for Zn (Table 2). Cadmium content varied in the range 0.032– 2.2 mg kg−1. Compared with the whole territory of the Republic of Macedonia (median 0.22 mg kg−1), cadmium showed reduced distribution in the study area (Barandovski et al. 2013; Harmens et al. 2015). Extraction of dominant geochemical associations The multivariate statistical method “hunting” was applied to interpret the numerous element distributions and identifying the dominant geochemical associations. The extraction was simplified to a seven-factor association, marked as follows:

Fig. 6 Distribution of element associations according to dominant geological units (Ng Neogene; Pg Paleogene; R Rifey; Pt Proterosoic; Pz Paleozoic)

F1: Al–Be–Co–Fe–Ga–Ge–Li–Mg–Sc–Ti–V–Y–(La– Gd)–(Eu–Lu) with a factor loading of 30.3 %; F2: Ba–Bi– Cd–Pb–Sb–Zn (factor loadings of 11.7 %); F3: As–Cs–Rb– Tl (factor loading of 10.2 %); F4: Ca–Sr–B (factor loading of 9.8 %); F5: Hf–Zr (factor loading of 8.3 %); F6: Cr–Cu–Ni (factor loading of 8.0 %); and F7: Br–I (factor loading of 5.7 %), with a total variability for the dominant loadings of 84 %. Table 4 summarises the factor loadings for dominant element associations. Cluster analysis was also applied, in order to determine the functional dependence of the internal associations (Fig. 5). Factor 1: [Al–Be–Co–Fe–Ga–Ge–Li–Mg–Sc–Ti–V– Y–(La–Gd)–(Eu–Lu)] can possibly be the presence of elements associated with mineral particles, mainly windblown


dust, and contributes 30 % of the variance. The highest values of the factor loading were obtained for the rare earth elements (0.95 and 0.90). The distribution of these elements represents a specific lithological phenomenon, due to the strong dominance of rare earth elements (Balabanova et al. 2015). As a dominant lithogenic marker for atmospheric distribution for the whole territory of Macedonia, the following association was identified: Al–As–Ba–Ce–Co–Cs–Dy–Fe–Hf–La–Li– Mg–Na–Nd–Sc–Sm–Ta–Tb–Th–Ti–U–V–W–Yb–Zr (Barandovski et al. 2015). The content of these elements in mosses is significantly influenced by secondary sources— mineral particles released into the atmosphere by wind erosion of local sources or particles attached to the moss in the periods when the soil surface is covered by water. Factor loading values were graphically represented according to the geological units (Fig. 6). The highest contents of the aforementioned group of elements were found in moss samples that were collected on Precambrian and Paleozoic schists, or close to them (Fig. 7). As a result of the actions of natural conditions, weathering of shales occurs very fast and consequently releases these elements into the environment. Factor 2 (Ba–Bi–Cd–Pb–Sb–Zn) is probably associated with long-range transport of air pollution and contributes 11.7 % of the variance. The highest loading values were obtained for Cd (0.88), Zn (0.82) and Pb (0.81). The elements Cd, Pb, Zn and Sb were also identified as anthropogenic

Fig. 7 Areal distribution of factor 1: Al–Be–Co–Fe–Ga–Ge–Li– Mg–Sc–Ti–V–Y–(La–Gd)–(Eu– Lu)

