air pollution in study (2013)
DESCRIPTION
Air pollutionTRANSCRIPT
-
Air Pollution Study in Croatia Using Moss Biomonitoringand ICPAES and AAS Analytical Techniques
Zdravko Spiric Ivana Vuckovic Trajce Stafilov
Vladimir Kusan Marina Frontasyeva
Received: 8 December 2012 / Accepted: 11 February 2013 / Published online: 7 March 2013
Springer Science+Business Media New York 2013
Abstract Moss biomonitoring technique was applied in a
heavy-metal pollution study of Croatia in 2006 when this
country participated in the European moss survey for the
first time. This survey was repeated in 2010, and the results
are presented in this study. For this purpose, 121 moss
samples were collected during summer and autumn 2010.
The content of 21 elements was determined by inductively
coupled plasmaatomic emission spectrometry and atomic
absorption spectrometry. Principal component analyses
was applied to show the association between the elements.
Six factors (F1F6) were determined, of which two are
anthropogenic (F3 and F6), two are mixed geogenic
anthropogenic (F1 and F5), and two are geogenic factors
(F2 and F4). Geographical distribution maps of the ele-
ments over the sampled territory were constructed using
geographic information systems technology. Comparison
of the median values of some of the anthropogenic ele-
mentssuch as arsenic, cadmium, chromium, copper,
mercury, nickel, lead, vanadium, and zincwith those
from the 2006 study shows that anthropogenic pollution
has changed insignificantly during the last 5 years. The
data obtained in the investigation in Norway are taken for
comparison with pristine area, which indicates that Croatia
is somewhat polluted but still, shows a more favourable
picture when compared with two neighbouring countries.
Environmental pollution is one of the largest problems the
world faces today. It is an issue that troubles our economy,
our health, and our daily lives. The contamination of the
environment is also being linked to some of the diseases
that are currently present. The major sources of air pollu-
tion are power and heat generation, burning of solid wastes,
industrial processes, and transportation. The best way to
determine the extent of contamination in living organisms
is by measurement of the levels of contaminants in the
organisms themselves, for which plants have proven to be
the most suitable. The use of mosses as biomonitors is a
convenient way of determining levels of atmospheric
deposition (Puckett 1988). Many studies have showed that
terrestrial mosses have good bioaccumulating ability, par-
ticularly for heavy metals, where metal concentrations
reflect deposition without the complication of additional
uptake by way of a root system (Bargagli et al. 1995;
Mulgrew and Williams 2000; Gjengedal and Steinnes
1990). The moss biomonitoring technique, first introduced
in Scandinavia, has proven to be feasible for studying the
atmospheric deposition of heavy metals as well as other
trace elements (Ruhling and Tyler 1971; Gjengedal and
Steinnes 1990; Ruhling 2002). A few studies have shown
that the content of metals in moss is correlated with
atmospheric deposition during the moss growth period
(Berg and Steinnes 1997; Thoni et al. 2011).
Many national and international organizations have
performed a number of monitoring and modelling studies
to determine the levels of heavy metals and toxic metals in
the air, soil, and water to take further precautions. The first
Z. Spiric (&) V. KusanInstitute for Applied Ecology, OIKON Ltd., Trg senjskih
uskoka 1-2, 10020 Zagreb, Croatia
e-mail: [email protected]
I. Vuckovic T. StafilovInstitute of Chemistry, Faculty of Natural Sciences and
Mathematics, Ss. Cyril and Methodius University,
POB 162, 1000 Skopje, Macedonia
M. Frontasyeva
Frank Laboratory of Neutron Physics, Joint Institute for Nuclear
Research, Str. Joliot-Curie6, 141980 Dubna, Moscow Region,
Russian Federation
123
Arch Environ Contam Toxicol (2013) 65:3346
DOI 10.1007/s00244-013-9884-6
-
local study of mercury (Hg) and some other elements from
the petroleum industry in Croatia was undertaken in
19962004 using lichens (Horvat et al. 2000). In
20052006, Croatia participated for the first time in a moss
survey in the framework of the International Cooperative
Programme on effects of air pollution on natural vegetation
and crops with heavy metals in Europe (UNECE ICP
Vegetation) (Spiric et al. 2012). The method of moss
monitoring has become well established and is used on a
national basis to evaluate the level of heavy-metal pollution
in European countries every 5 years. The first survey
started in 1990.
In this study, a portion of the results obtained for the
2010 moss survey will be presented. The recommended
species of mosses for metal-deposition monitoring (Hyp-
num cupressiforme, Pleurozium schreberi, Brachythecium
rutabulum, and Homalothecium sericeum) (Ruhling et al.
1987; Ruhling 1994; Buse et al. 2003) were collected from
121 sampling sites evenly distributed throughout the ter-
ritory of Croatia. The purposes of this study were as fol-
lows: (1) to present the results from the 2010 moss survey
in Croatia based on moss biomonitoring technique, (2) to
compare these with the results obtained in the 2005 survey
in Croatia to evaluate temporal deposition trends, (3) to
compare the results obtained with similar 2005 surveys
performed in some neighboring countries, and (4) to
compare the result with those obtained in the latest survey
(2010) in Norway, which were used for comparison with a
pristine area.
