J Atmos Chem (2007) 58:167179

DOI 10.1007/s10874-007-9085-7

Chemical composition of wet deposition

in a Mediterranean site Athens, Greece
related to the origin of air masses
Panagiotis T. Nastos & Dimitrios Alexakis &
Helen A. Kanellopoulou & Akindinos E. Kelepertsis

Received: 29 March 2007 / Accepted: 17 August 2007 /

Published online: 21 September 2007
# Springer Science + Business Media B.V. 2007

Abstract The goal of this study is to determine the chemical composition of rain, in the
wider region of Athens, Greece for the time period 1st September 2001 to 31st August
2002. Two model automatic rain samplers were installed in the Meteorological Station of
Laboratory of Climatology (latitude: 37 58 N, longitude: 23 47 E) inside the Athens
University Campus and in a site at Heraklio Attica, a northern suburb of Athens (latitude:
38 03 N, longitude: 23 45 E). The concentrations (eq l1) of the major cations (H+,


2

Na+, K+, Ca2+ and Mg2+) and anions (Cl, NO

3 , HCO3 SO4 ), as well as pH and
conductivity of rain in 39 total samples were determined. The figures of pH range from 6.4
to 8.4 and conductivity from 8 to 207 S cm1. The analysis showed that Ca2+ ions are
2

present the highest
abundant within all examined samples, while HCO

3 and SO4
concentrations from the anions. In order to find out the origin of the air masses, the air mass
back trajectories were calculated. Five sectors of the origin of air masses were revealed: the
North, the South, the Local, the West and the East sector. Multivariate methods included
Factor Analysis and Discriminant Analysis were applied to the examined ion concentrations
and three main factors were extracted, which discriminated the ions according to their
origin. The first group of ions is interpreted as the result of the anthropogenic activity, the
second group represents the acidityalkalinity independently of their source and the third
one the marine influence.
P. T. Nastos : H. A. Kanellopoulou
Laboratory of Climatology & Atmospheric Environment, University of Athens, Panepistimiopolis,
GR 157 84 Athens, Greece
D. Alexakis : A. E. Kelepertsis
Laboratory of Geochemistry & Economic Geology, University of Athens, Panepistimiopolis,
GR 157 84 Athens, Greece
P. T. Nastos (*)
Laboratory of Climatology and Atmospheric Environment, Department of Geography and Climatology,
Faculty of Geology and Geoenvironment, University of Athens, Panepistimiopolis,
GR 157 84 Athens, Greece
e-mail: [email protected]

168

J Atmos Chem (2007) 58:167179

Keywords Air mass back trajectories . Discriminant analysis . Factor analysis .

Major anions . Major cations . Wet deposition

1 Introduction
The phenomenon of acid rain presents great scientific interest, and is well known that it is
one of the important environmental problems and for this reason a lot of studies and
analyses have been carried out concerning the emissions, the way of transport,
transformation and deposition of air pollutants which causes acid rain (Beilke 1983; CEC
1985; Pierson and Chang 1986). Particularly in the central-eastern Mediterranean, acid
episodes (25%, pH<5), and a great percentage (>40%) of episodes with pH>6 appeared
(Glavas 1988; Dikaiakos et al. 1990; Samara et al. 1992; LeBolloch and Guerzoni 1995;
Tuncel and Ungoer 1996; Guelsoy et al. 1999). Also acid and alkaline rains have been
reported in Israel (Mamane and Gottlieb 1995), in Spain (Camarero and Catalan 1993;
Avila and Alarcon 1999; Hontoria et al. 2003) and in southern France (Loye-Pilo and
Morelli 1988). In Greece, the study of the phenomenon of acid rain began 25 years ago,
approximately, when some researchers showed scientific interest for the reasons responsible
for the observed deterioration of the marbles of Parthenon (Vocom 1979; Skoulikidis 1983;
Cheng et al. 1987). Then, a lot of researchers studied systematically the phenomenon of
acid rain in Athens (Dikaiakos and Nastos 1987; Dikaiakos et al. 1990; Smirnioudi and
Siskos 1992; Kelepertsis et al. 2002; Kita et al. 2004; Nastos et al. 2004), in Thessalonica
(Samara et al. 1992), in Patras (Glavas 1988; Glavas and Moschonas 2002), in Crete
(Mihalopoulos et al. 1997) and in central Greece (Tsikritsis 2006).
This work presents the results of the chemical composition of the wet deposition, with
respect to the concentrations of major anions and cations, in two locations in the wider
region of Athens, for the period September 2001August 2002 and aims in finding out the
sources of the ions using multivariate statistical methods and air mass back trajectories.