markers for the whole territory of Macedonia (Barandovski et al. 2015). The atmospheric deposition of these elements was intensified in the areas of Neogene pyroclastite and Proterozoic and Paleozoic schists (Figs. 8 and 9). There are two dominant sub-areas that mostly affect the anthropogenic enrichment of these elements. The first is an area with dominant occurrence of Neogene pyroclastites and vulcanites, where hydrothermal exploitation of Pb–Zn is in operation (Zletovo mine). The second is an area of very old volcanic minerals (Paleogene volcanic sedimentary rocks and Paleogene flysch) where hydrothermal exploitation of the Pb–Zn is also in operation (Sasa mine). This area of very old volcanic rocks continues in the NW direction with another area of Pb–Zn mineral exploitation (Toranica mine). These mentioned sub-areas are predominantly affected by the dispersion of particles from flotation tailing dam sites near the Zletovo and Sasa mines. It should be noted that dust particles are significantly enriched in Pb and Zn. Soil surface windblown dust also affects Pb–Zn enrichment due to the significant soil contamination that occurs in these areas (Stafilov 2014). A similar anomaly was recorded in northern France, exhibiting a similar pattern for the environmental distribution of Pb vs. Zn (Buorennane et al. 2010). This anomaly was extended to the Cd enrichment due to the presence of the urbanised areas (Buorennane et al. 2010). The present data point to pollution as source for this enrichment (Zletovo and


Sasa mines). Lead and zinc distribution patterns were separately given, suggesting industrial pollution (Fig. 8a). The affected areas with significantly enriched Pb/Zn deposition

were predominantly detected in the areas of Neogene and Paleogene volcanism. The geochemistry of the igneous rock is known to correlate with extended poly-metallic enrichments

Fig. 8 Areal distribution of factor 2: Ba–Bi–Cd–Pb–Sb–Zn. a Areal distribution of lead (left) and zinc (right) and average content distribution according to dominant lithological units (Ng Neogene; Pg Paleogene; R Rifey; Pt Proterosoic; Pz Paleozoic)

Author's personal copy Environ Sci Pollut Res Fig. 9 Areal distribution of factor 3: As–Cs–Rb–Tl

Fig. 10 Areal distribution of factor 4: Ca–Sr–B

Author's personal copy Environ Sci Pollut Res Fig. 11 Areal distribution of factor 5: Hf–Zr

Fig. 12 Areal distribution of factor 6: Cr–Cu–Ni


(Keller et al. 2015). Furthermore, this association can be used as a very intensive probabilistic model of anthropogenic polymetallic enrichments. Factor 3 (As–Cs–Rb–Tl) indicates a characteristic deposition of these elements linked to old volcanism in the test area (Fig. 6). The spatial distribution of these elements follows the area distribution of factor 2. This geogenic marker probably occurs as secondary poly-metallic enrichments in areas where vulcanite dominates (Fig. 9). Predominantly, this occurrence marked areas with Neogene vulcanites vs. Paleogene vulcanites. The specific distribution of As, Cs, Rb and Tl was extended in the SE direction into the vicinity of Berovo town. There are existing data that suggest a poly-metallic enrichment in this area (Arsovski 1997). This association can be used for the identification of anthropogenic hydrothermal mineral exploitation, due to the area distribution patterns (Fig. 9). Higher values for these elements were obtained in the Sasa Pb–Zn mine area, the environs of the Zletovo Pb–Zn mine and in the Bučim copper mine area. Factor 4 (Ca–Sr–B) was extracted with a total variability of 9.8 %. The distribution of Ca and Sr appears in the rocks of the Earth’s crust, mainly as insoluble carbonates, sulphates and silicates (Greenwood and Earnshaw 2005). Their total representation depends on the geochemical model they used, particularly the relative weights given to different types of volcanic and sedimentary rocks. According to their distribution in different geological units, this association dominates in Paleogene flysch (Fig. 6). These old sedimentary rocks are Fig. 13 Areal distribution of factor 7: Br–I