Study Area
The Republic of Croatia covers an area of 87.677,
56.610 km2 of which is land mass. Croatia is a Mediter-
ranean and southeastern European country geographically
located between 13.5s and 19.5s eastern longitudes and
42.5s and 46.5s northern latitudes. It is located between
BosniaHerzegovina and Serbia in the east, Slovenia in
the west, Hungary in the north, and Montenegro and
the Adriatic Sea in the south. Croatian territory is divided
into three large natural and geographic entities: Half of
the territory (54.5 %) is located in the Pannonian and
Fig. 1 Geographical position ofCroatia
34 Arch Environ Contam Toxicol (2013) 65:3346
123
-
peri-Pannonian region, one-third (31.6 %) belongs to the
Adriatic part, and the rest is hilly and mountainous area
(14 %). Most of Croatia has a moderately warm and
rainy continental climate as defined by Koppen climate
classification (Halamic and Miko 2009). The climate is
strongly conditioned by relief, ranging from continental
temperate in the north to the Mediterranean climate along the
coastline and in the adjacent hinterland. Temperatures
increase from west to east, whereas precipitation varies
reversely. Mountain ranges are characterized by high amounts
of precipitation (B2000 mm/years), whereas areas in their
shadow or in the mainland (Pannonian region) receive little
precipitation (\800 mm/years) (Halamic et al. 2012).Two geologically specific and different areas can be
distinguished within the area of the Republic of Croatia: (1)
the area of north Croatia, or the Pannonian part, composed
predominantly of clastic sedimentary rocks and metamor-
phic and magmatic rocks; and (2) the mountainous and
coastal part, which is made up of predominantly of car-
bonate rocks (Halamic and Miko 2009).
Croatia has a large reserve of freshwater and is one of the
richest areas in Europe in terms of biodiversity. Industries are
mostly located near the larger cities (Zagreb, Sisak, Kutina,
Osjek, Split, Sibenik, etc.). The main industries are heavy
industry, light industry, metallurgy, chemical, steel, and tex-
tile industry. There are also thermoelectric power plants and
oil deposits. All of these previously mentioned industries are
reasons for anthropogenic air pollution with heavy metals. As
a consequence of the overall economic recession since 1990,
air emissions of the main pollutants from stationary and
mobile sources in Croatia have decreased and have a low
influence to the environment. Croatia has a relatively clean
environment compared with European Union industrial
countries (Fig. 1)
Materials and Methods
Sampling
Moss samples of the four most dominant moss species
H. cupressiforme, P. schreberi, B. rutabulum, and H. seri-
ceumwere collected during the summer and autumn in
Fig. 2 Locations of mosssampling points in 2010
Arch Environ Contam Toxicol (2013) 65:3346 35
123
-
2010 at 121 sampling sites evenly distributed over the
territory of Croatia. The map of sampling points is shown
in Fig. 2. The sampling was performed in accordance with
the strategy of the European Moss Survey Programme and
the monitoring manual of the ICP Vegetation Coordination
Centre (Ruhling and Steinnes 1998; Harmens 2009).
Sampling sites were located at least 300 m from main
roads, 100 m from local roads, and 200 m from villages.
Each sampling point was situated at least 3 m away from
the nearest tree canopy. Samples in forests were collected
primarily in small gaps, without influence from canopy drip
from trees, preferably on the ground. Each sample was
composed of 510 subsamples collected within an area of
50 9 50 m2 to make the moss samples representative. A
separate set of polyethylene gloves was used for collection
of each sample. Collected material was stored in paper
bags.
Sample Preparation
Moss samples were air-dried and cleaned of soil particles
and other contaminants. Only the 34 cm of green and
greenbrown shoots from the top of the moss, which rep-
resents the last 3 years of growth, were separated,
homogenized, and used for analyses (Harmens 2009). All
of the reagents used for this study were of analytical grade:
nitric acid; trace pure (Merck, Germany); hydrogen per-
oxide, p.a. (Merck, Germany); and bidistilled water.
Approximately 0.5 g of moss material was placed in a
Teflon vessel and treated with 7 ml of concentrated nitric
acid (HNO3) and 2 ml of hydrogen peroxide (H2O2)
overnight. The moss material was put in microwave
digestion system (Mars; CEM, USA) for full digestion.
Digestion was performed in two steps: (1) ramp: temper-
ature 180 C, time 10 min, power 400 W, and pressure20 bar; (2) hold: temperature 180 C, hold time 20 min,power 400 W, and pressure 20 bar. Digests were filtrated
and quantitatively transferred to 25-ml calibrated flasks
with the addition of bidistilled water to a whole volume of
25 ml.