2 Data and analysis

The collection of the rain samples was achieved using prototype automatic samplers, the
specifications of which have been reported in previous studies, (Dikaiakos and Nastos
1987; Dikaiakos et al. 1990). The automatic samplers had been installed in the
Meteorological Station of Laboratory of Climatology inside the Athens University Campus
(latitude: 37 58 N, longitude: 23 47 E) and in a site at Heraklio, Attica (latitude: 38 03 N,
longitude: 23 45 E). Figure 1 depicts the locations of the two rain samplers. The collection
of 39 rain samples occurred on 24 hours basis (end of procedure at 08:00 am, local time)
during the period from September 2001 to August 2002 and concerns 17 common in both
stations and 5 isolated in the one station, precipitation events (Table 2).
The samples of rain were kept in polyethylene bottles (Moody and Lindstrom, 1977) and
forwarded in the Laboratory for the chemical analyses. The preparation of samples and the
chemical analyses, took place in the Laboratory of Geochemistry of the University of Athens.
The measurements of pH (pH meter Hanna HI 8424) and conductivity (Hanna HI 9033
multi-range conductivity meter) were done at once after the sample collection at 8 o clock
in the morning. The pH meter was calibrated in the range 47 before every use. In the
process, every rain sample was kept in the refrigerator (4C) for the chemical analyses
forward.

J Atmos Chem (2007) 58:167179

169

The rainwater samples were vacuum filtered through 0.45 m pore size membrane
filters. A part of each sample was vacuum filtered with the same membrane filters and
acidified to a final concentration of about 1% nitric acid. In the process, every rainwater
sample was kept in the refrigerator (4C) for the chemical analyses forward. Sulphate,
nitrate and chloride were measured photometrically using a Hach DR/2000 apparatus.
Bicarbonate was measured using a Hach alkalinity titrator. Calcium and magnesium were
measured using an Atomic Absorption Spectrophotometer Perkin Elmer 1100B while
sodium and potassium were measured using a flame photometer Jenway PFP7.
The analyses applied to the data under consideration were the Factor Analysis (FA), the
Discriminant Analysis (DA) and the back trajectories analysis which are briefly discussed
below:
The main applications of FA are to reduce the number of variables and to detect structure
in the relationships between variables, that is to classify variables. Therefore FA is applied
as a data reduction or structure detection method. Therefore each of the p initial variables
X1, X2, ..., Xp can be expressed as a linear function of m (m<p) uncorrelated factors: Xi =
ai1F1 + ai2F2 + ... + aimFm (i=1,..., p) where F1, F2, ... ,Fm are the factors and ai1, ai2, ... ,
aim are the factor loadings which express the correlation between the factors and the initial
variables. The values of each factor are called factor scores and they are presented in
standardized form, having zero mean and unit variance (Jolliffe 1986; Manly 1986). The
number m of the retained factors has to be decided, by using various rules (eigenvalue 1,

Fig. 1 Locations of the automated rainwater samplers

170

J Atmos Chem (2007) 58:167179

Table 1 Statistical figures for pH and conductivity

University Campus
Heraklio, Attica
University Campus
Heraklio, Attica

pH
Conductivity(S cm1)

Mean

Std. Dev.

Minimum

Maximum

7.4
7.4
63.4
88.4

0.4
0.5
42.0
60.4

6.4
6.5
10.0
8.0

8.4
8.1
137.0
207.0

scree plot) and considering the physical interpretation of the results (Bartzokas and Metaxas
1993). Another important point of the analysis is the rotation of the axes, which maximizes
some factor loadings and minimizes some others and in that way a better separation among
the initial variables is succeeded. Varimax rotation is generally accepted as the most
accurate orthogonal rotation, which maximizes the sum of the variances of the square factor
loadings, keeping the factors uncorrelated (Richman 1986).
Discriminant analysis is useful for situations where the subject of research is to build a
predictive model of group membership based on observed characteristics of each case. The
procedure generates a discriminant function (or, for more than two groups, a set of
discriminant functions) based on linear combinations of the predictor variables that provide
the best discrimination between the groups. The functions are generated from a sample of

Fig. 2 Histograms of pH and conductivity for the two sampling sites

J Atmos Chem (2007) 58:167179

171

Fig. 3 Conductivity as a
function of pH for all rainwater
samples

cases for which group membership is known; the functions can then be applied to new
cases with measurements for the predictor variables but unknown group membership.
The application of multivariate analyses was performed by using SPSS for Windows,
version 11. All the multivariate methods were applied to the log10 converted data (Glavas
and Moschonas 2002).
First and foremost, using Factor Analysis, three main factors with eigenvalues >1 were
evaluated. In the process, the application of Discriminant Analysis to the factor scores
extracted by the Factor Analysis resulted in the best discrimination of the main factors
between the sectors, which were defined by the analysis of air mass back trajectories. For
the calculation of the back trajectories analysis the HYSPLIT 4 model of Air Resources
Laboratory of NOAA (Draxler and Hess 1997, 1998) is used for two different levels: 1500
and 3000 m (a.m.s.l.).