natural repositories for these elements. This factor was dominant along the whole course of the Bregalnica River which was the determined dominance of this factor, with emphasis on the area of the Bregalnica River, especially where it flows into the Vardar River (intensive occurrence of Paleogene flysch, Fig. 10). This lithogenic marker can be used for identification of natural enrichments due to wind dusting in areas of very old sedimentary rock in river basins. Factor 5 (Hf–Zr) presents a natural phenomenon in the investigated area, due to the specific geochemistry of these elements in the environment (Fig. 11). Their very similar chemical background causes them to be correlated, and both Zr and Hf behaviour is similar to the HREEs (Prudêncio 2006). No significant variation occurred according to the dominant lithological units, but some intensive deposition can be singled out in an area of Neogene volcanism. This area was characterised by polymetallic enrichments (with special emphasis on Pb, Zn, Sb and Cd). Lower enrichments of Hf and Zr in this kind of area have also been identified in Portugal, in correlation with significant Zn enrichments (Prudêncio 2006). Barandovski et al. (2015) reported a decreasing trend in Hf–Zr distribution from 2005 to 2010 in the whole territory of the Republic of Macedonia. The data obtained in this work compared to values from the 2010 investigation (Barandovski et al. 2015) show lower medians. Despite the decreasing trend in Hf–Zr deposition, this geogenic association can still be correlated with some poly-metallic anomalies. Factor 6 (Cr–Cu–Ni) associates elements primarily affected by natural factors such as the lithological background.


Associations with Paleogene flysch sediments and Neogene sediments such as those that occur in the Vardar zone have been repeatedly demonstrated in Macedonia (Stafilov et al. 2010; Balabanova et al. 2010, 2011, 2012; Bačeva et al. 2012), Kosovo (Šajn et al. 2013) and Bosnia and Herzegovina (Alijagić and Šajn 2011) and in other Balkan countries (Salminen et al. 2005). These elements can also be connected to air pollution. Most emissions of these elements to the environment originate from local sources, mainly from mines. The Bučim copper mine contributes to anthropogenic copper emissions, in the south-east part of the study area. However, significant enrichment of this geochemical association occurs in the area of Paleogene flysch, overlapping with part of the Vardar River basin. Barandovski et al. (2015) identified a strong anomaly in Cr–Co–Ni distribution along the Vardar River basin. This anomaly affects a part of the Bregalnica River basin and was also demonstrated in the present study (Fig. 12). Factor 7 (Br–I) associates elements that are primarily affected by marine influence (Frontasyeva and Steinnes 2004). The coastline of the Aegean Sea is less than 60 km from the southern state border. Barandovski et al. (2015) reported the influence of the wind blowing along the Vardar River valley from northern parts of Africa across the Mediterranean and the Aegean Sea. High contents of halogens in mosses and their exponential decrease with the distance from the coastline are also observed in Norway (Frontasyeva and Steinnes 2004). Elevated values of these elements were found in the mosses collected in an area connected with the presence of Neogenic sediments along the river Begalnica (Fig. 13). Lower enrichments of these elements were also found in an area of Rifeous schist.

Conclusion This paper focuses on the mapping of the lithogenic and anthropogenic atmospheric distribution of 69 elements in the Bregalnica River basin, Republic of Macedonia. Specific poly-metallic enrichments occur in the areas of very old Neogene volcanism and Paleogene volcanism. Strong relationships occur between these dominant lithological units and the hydrothermal exploitation in the area of the Sasa Pb–Zn mine, the Zletovo Pb–Zn mine and the Bučim copper mine. The multivariate approach and spatial distribution method hunting identified seven dominant clusters for the atmospheric distribution of the 69 analysed elements: F1: Al– Be–Co–Fe–Ga–Ge–Li–Mg–Sc–Ti–V–Y–(La–Gd)–(Eu–Lu); F2: Ba–Bi–Cd–Pb–Sb–Zn; F3: As–Cs–Rb–Tl; F 4: Ca–Sr–B; and F5: Hf–Zr; F6: Cr–Cu–Ni; F7: Br–I. Intensive effects were determined for the factor 1 distribution that is strongly correlated with the dominant lithologic unit for the whole territory of the Republic of Macedonia (extracted as the dominant lithogenic marker). The second geochemical association (including Cd– Pb–Zn–Sb) was defined as a dominant anthropogenic marker for the atmospheric element distributions. Significantly polluted

sub-areas were determined, affected by significant enrichments in the atmospheric depositions of Pb and Zn and lower enrichments of Cd and Sb. The Zletovo and Sasa mine areas were singled out as poly-metallic enriched areas not only from the anthropogenic emission but also from the influence of windblown dusting on the soil surface. Accordingly, the resulting distribution maps were used to support the assessment of anthropogenic inputs and enrichments with high certainty. Acknowledgments This work was funded by Core Program, under the support of ANCSI, Project No. PN16.40.02.01.