Instrumentation
Inductively coupled plasmaatomic emission spectrometry
(ICP-AES; 715ES; Varian, USA) was used for analyzing the
content of 19 elements (silver [Ag], aluminum [Al], barium
[Ba], calcium [Ca], cadmium [Cd], chromium [Cr], copper
[Cu], iron [Fe], lithium [Li,] potassium [K], manganese
Table 1 Descriptive statistic for all chemical elements analysed in 121 moss samples (mg kg-1)
Xa Xg Median Minimum Maximum P10 P90 rx
Ag 0.035 0.027 0.032 0.001 0.155 0.010 0.061 0.024
Al 1061 802 878 112 4493 320 1995 808
As 0.39 0.30 0.36 0.05 1.00 0.09 0.70 0.24
Ba 24.53 20.80 20.64 4.49 94.30 10.40 39.95 15.87
Ca 7399 6786 6632 2649 20795 4136 11984 3241
Cd 0.43 0.37 0.38 0.10 1.42 0.19 0.74 0.24
Cr 2.25 1.86 1.94 0.41 8.55 0.90 3.91 1.48
Cu 9.23 8.75 8.53 4.72 22.69 5.96 13.13 3.29
Fe 881 689 789 85.00 4028 305 1658 638
Hg 0.043 0.039 0.043 0.010 0.145 0.023 0.063 0.018
K 4366 4008 3891 1552 9279 2411 6883 1859
Li 0.74 0.57 0.55 0.11 4.27 0.24 1.42 0.60
Mg 3099 3023 3059 1619 4740 2254 4051 683
Mn 161 106 99.10 16.10 928 31.60 324 165
Na 131 125 120 65.00 304 89.00 181 43.96
Ni 3.70 3.15 3.16 1.04 14.66 1.61 6.39 2.35
P 1237 1121 1134 419 3117 631 1954 570
Pb 3.79 3.27 3.21 1.11 36.64 1.87 5.48 3.42
Sr 17.09 15.59 16.00 4.74 54.03 9.02 25.54 7.56
V 3.50 2.42 2.55 0.23 37.26 0.86 6.17 4.05
Zn 27.13 25.18 24.80 11.64 77.13 16.31 41.61 11.86
Xa aritmetical mean; Xg geometrical mean; P10 10th percentile; P90 90th percentile; rx SD
36 Arch Environ Contam Toxicol (2013) 65:3346
123
-
Ta
ble
2C
om
par
iso
no
fm
edia
nv
alu
es(m
gk
g-
1)
and
ran
ges
(mg
kg
-1)
ob
tain
edin
the
pre
sen
tst
ud
yw
ith
the
sam
ep
aram
eter
so
bta
ined
insi
mil
arst
ud
ies
inS
erb
ia,
Slo
ven
ia,
and
No
rway
Cro
atia
(pre
sen
tw
ork
)C
roat
ia2
00
5(S
pir
icet
al.
20
12
)S
erb
ia2
00
5(H
arm
ens
etal
.2
00
8)
Slo
ven
ia2
00
5(H
arm
ens
etal
.2
00
8)
No
rway
20
10
(Ste
inn
eset
al.
20
11
)
No
.o
fsa
mp
les
n=
12
1n
=9
4n
=1
93
n=
57
n=
46
4
Ele
men
tM
edia
nR
ang
eM
edia
nR
ang
eM
edia
nR
ang
eM
edia
nR
ang
eM
edia
nR
ang
e
Ag
0.0
32
0.0
01
0
.15
5
0
.02
\0
.01
4
.22
Al
87
81
12
4
49
31
35
03
98
2
14
63
94
61
11
7
31
18
0
2
83
46
4
58
1
As
0.3
60
.05
1
.00
0.3
70
.10
6
1.4
10
.22
2
1.6
0.4
30
.15
1
.36
0.1
30
.02
4
.84
Ba
20
.64
4.4
9
94
.30
32
7
19
2
2
54
3
25
Ca
66
32
26
49
2
07
95
76
23
28
32
2
67
40
27
87
87
3
85
15
Cd
0.3
80
.10
1
.42
0.2
70
.07
1
.90
.26
0.0
40
1
.11
0.3
30
.13
1
.21
0.0
81
0.0
09
1
.87
5
Cr
1.9
40
.41
8
.55
2.8
0.7
6
33
6.4
42
.00
7
8.8
2.1
40
.85
1
0.3
0.5
90
.16
4
7.8
7
Cu
8.5
34
.72
2
2.6
97
.53
.7
22
.71
1.1
3.0
4
45
18
.17
3.6
9
44
.54
.01
.4
44
3.4
Fe
78
98
5.0
0
40
28
10
00
32
0
12
14
02
26
76
70
1
61
00
94
33
47
4
33
02
78
27
2
46
84
Hg
0.0
43
0.0
10
0
.14
50
.06
40
.00
7
0.3
01
0.0
95
0.0
50
0
.18
0.0
60
\0
.02
4
0.3
38
K3
89
11
55
2
92
79
80
85
25
65
2
37
20
38
67
17
63
8
65
9
Li
0.5
50
.11
4
.27
0.1
17
0.0
11
1
.27
7
Mg
30
59
16
19
4
74
02
12
06
76
1
27
40
13
35
50
2
31
28
Mn
99
.11
6.1
0
92
81
06
20
1
42
1
2
92
19
2
65
3
Na
12
06
5.0
0
30
41
69
67
2
33
2
1
23
11
8
64
Ni
3.1
61
.04
1
4.6
62
.70
.66
1
84
.43
1.7
0
23
.82
.75
0.9
2
8.5
21
.16
0.1
5
85
6.6
6
P1
13
44
19
3
11
7
Pb
3.2
11
.11
3
6.6
42
.46
0.0
6
82
.41
6.7
1.0
3
24
91
0.1
2.5
8
29
.01
.54
0.3
3
20
.83
Sr
16
.00
4.7
4
54
.03
21
4
12
5
1
5.1
1.9
7
2.0
V2
.55
0.2
3
37
.26
3.1
0.9
1
32
5.7
61
.94
3
2.7
3.3
81
.34
1
3.1
1.4
10
.29
2
5.8
8
Zn
24
.80
11
.64
7
7.1
32
91
2
28
32
9.0
13
.2
25
93
8.6
16
.5
99
.33
0.7
7.4
3
68
.4
Arch Environ Contam Toxicol (2013) 65:3346 37
123
-
[Mn], magnesium [Mg], sodium [Na], nickel [Ni], phos-
phorus [P], lead [Pb], strontium [Sr], vanadium [V], and zinc
[Zn]) in moss samples. Ultrasonic nebulizer CETAC (ICP/
U-5000AT?; Varian, USA) was used for better sensitivity
and signal stability. The calibration solutions were prepared
from a 1000 mg L-1 stock solution (ICP-multielement
standard solution IV; Merck, Germany).