3 Results and discussion

Table 1 presents the statistical characteristics of pH and conductivity, for both locations.
The pH figures are found in agreement with the respective findings of others researchers
(Glavas 1988; Samara et al. 1992; Dikaiakos et al. 1990; Smirnioudi and Siskos 1992;
Kelepertsis et al. 2002; Kita et al. 2004).
It is well known that a water solution containing 325 ppm CO2, which is approximately
the concentration of CO2 in the atmosphere, presents a typical pH value of approximately
5.6. Rainfall pH figures smaller than the threshold of 5.6 are associated with the effect of
natural (such as volcanic emissions) or anthropogenic (such as central heating,
transportation and industry emissions) source of pollution. With respect to the examined
period (1/9/200131/8/2002), episodes of acid rain (pH<5.6) were not recorded in both the
two sites of observation. The frequency distribution of pH (Fig. 2) shows that the class 7.5
8.0 is the prevailing one (39% for University Campus and 38% for Heraklio), meaning a
shift to alkalinity compared to the respective frequency distribution for the previous studied
period (1/9/200031/8/2001), where the class 7.07.5 was the predominant regarding 40
rain samples (Kelepertsis et al. 2002). With respect to conductivity histograms (Fig. 2), we
observe that the class 050 S cm1, dominates (50% for University Campus and 38% for
the Heraklio).

172

J Atmos Chem (2007) 58:167179

Table 2 Mean monthly and annual figures of pH, total acidity (eq l1), conductivity (S cm1) and
concentrations of major anions and cations (eq l1), along with the corresponding standard deviations, in
the collected rainwater samples at Heraklio, Attica and at University Campus of Athens
Month

11/01

12/01

01/02

02/02

03/02

04/02

05/02

07/
02

08/
02

Annual

Number of rainwater
samples
Rainfall (mm) at the
University Campus
pH
[H+]Total
Conductivity

12

10

39

154.0

85.4

9.6

7.2

58.8

26.8

35.0

32.0 0.8

7.30.5
0.60.5
60.8
48.4
117.4
158.5
28.3
19.4
153.1
137.0
158.3
108.9
123.8
111.8
23.2
5.8
345.8
267.9
33.0
27.2

7.20.2
0.60.4
51.9
32.0
175.0
180.1
36.1
21.0
189.8
91.1
126.5
68.9
124.3
83.5
21.8
3.3
278.5
181.7
32.9
17.9

7.40.3
0.70.2
70.0
41.0
302.1
250.4
46.1
15.0
182.8
86.9
135.2
63.7
137.0
52.3
25.4
4.0
376.2
284.6
45.8
8.2

8.00.1
0.80.0
179.0
17.0
854.2
206.2
106.8
9.6
563.9

7.70.1
0.30.5
69.5
46.7
281.2
270.8
49.0
36.0
348.4
95.9
112.7
23.0
67.4
94.4
21.2
6.9
662.5
379.4
33.3
34.0

7.80.4
0.50.1
152.5
31.8
781.2
132.6
244.8
155.6
854.1
252.7
169.0
39.8
104.4
12.3
43.7
2.0
1850.0
495.0
76.7
21.2

7.4 6.5 7.40.5

0.6 0.2 0.60.4
136.0 43.0 76.9
53.5
479.2 187.5 246.7
265.1
198.4 127.7 59.6
63.5
563.9
287.0
212.1
84.5 22.5 141.3
79.7
130.4 39.1 118.1
83.4
56.4 18.0 25.4
8.6
655.0 210.0 527.0
506.4
100.0 45.0 40.7
26.6

 2

SO4


NO3


HCO

3
[Cl]
[Na+]
[K+]
[Ca2+]
[Mg2+]

7.70.6
0.60.2
97.8
62.4
133.3
95.0
61.3
21.0
350.8
102.8
147.9 180.3
10.0
88.2
167.4 124.4
3.1
51.4
36.4
25.0
4.4
1.4
1661.2 465.7
723.0
96.3
79.2
35.7
17.7
14.7

409.6

The conductivity of all analyzed samples estimated 63.442.0 S cm1 for University
Campus and 88.460.4 S cm1 for Heraklio, (Table 1). The conductivity is positively
correlated (r=0.67, p<0.05) with pH, concerning the total of rain samples for the two cites
(Fig. 3). This relationship is expressed by the linear model: y=498.8 + 77.8 x and it can be
attributed to the neutralization of acids in the rain by ammonia (from fertilizers), marine
water and dust from the ground. Neutral rains in Athens must be caused by the
counteraction of artificially and naturally acidified rains with calcium carbonate in the
dust carried not only from the urban area of Athens by local winds but from arid areas in
the world by global atmospheric currents, as well (Kita et al. 2004).