References Aboal JR, Fernández JA, Boquete T, Carballeira A (2010) Is it possible to estimate atmospheric deposition of heavy metals by analysis of terrestrial mosses? Sci Total Environ 408:6291–6297 Acton QA (2013) Issues in environmental health and pollution. Scholatity Edition press, Atlanta, Georgia Alderton DHM, Serafimovski T, Mullen B, Fairall K, James S (2005) The chemistry of waters associated with metal mining in Macedonia. Mine Water Environ 24(3):139–149 Alijagić J, Šajn R (2011) Distribution of chemical elements in an old metallurgical area, Zenica (Bosnia and Herzegovina. Geoderma 162:71–85 Ares A, Aboal JR, Carballeira A, Giordano SP, Adamo P, Fernández JA (2012) Moss bag biomonitoring: a methodological review. Sci Total Environ 432:143–158 Arsovski M (1997) Tectonics of Macedonia. Faculty of Mining and Geology, Štip:1–306 Athar M, Vohora S (1995) Heavy metals and environment. New Age International publishers, New Delhi Bačeva K, Stafilov T, Šajn R, Tănăselia C (2012) Moss biomonitoring of air pollution with heavy metals in the vicinity of a ferronickel smelter plant. J Environ Sci Health A 47:1–13 Balabanova B, Stafilov T, Bačeva K, Šajn R (2010) Biomonitoring of atmospheric pollution with heavy metals in the copper mine vicinity located near Radoviš, Republic of Macedonia. J Environ Sci Health A 45:1504–1518 Balabanova B, Stafilov T, Šajn R, Bačeva K (2011) Distribution of chemical elements in attic dust as reflection of lithology and anthropogenic influence in the vicinity of copper mine and flotation. Arch Environ Contam Toxicol 6:173–184 Balabanova B, Stafilov T, Šajn R, Bačeva K (2012) Characterisation of heavy metals in lichen species Hypogymnia physodes and Evernia prunastri due to biomonitoring of air pollution in the vicinity of copper mine. Int J Environ Res 6(3):779–792 Balabanova B, Stafilov T, Šajn R (2015) Litogenic distribution of rare earth elements in automorphic and alluvial soils from Bregalnica River basin. Maced J Chem Chem Eng 34(1):201–212 Barandovski L, Cekova M, Frontasyeva MV, Pavlov SS, Stafilov T, Steinnes E, Urumov V (2008) Atmospheric deposition of trace element pollutants in Macedonia studied by the moss biomonitoring technique. Environ Monit Assess 138:107–118 Barandovski L, Frontasyeva MV, Stafilov T, Šajn R, Pavlov S, Enimiteva V (2012) Trends of atmospheric deposition of trace elements in Macedonia studied by the moss biomonitoring technique. J Environ Sci Health A 47(13):2000–2015 Barandovski L, Stafilov T, Šajn R, Frontasyeva MV, Bačeva K (2013) Air pollution study in Macedonia using a moss biomonitoring technique, ICP-AES and AAS. Maced J Chem Chem Eng 32(1):89–107