The determination of As and Hg using ICP-AES is
difficult at low concentration levels. Therefore, these ele-
ments were determined by AAS. As was determined by
Zeeman electrothermal AAS (SpectrAA 640 Z; Varian,
USA), whereas Hg was determined by cold vapour AAS
(SpectrAA 55B; Varian, USA) using a continuous flow
vapour generation accessory (VGA-76; Varian, USA). The
optimal instrumental parameters for these techniques are
given in previously published articles (Balabanova et al.
2010; Serafimovski et al. 2008).
Statistical Analysis and Mapping
Data matrix for statistical analysis was prepared using all
field observations, analytical data, and measurements. Data
processing was performed using software Statistica (Stat-
Soft version 6). Descriptive statistic for all (n = 21)
chemical elements (Ag, Al, As, Ba, Ca, Cd, Cr, Cu, Fe, Hg,
K, Li, Mg, Mn, Na, Ni, P, Pb, Sr, V, and Zn) from 121
locations were calculated (Tables 1 and 2). Parameters of
descriptive statistics used were as follows: arithmetical
mean, geometrical mean, median, minimum, maximum,
10th percentile, 90th percentile, and SD.
Relationships between heavy-metal contents in mosses
and some environmental, geologic, and anthropogenic fac-
tors have been established by multivariate coordination
methods (Figueira et al. 2002). Principal component analysis
and cluster analysis are frequently used chemometric
methods for the assessment of relationships among heavy
metals in mosses and the identification of their pollution
sources (Gramatica et al. 2006; Pesch and Schroeder 2006).
To discover the associations of chemical elements and to
decrease the number of variables for the obtained data
principal component analysis and factor analysis (PCFA),
R-mod was used (Johnson and Wichern 2002). All sites and
all 21 chemical elements were included in the PCFA. No
standardisation of variables was performed because the
variable standardisation did not give any improvement dur-
ing the analysis preparation and testing phase. For orthogo-
nal rotation, varimax method was used. Other parameters of
analysis were performed according to Stevens (2002): The
number of calculated factors was set to 6 and the minimum
eigenvalue to 1. The threshold of statistical significance was
set to 0.65. With such a high threshold, we decreased the
number of significant loadings in particular factors. It is
clearly visible in factor 2, where Zn has a loading of 0.645,
and in factor 4, where Sr has a loading of 0.646 (Table 3).
For the spatial distribution of factor scores, universal
krigging, i.e., the linear variogram interpolation method,
was used (Barandovski et al. 2012). The parameters of
interpolation were as follows: the number of closest points
was 8, the maximum distance was 200 km, and the basic
grid cell size was 500 m 9 500 m. Spatial representation
for interpolated factors were made by using percentile class
limits of 10 (010, 1020, 2030 90100).