Table 3 The ratios of mean annual concentrations of major ions Cl, K+, Mg2+, Ca2+ and SO2

to Na+, in
4
the rainwater samples and seawater
Ions ratio

[Cl ]/[Na ]
[K+]/[Na+]
[Mg2+]/[Na+]
2+
+
[Ca
 2
]/[Na
 ]
SO4
Na 

Rainwater samples

Seawater (Galloway et al. 1983)

1.19
0.22
0.34
4.46
2.09

1.16
0.02
0.23
0.04
0.12

J Atmos Chem (2007) 58:167179

173

Fig. 4 Annual march of the surface concentrations (mg m2 month1) for the elements S, N, Cl, Na, K, Ca
and Mg, for the two sampling locations

174

J Atmos Chem (2007) 58:167179

Fig. 5 Forty-eight hours air mass back trajectories for the height 1500 m (850 hPa) ending at the following
specific time and dates. The numbers stand for the corresponding rain event and the letter in parenthesis
defines the sector (N for North, S for South, W for West, E for East and L for Local): #1(N) at 07 UTC, 03
November 2001; #2(E) at 08 UTC, 04 November 2001; #3(N) at 22 UTC, 05 November 2001; #4(L) at 17
UTC, 19 November 2001; #5(S) at 01 UTC, 20 November 2001; #6(W) at 08 UTC, 23 November 2001; #7
(W) at 09 UTC, 24 November 2001; #8(S) at 20 UTC, 28 November 2001; #9(N) at 12 UTC, 01 December
2001; #10(N) at 22 UTC, 02 December 2001; #11(N) at 22 UTC, 07 December 2001; #12(N) at 05 UTC, 09
December 2001; #13(N) at 07 UTC, 13 December 2001; #14(S) at 05 UTC, 16 December 2001; #15(N) at
06 UTC, 21 December 2001; #16(S) at 22 UTC, 17 January 2002; #17(L) at 10 UTC, 11 February 2002; #18
(S) at 13 UTC, 20 February 2002; #19(S) at 23 UTC, 10 March 2002; #20(N) at 22 UTC, 15 March 2002;
#21(N) at 04 UTC, 16 March 2002; #22(L) at 00 UTC, 16 April 2002; #23(L) at 08 UTC, 18 April 2002;
#24(W) at 09 UTC, 21 May 2002; #25(N) at 09 UTC, 08 July 2002; #26(W) at 04 UTC, 30 August 2002

Another parameter that was calculated is the total acidity of rain [H+]Total (Galloway et al.
1983):
H Total H pH Alkss
where [H+]pH represents the free acidity of rain ([H+]pH =10pH) and [Alk]ss is the alkalinity
from the marine salt and is given by the formula:
Alkss 0:00484Na ppt

J Atmos Chem (2007) 58:167179

175

Table 4 pH, conductivity (S cm1) and mean concentrations of major anions and cations (eq l1), for
every section estimated from air mass trajectories analysis
Section