Author's personal copy Environ Sci Pollut Res Barandovski L, Frontasyeva MV, Stafilov T, Šajn R, Ostrovnaya T (2015) Atmospheric deposition of trace elements in Macedonia studied by the moss biomonitoring technique. Environ Sci Pollut Res 22(20): 16077–16097 Boquete MT, Fernández JA, Carballeira A, Aboal JR (2013) Assessing the tolerance of the terrestrial moss Pseudoscleropodium purum to high levels of the atmospheric heavy metals: a reciprocal transplants study. Sci Total Environ 461-462:552–559 Bourennane H, Douay F, Sterckeman T, Villanneau E, Ciesielski H, King D, Baize D (2010) Mapping of anthropogenic trace elements inputs in agricultural topsoil from Northern France using enrichment factors. Geoderma 157(3):165–174 Carballeira CB, Aboal JR, Fernández JA, Carballeira A (2008) Comparison of the accumulation of elements in two terrestrial moss species. Atmos Environ 42:4904–4917 Cekova M (2005) Review of the bryoflora of the Republic of Macedonia. Ss. Cyril and Methodius University, Skopje in Macedonian Davis Jr FD (1986) A technology acceptance model for empirically testing new end-user information systems: Theory and results (Doctoral dissertation, Massachusetts Institute of Technology). Dumurdzanov N, Serafimovski T, Burchfiel BC (2004) Evolution of the Neogene-Pleistocene basins of Macedonia. Geological Society of America, Digital Map and Chart Series 1 Dumurdzanov N, Serafimovski T, Burchfiel BC (2005) Cenozoic tectonics of Macedonia and its relation to the South Balkan extensional. Geosphere 1:1–22 Fernández JA, Boquete MT, Carballeira A, Aboal JR (2015) A critical review of protocols for moss biomonitoring of atmospheric deposition: sampling and sample preparation. Sci Total Environ 517:132–150 Filzmoser P, Garrett RG, Reimann C (2005). Multivariate outlier detection in exploration geochemistry. Comput Geosci 31(5):579-587 Frontasyeva MV, Steinnes E (2004) Marine gradients of halogens in moss studied by epithermal neutron activation analysis. J Radioanal Nucl Chem 261:101–106 Gjengedal E, Steinnes E (1990) Uptake of metal ions in moss from artificial precipitation. Environ Monit Assess 14:77–87 Greenwood NN, Earnshaw A (2005) Chemistry of the elements, 2nd edn. Elsevier Butteworth-Heinemann, Oxford Harmens H, Buse A, Büker P, Norris D, Mills G, Williams B (2004) Heavy metal concentrations in European mosses: 2000/2001 survey. J Atmos Chem 49:425–436 Harmens H, Noris D, Viňas JA, Alber R, Aleksiayenak Y, Ashmore M, Barandovski L, et al. (2008) Spatial and temporal trends in heavy metal accumulation in mosses in Europe (1990–2005). In: Harmens H, Noris D (eds) Programme Coordination Centre for the ICP Vegetation, Centre for Ecology & Hydrology. Natural Environment Research Council, Bangor Harmens H, Norris DA, Steinnes E, Kubin E, Piispanen J, Alber R, Aleksiayenak Y, Blum O, Coşkun M, Dam M, De Temmerman L, Fernández JA, Frolova M, Frontasyeva M, González-Miqueo L, Grodzińska K, Jeran Z, Korzekwa S, Krmar M, Kvietkus K, Leblond S, Liiv S, Magnússon SH, Maňkovská B, Pesch R, Rühling A, Santamaria JM, Schröder W, Spiric Z, Suchara I, Thöni L, Urumov V, Yurukova L, Zechmeister HG (2010) Mosses as biomonitors of atmospheric heavy metal deposition: spatial and temporal trends in Europe. Environ Pollut 158:3144–3156 Harmens H, Norris DA, Sharps K, Mills G, Alber R, Aleksiayenak Y, et al. (2015) Heavy metal and nitrogen concentrations in mosses are declining across Europe whilst some “hotspots” remain in 2010. Environ Pollut 200:93–104 Keller CB, Schoene B, Barboni M, Samperton KM, Husson JM (2015) Volcanic-plutonic parity and the differentiation of the continental crust. Nature 523(7560):301–307 Lehmann S, Barcikowski J, von Quadt A, Gallhofer D, Peytcheva I, Heinrich CA, Serafimovski T (2013) Geochronology, geochemistry