Results and Discussion
Results of the descriptive statistics made for all 21 chem-
ical elements are listed in Table 1. Values of medians and
ranges obtained for the elements in the present study are
compared with the values of the same parameters obtained
in the 2005 study in Croatia and two neighbouring
Table 3 Factor analysis of the data obtained in moss samples fromCroatia
Rotated matrix component
Factor
1
Factor
2
Factor
3
Factor
4
Factor
5
Factor
6
Ag 0.12 0.12 0.72 0.10 0.29 0.09
Al 0.97 0.04 0.01 0.10 0.13 0.02
As
(ETAAS)
0.39 0.26 0.07 0.28 0.36 0.26
Ba 0.09 0.30 0.82 0.11 0.08 0.03
Ca 0.31 0.01 0.31 0.66 0.32 0.04
Cd 0.81 0.24 0.02 0.07 0.14 0.32
Cr 0.93 0.04 0.02 0.07 0.26 0.08
Cu 0.30 0.43 0.04 0.01 0.47 0.50
Fe 0.97 0.02 0.02 0.06 0.12 0.05
Hg (AAS) 0.21 0.03 0.12 0.17 0.31 0.58
K 0.15 0.90 0.03 0.11 0.06 0.05
Li 0.97 0.03 0.02 0.03 0.00 0.00
Mg 0.27 0.75 0.12 0.03 0.01 0.05
Mn 0.06 0.04 0.91 0.03 0.03 0.08
Na 0.03 0.29 0.22 0.65 0.23 0.10
Ni 0.42 0.05 0.13 0.05 0.74 0.12
P 0.04 0.83 0.27 0.15 0.12 0.00
Pb 0.13 0.08 0.11 0.01 0.19 0.81
Sr 0.19 0.36 0.07 0.65 0.12 0.05
V 0.39 0.06 0.10 0.10 0.72 0.09
Zn 0.16 0.64 0.10 0.01 0.36 0.40
%
Explained
25.0 14.9 11.0 7.1 9.9 7.8
Prp total 0.25 0.15 0.11 0.07 0.09 0.07
Bolded values for separate element shows that the element belongs to
the corresponding factor group
ETAAS Electrothermal atomic absorption spectrometry; Prp. total
Total proportion
38 Arch Environ Contam Toxicol (2013) 65:3346
123
-
countries, Serbia and Slovenia. The data obtained from the
latest research in Norway in 2010 were used for compari-
son with a pristine area. These comparisons are listed in
Table 2. The Norwegian values were obtained by ICP
mass spectrometry method and are based on nitric acid
solutions (the same as in the present study), possibly
leaving out fractions of the elements contained in silicate
minerals (soil particles).
From data listed in Table 2, it can be seen that the
median values and ranges of all elements obtained in this
study are similar to the median values and ranges obtained
in the 2005 study (Spiric et al. 2012). Only a few elements
have slightly greater values for medians. The median value
for Cd is 1.4 times, for Cu 1.13 greater, for Mg 1.44, for Ni
1.17 times, and for Pb 1.3 times greater. For some typical
anthropogenic elements, such as Cr, Hg, V and Zn, lower
median values are recorded. For these elements in 2005,
median values of 2.8 mg, 0.064 mg, 3.1 mg, and
29 mg kg-1, respectively, are measured, whereas in the
present study, for the same elements, median values of
1.94, 0.043, 2.55, and 24.8 mg kg-1, respectively, are
measured. According to this comparison and the decrease
of median values for some important anthropogenic ele-
ments, it is obvious that the state of anthropogenic pollu-
tion in Croatia in the last 5 years has not changed
significantly even as the anthropogenic influence is
decreasing.
From the comparison of median values for those ele-
ments for which there is data from Serbian moss samples
(Harmens et al. 2008), it can be seen that the median values
for all heavy metals in Croatian moss samples are lower,
except for Cd (0.38 mg kg-1), which has a slightly higher
value. Comparing Croatian with Slovenian moss analyses
(Harmens et al. 2008), it can be concluded that these two
countries have similar median values for all elements. Only
Cd, Cu, and Ni have (insignificantly so) 1.15, 1.04, and
Fig. 3 Geographicaldistribution of factor 1 scores
(Al, Cd, Cr, Fe, and Li)
Arch Environ Contam Toxicol (2013) 65:3346 39
123
-
1.15 times greater median values in Croatian moss sam-
ples. It should be pointed out that the Croatian data refers
to sampling in 2010, whereas the data from the other two
Balkan countries are from the 2005 to 2006 moss survey,
and the situation in those countries may have changed in
the meantime.
The data for Norwegian moss samples refers to the
mosses taken from the latest study in 2010 (Steinnes et al.
2011). Norway, as pristine area, has lower median values
for typical air pollution elements (As, Cd, Cr, Cu, Ni, Pb,
and V), whereas Croatia has lower median values only for
Ba (20.64 mg kg-1), Hg (0.043 mg kg-1), and Zn
(24.8 mg kg-1).
Association of Chemical Elements
A multivariate statistical approach was adopted to assist the
interpretation of metal concentration data in moss samples
and to visualize the sites with greater metal content. Factor
analysis is a multivariate statistical technique commonly
used in atmospheric deposition and other environmental
studies to deduce source from data set (Garson 2000). With
PCFA, the characteristics of the 21 individual elements
were decreased to 6 synthetic variables (F1 to F6), which
account for 75.7 % of the total variability of treated ele-
ments (Table 3). The first three factors are reliable (they
have more than three significant loadings) and explain
50.9 % of total variability. The other three factors are less
reliable. Factors 4 and 5 have two significant loadings, and
factor 6 only has one significant loading. Together, they
explain 24.8 % of the total variability.
Factor 1
Factor 1 has high factor loadings for Al, Cd, Cr, Fe,
and Li and represents mixed (geogenicanthropogenic)
Fig. 4 Geographicaldistribution of factor 2 scores
(K, Mg, and P)
40 Arch Environ Contam Toxicol (2013) 65:3346
123
-
association of elements. It is the strongest factor repre-
senting 25 % of the total variability. The geographical
distribution of factor 1 is shown in Fig. 3. The regional
distribution of Al, Fe, and Li are typical for the group of
crustal elements predominantly supplied to the moss by
windblown soil dust; thus, the content is not solely due to
atmospheric deposition. Al compounds are insoluble, and
most of the Al found in biological systems comes from dust
contamination (Pais and Jones 1997). In contrast, high
values for Cd and Cr, especially around industrial areas
near Zagreb (sampling site no. 93B, Sisak [no. 104], Kutina
[no. 51], Osjek [no. 40], and Sibenik [no. 81A]), indicate
that the contents of these elements are the result of
anthropogenic sources. According to the Geochemical
Atlas of Croatia (Halamic and Miko 2009), soils of coastal
Croatia have greater contents of these two elements, and it
can be concluded that greater values for their content in
moss samples collected from Coastal region have a natural
origin. The reasons for higher Cd and Cr values near
Zagreb, Sisak, Kutina, and Osijek are probably due to
chemical and heavy-metal industries and near Sisak and
Sibenik are due to metallurgy.
Factor 2
Factor 2 (K, Mg, and P) is the second factor accounting for
14.9 % of total variability (Fig. 4). These elements are
typical for crustal material,and are significantly influenced
by soil particles attached to the moss samples. K, Mg, and
P are lithophile elements, and their increased contents are
typical for the northern part of the country in the regions of
central Croatia, Posavina, Podravina, and part of moun-
tainous Croatia. Igneous and metamorphic rocks rich in K
minerals are represented in the central (Slavonic moun-
tains) and western parts (Moslavacka, Medvednica, and
Ivanscica Mountains) of northern Croatia, which lead to
K-enrichment in the soil and moss samples (Halamic et al.
2012). In these regions, Mg content is a direct consequence
Fig. 5 Geographicaldistribution of factor 3 scores
(Ag, Ba, and Mn)
Arch Environ Contam Toxicol (2013) 65:3346 41
123
-
of the geological substrate (bedrock) because these areas
are built of Mesozoic dolomite (Halamic and Miko 2009).
In the Podravina region, high contents of K and Mg in moss
samples are a result of natural occurrence. Flood plain
sediments of the Drava River are composed of particles
originating from weathering of igneous and metamorphic
rocks in the Alps (Slovenia, Austria), which abound in K
minerals and from which K is deposited on moss samples.
The high content of ferromagnesian minerals in the
deposits of the Sava and Danube flood plains can explain
the high Mg contents in moss (Halamic and Miko 2009;
Halamic et al. 2012). In coastal Croatia, K content is
generally associated with siliciclastic lithology, namely,
turbidite deposits and micaceous sandstones (Halamic et al.
2012). Greater contents of P are registered in areas with
intensive agricultural production (central Croatia, Posavi-
na, Podravina, and coastal Croatia). P is introduced into
soils and mosses due to intensive agricultural production
and the use of P-based artificial fertilizer. P also can be
found in soils and moss samples as a result of geogenic
origin in heavy mineral fraction apatite (Halamic and Miko
2009).
Factor 3
Factor 3 has high loadings for Ag, Ba, and Mn. The dis-
tribution map is shown in Fig. 5. From the map of distri-
bution, it can be noted that the highest distribution of those
elements is focused in major industrial centres. According
to this, it can be concluded that this is an anthropogenic
factor. The highest values of Ag, Ba, and Mn are recorded
in the moss samples collected near Zagreb at sampling site
no. 94 and in Sisak at sampling sites no. 104 and 105. Ag
has the highest value in the sample collected at sampling
site no. 104 (0.115 mg kg-1). Ba has highest content in
Fig. 6 Geographicaldistribution of factor 4 scores
(Ca and Na)
42 Arch Environ Contam Toxicol (2013) 65:3346
123
-
the sample at sampling site no. 94 (928 mg kg-1), whereas
Mn has the highest content in the sample collected at
sampling site no. 105 (94.3 mg kg-1). These three ele-
ments probably occur in greater contents in moss samples
as a result of industrial activities, processing of ores, steel
refining, cement manufacture, and similar activities near
Zagreb, Sisak, Kutina, and some other industrially devel-
oped cities.
Factor 4
Factor 4 is a geogenic factor that is present with Ca and Na
(Fig. 6).Ca and Na are the main elements in the Earths
crust. Their contents in moss samples are mainly from
geogenic origin. Those elements are found in igneous rocks
(ultrabasites, basalts, and granites). They are also
commonly found in the form of carbonates. High contents
of Ca are registered in mosses collected in central Croatia
where carbonate rocks as bedrock are dominant. Here, high
contents of Na are found as a result of igneous and clastic
rocks that contain Na-rich minerals. Increased Ca contents
are found in samples collected near the Valley of Sava
River where flood sediments predominantly consist of
carbonate pebbles and fine-grained deposits, which are also
deposited on moss samples. The highest contents of Na are
found in the Podravina region where sediments are enri-
ched with minerals having high Na concentrations. The
coastal region contains the highest content of Ca, which is
associated with undeveloped soils on flysch bedrock
(Halamic and Miko 2009). High content of Na in moss
samples collected from the coastal region is probably a
result of the sea influence.