and isotope tracing of the Oligocene magmatism of the BuchimDamjan-Borov Dol ore district: implications for timing, duration and source of the magmatism. Lithos 180–181:216–233 Li W, Wang Y, Collet JL, Chen JJ, Zhang X, Wang Z, Wang W (2013) Microscopic evaluation of trace metals in cloud droplets in an acid precipitation region. Environ Sci Technol 47(9):4172–4180 Prudêncio M (2006) Biogeochemistry of trace and major elements in a surface environment (volcanic rock, soil, mosses, lichens) in the S. Miguel Island, Azores. Portugal J Radioanal Nucl Ch 271(2):431–437 Rakićević Т, Dumurdzanov N, Petkovski М (1968) Basic geological map of SFRJ, sheet Štip, M 1:100,000 (map & interpreter). Federal Geological Survey, Beograd Reimann C, Filzmoser P, Garrett RG (2002) Factor analysis applied to regional geochemical data: problems and possibilities. Appl Geochem 17:185–206 Rühiling Ǻ, Tyler G (1968) An ecological approach to the lead problem. Bot Notiser 122:248–342 Šajn R, Aliu M, Stafilov T, Alijagić J (2013) Heavy metal contamination of topsoil around a lead and zinc smelter in Kosovska Mitrovica/ Mitrovicë, Kosovo/Kosovë. J Geochem Explor 134:1–16 Salminen R, Batista MJ, Bidovec M, Demetriades A, De Vivo B, De Vos W, Duris M, Gilucis A, Gregorauskiene V, Halamić J, Heitzmann P, Jordan G, Klaver G, Klein P, Lis J, Locutura J, Marsina K, Mazreku A, O’Connor PJ, Olsson SA, Ottesen RT, Petersell V, Plant JA, Reeder S, Salpeteur I, Sandström H, Siewers U, Steenfelt A, Tarvainen T (2005) Geochemical atlas of Europe, part 1, background information, methodology and maps. Geological Survey of Finland, Espoo (http://weppi. gtk.fi/publ/foregsatlas/) Serafimovski T, Tasev G, Dolenec T (2006a) Petrological and geochemical features of the Neogene volcanites of the Oso govo mountains, eastern Macedonia. RMZ Mat Geoenviron 52(3):523–534 Serafimovski T, Tasev G, Dolenec T (2006b) Actinolite-phengite-chlorite metasomatites from the Toranica Pb-Zn ore deposit in Macedonia. RMZ Mat Geoenviron 53(4):445–453 Serafimovski T, Dolenec T, Tasev G, Dolenec M, Rogan N (2007) Acid mine drainage systems and metal pollution around the active polymetallic mines in the eastern Macedonia. Geol Macedonia 21:69–73 Serafimovski T, Stefanova V, Volkovb AV (2010) Dwarf coppergold porphyry deposits of the Buchim-Damjan Borov Dol ore district, Republic of Macedonia (FYROM). Geol Ore Deposit 52(3):179–195 Stafilov T (2014) Environmental pollution with heavy metals in the Republic of Macedonia. Contrib Sect Nat Math Biotech Sci MASA 35(2):81–119 Stafilov T, Šajn R, Pančevski Z, Boev B, Frontasyeva MV, Strelkova LP (2010) Heavy metal contamination of surface soils around a lead and zinc smelter in the Republic of Macedonia. J Hazard Mater 175:896–914 Steinnes E, Rühling Å, Lippo H, Mäkinen A (1997) Reference material for large-scale metal deposition surveys. Accred Qual Assur 2:243–249 Tyler S (1990) Bryophytes and heavy metals: a literature review. Bot J Linn Soc 104:231–253 Vallero D (2014) Fundamentals of air pollution, 5th edn. Academic press, Oxford Zechmeister HG (1995) Correlation between altitude and heavy metal deposition in the Alps. Environ Pollut 89:73–80 Zechmeister HG, Hohenwallner D, Riss A, Hanus-Illnar A (2003) Variations in heavy metal concentrations in the moss species Abietinella abietina (Hedw.) Fleisch. according to sampling time, within site variability and increase in biomass. Sci Total Environ 301(1):55-65 Žibret G, Šajn R (2010) Hunting for geochemical associations of elements: factor analysis and self-organizing maps. Math Geosci 42: 681–703