Fig. 7 Geographicaldistribution of factor 5 scores
(Ni and V)
Arch Environ Contam Toxicol (2013) 65:3346 43
123
-
Factor 5
Factor 5 shows high loadings for Ni and V (Fig. 7). It is
generally an anthropogenic factor but can also be of geo-
genic origin. High contents of these elements in the
northern parts of Croatia are of anthropogenic origin. Ni
and V may be associated with burning of heavy fuel oil for
heating and electricity production or with metal industry.
The two metals show sufficient covariation to support the
conclusion that combustion of heavy fuel oil is a significant
factor. Highest contents of Ni (14.66 mg kg-1) and V
(37.26 mg kg-1) are found in the moss sample collected
near Rijeka at sampling site no. 108 where a thermoelectric
plant is located, which is an anthropogenic source of these
elements. In the vicinity of developed industrialized cities,
such as Zagreb and Sisak, greater contents of these ele-
ments are also found. According to the Geochemical Atlas
of Croatia, it can be noted that high contents of Ni and V in
the regions of coastal and mountainous Croatia are found
as a result of geogenic influence. These elements are
associated with a complex of Palaeozoic siliciclastic rocks
(Halamic and Miko 2009).
Factor 6
Factor 6 is dominated by Pb and is an anthropogenic
factor (Fig. 8). Pb is a nonessential element, and it is
harmful. It is introduced into the environment by leaded
gasoline, mining and smelting activities, coal, and waste.
The highest content of Pb (36.64 mg kg-1) in moss is
found at sampling site no. 98, near Zagreb, as a result of
anthropogenic influence of the developed industries in this
region. In the Drava River valley, high contents of this
element are also found as a consequence of Pb ore deposits
situated more upstream (Austria, Slovenia) where intense
mining activity has existed for the last two centuries
Fig. 8 Geographicaldistribution of factor 6 scores
(Pb)
44 Arch Environ Contam Toxicol (2013) 65:3346
123
-
(Bleiberg, Mezica) (Sajn et al. 2011). High contents of Pb
are also found in moss samples collected from areas with
developed industry (Sisak, Kutina, Rijeka, Split) as a
result of anthropogenic activities.
Conclusion
In this study, moss biomonitoring technique was applied
for investigation of air pollution in Croatia. The content of
21 elements was determined in 121 moss samples. Statis-
tical analyses were performed on the obtained results. With
PCFA, the characteristics of 21 individual elements were
decreased to 6 synthetic variables (F1 to F6), of which 2
are anthropogenic (F3 and F6 [Ag, Ba, Mn, and Pb]), 2 are
mixed geogenic-anthropogenic (F1 and F5 [Al, Cd, Cr, Fe,
Li, Ni, and V[), and 2 are geogenic (F2 and F4 [K, Mg, P,
Ca, and Na]). By comparing the results of this study with
those obtained in 2005, it can be concluded that the air
pollution, along with some of the anthropogenic elements,
in Croatia has not significantly changed. Compared with
neighboring countries, Croatia is less polluted than Serbia
and has similar values of the content of typical anthropo-
genic elements as those obtained in the investigation in
Slovenia. By comparing with the data obtained in the latest
study in Norway as the least anthropogenically polluted
area of Europe, it may be noted that Croatia is negligible
polluted. The main anthropogenic factors are highly
developed light and heavy industry, transportation, steel
industry, textile industry, thermoelectric plant, and oil
deposits, which activities are performed near large indus-
trialized cities, such as Zagreb, Sisak, Kutina, Split, Rijeka,
and Sibenik. Moss samples collected near these regions
showed the highest content of heavy metals typical for air
pollution of anthropogenic derivation.
References
Balabanova B, Stafilov T, Baceva K, Sajn R (2010) Biomonitoring of
atmospheric pollution with heavy metals in the copper mine
vicinity located near Radovis, Republic of Macedonia. J Environ
Sci Heal A 45:15041518
Barandovski L, Frontasyeva MV, Stafilov T, Sajn R, Pavlov S,
Enimiteva V (2012) Trends of atmospheric deposition of trace
elements in Macedonia studied by the moss biomonitoring
technique. J Environ Science Health A 47:116
Bargagli R, Brown DH, Nelli L (1995) Metal biomonitoring with
moss: Procedures for correcting for soil contamination. Environ
Pollut 89(2):169175
Berg T, Steinnes E (1997) Use of mosses (Hylocomium splendens and
Pleurozium schreberi) as biomonitors of heavy metal deposition:
From relative to absolute values. Environ Pollut 98:6171
Buse A, Norris D, Harmens H, Buker P, Ashenden T, Mills G (2003)
European atlas: Heavy metals in European mosses: 2000/2001
Survey. University of Wales, Bangor, UK, UNECE ICP
Vegetation. Centre for Ecology & Hydrology
Figueira R, Sergio C, Sousa AJ (2002) Distribution of trace metals in
moss biomonitors and assessment of contamination sources in
Portugal. Environ Pollut 118:153163
Garson D (2000) Factor analysis. PA 765: Quantitative research in
public administration, North Carolina State University
Gjengedal E, Steinnes E (1990) Uptake of metal ions in moss from
artificial precipitation. Environ Monit Assess 14:7787
Gramatica P, Battaini F, Gianti E, Papa E, Jones R, Preatoni D et al
(2006) Analysis of mosses and soils for quantifying heavy metal
concentrations in Sicily: A multivariate and spatial analytical
approach. Environ Sci Pollut Res Int 13:2836
Halamic J, Miko S (eds) (2009) Geochemical atlas of the Republic of
Croatia. Croatian Geological Survey, Zagreb, Croatia
Halamic J, Peh Z, Miko S, Galovic L, Sorsa A (2012) Geochemical
atlas of Croatia: Environmental implications and geodynamical
thread. J Geochem Explor 115:3646
Harmens H (2009) Heavy metals in European mosses: 2010 survey-
monitoring manual. ICP Vegetation Coordination Centre, Centre
for Ecology and Hydrology, Bangor, UK
Harmens H, Norris D, and the participants of the Moss Survey (2008):
Spatial and temporal trends in heavy metal accumulation in mosses
in Europe (1990-2005) Programme Coordination Centre for the
ICP Vegetation, Centre for Ecology and Hydrology, Bangor
Horvat M, Jeran Z, Spiric Z, Jacimovic R, Miklavcic V (2000)
Mercury and other elements in lichens at INA-naftaplin gas
treatment plant, Molve, Croatia. J Environ Monit 2:139144
Johnson RA, Wichern DW (2002) Applied multivariate statistical
analyses (5th ed). Prentice Hall, Upper Saddle River
Mulgrew A, Williams P (2000) Biomonitoring of air quality using
plants. Air Hygiene Report 10. WHO Collaborating Centre for
the Air Quality Management and Air Pollution Control, Berlin
Pais I, Jones JB (1997) The handbook of trace elements. Lucie Press,
Boca Raton, FL, St
Pesch R, Schroeder W (2006) Mosses as bioindicators for metal
accumulation: Statistical aggregation of measurement data to
exposure indices. Ecol Indic 6:137152
Puckett KJ (1988) Bryophytes and lichens as monitors of metal
deposition. In: Nash TH III, Wirth V (eds) Lichens, bryophytes and
air quality, 30. Bibliotheca Lichenologica, Berlin, pp 231267
Ruhling A (2002) A European survey of atmospheric heavy metal
deposition in 20002001. Environ Pollut 120(1):2325
Ruhling A (1994) Atmospheric heavy metal deposition in Europe
Estimations based on moss analyses. Nord 9. Nordic Council of
Ministers, Copenhagen
Ruhling A, Steinnes E (1998) Atmospheric heavy metal deposition in
Europe 19951996. Nord 15. Nordic Council of Ministers,
Copenhagen, Denmark
Ruhling A, Tyler G (1971) Regional differences in the heavy metal
deposition over Scandinavia. J Appl Ecol 8:497507
Ruhling A, Rasmussen L, Pilegaard K, Makinen A, Steinnes E (1987)
Survey of atmospheric heavy metal deposition in the Nordic
countries in 1985Monitored by moss analyses. Nord 21.
Nordic Council of Ministers, Copenhagen, Denmark
Sajn R, Halamic J, Peh Z, Galovic L, Alijagic J (2011) Assessment of
the natural and anthropogenic sources of chemical elements
in alluvial soils from the Drava River using multivariate statis-
tical methods. J Geochem Explor 110(3):278289
Serafimovski I, Karadjova I, Stafilov T, Cvetkovic J (2008) Deter-
mination of inorganic and methylmercury in fish by cold vapor
atomic absorption spectrometry and inductively coupled plasma
atomic emission spectrometry. Microchem J 89:4247
Spiric Z, Frontasyeva M, Steinnes E, Stafilov T (2012) Multi-element
atmospheric deposition study in Croatia. Int J Environ Anal
Chem 92(10):12001214
Arch Environ Contam Toxicol (2013) 65:3346 45
123
-
Steinnes E, Berg T, Uggerud TH, Pfaffhuber KA (2011) Atmospheric
deposition of heavy metals in Norway, National Study in 2010,
Rapport 1109/2011 [in Norwegian]. Norwegian University of
Science and Technology, Trondheim
Stevens J (2002) Applied multivariate statistics for the social sciences
(4th ed.). Lawrence Erlbaum Associates, Mahwah, NJ
Thoni L, Yurukova L, Bergamini A, Ilyin I, Matthaei D (2011)
Temporal trends and spatial patterns of heavy metal concentra-
tions in mosses in Bulgaria and Switzerland: 19902005. Atmos
Environ 45:18991912
46 Arch Environ Contam Toxicol (2013) 65:3346
123
-
Reproduced with permission of the copyright owner. Further reproduction prohibited withoutpermission.
c.244_2013_Article_9884.pdfAir Pollution Study in Croatia Using Moss Biomonitoring and ICP--AES and AAS Analytical TechniquesAbstractStudy AreaMaterials and MethodsSamplingSample PreparationInstrumentationStatistical Analysis and Mapping
Results and DiscussionAssociation of Chemical ElementsFactor 1Factor 2Factor 3Factor 4Factor 5Factor 6
ConclusionReferences