Cases

pH

Conductivity

SO42

NO3

HCO3

Cl

Na+

K+

Ca2+

Mg2+

North
South
Local
West
East

16
11
5
5
2

7.2
7.6
7.3
7.6
7.6

55.4
84.6
91.6
125.4
48.0

173.2
204.2
370.8
483.3
145.8

45.2
48.4
59.5
144.8
24.2

212.6
298.6
218.9
706.0
344.3

121.7
135.0
156.1
205.1
126.8

108.1
112.8
136.2
183.3
19.6

24.2
25.2
25.6
33.0
16.7

342.5
462.1
724.3
1073.5
500.0

34.1
38.0
44.7
73.0
16.7

where [Na+]ppt is the sodium concentration in the rain samples expressed in (eq l1). The
figures of total acidity range from 0.2 to 2.0 (eq l1).
Table 2 presents the mean monthly and annual figures of pH, total acidity (eq l1),
conductivity (S cm1) and concentrations of major anions and cations (eq l1), with the
corresponding standard deviation, for the total of the rain samples during the investigated
period (September 2001August 2002).
As it was mentioned before, neutral or even alkaline pH figures show that the main
process in the rain is the neutralization. The ratio of the mean annual concentrations (for 39
samples) of chloride to sodium (Table 3) is 1.19, and this compared to the corresponding
ratio 1.16 existed in marine water (Galloway et al. 1983), drives to a conclusion that the
chemical composition of rain is influenced more by the sea. Besides, the marine origin of
magnesium and sodium ions is proved by their ratio, which is 0.34, (0.23 in marine water).
Taking into consideration the ratios of ions determined in the rain samples and in the
seawater (Galloway et al. 1983), it is obvious that the marine water is the main source for
the Cl, Na+ and Mg2+, while the origin of the ions K+, Ca2+, SO2

4 is associated with other
2

sources. Also, the results of the analysis showed that the concentration
ofSO
 2

 4
ions



exceeded the concentration of NO3 ions in all samples and their ratio SO4
NO3 in
the rain was calculated equal to 4.14. This finding is in agreement with Tsikritsis (2006),
who concluded that rainwater acidity in Larissa (Greece) is mainly due to H2SO4 and
secondary to HNO3.
Figure 4 depicts the annual course of surface concentrations (mg m2 month1) for
sulphur, nitrogen, chloride, sodium, potassium, calcium and magnesium for both the two
sampling locations. It is remarkable that a very high concentration of the aforementioned
elements is evident during July, very likely due to the accumulation of pollutants in the

Table 5 Factor loadings

extracted from factor analysis
with varimax rotation of the axes

Factors
1
 2

SO4

NO3

HCO3
Cl
Na+
K+
Ca2+
Mg2+
pH

0.8
0.8
0.8
0.9
0.8
0.9
0.7
0.8

0.6
0.9

176

J Atmos Chem (2007) 58:167179

Fig. 6 Scatter plot of rain sample factor scores, derived by Factor analysis, for the sectors extracted by air
mass back trajectories

atmosphere combined with the absence of rainfalls for a long time period before. Another
explanation is the transboundary air pollution which is driven by the northeastern flow of the
Etesian winds established in the Aegean sea during summer time. According to OSullivan
(1989) the percentage of sulphur deposition in Greece from other countries is 63%.
Further to, the effect of the origin of air masses on the configuration of the chemical
composition of rain is investigated. For this reason, the 48 hours air mass back trajectories
for the 26 different days of sample collection were calculated, using the HYSPLIT 4 model
of Air Resources Laboratory of NOAA and are depicted in Fig. 5. Two different levels were
used: 1500 and 3000 m (850 and 700 hPa, respectively). In 9 cases the air masses come
from different directions in the two examined trajectory heights. Based on the results of the
air mass back trajectories analysis we distinguished 5 sectors according to the origin of the
air masses. Trajectories coming from the north (NWNE Europe) consist the northern
sector (N), from Northern Africa, the southern sector (S), from the grid box defined by the
coordinates 1828 E, 3542 N, the local sector (L), from the Western Mediterranean,
the western sector (W) and finally from the Asia Minor and Cyprus, the eastern sector (E).
If the trajectory remained within the mentioned grid box for 48 hours it was characterized
as local.
The mean chemical concentration for each sector, calculated by the air masses back
trajectory analysis, is appeared in Table 4. We observe low mean ions concentrations with
respect to the northern sector and high to the western sector, while intermediate
concentrations are associated with the southern and local sectors. Regarding the eastern

J Atmos Chem (2007) 58:167179

177

Fig. 7 Scatter plot of rain sample discriminant scores, derived by discriminant analysis

+
+
2+
sector, the concentrations of the ions NO

present the smallest figures
3 , Na , K and Mg
from all the other sectors, while pH and conductivity are lower in the northern sector.
In the process, FA with varimax rotation of the axes was applied, to the log10 converted
data and three main factors (eigenvalue 1) were extracted, which explain 90% of the total
variance, (Table 5). The first main factor that interprets 61% of the total variance is strongly
+
2+


associated with the ions SO2

4 , NO3 , K and Mg , and implies mainly the anthropogenic
2+
and pH and for this
contribution. The second factor (17%) comprises the ions HCO

3 , Ca
reason represents the alkalinityacidity factor. The third factor (12%) is strongly correlated
with the ions Cl and Na+, as well as moderately correlated with Mg2+ and stands for the
marine origin of pollutants. Similar conclusions are reported in the study of Glavas and
Moschonas (2002).
The plot of factor scores (the coefficients of each rain sample within each extracted main
factor) of each rain sample, classified to sectors according to the back trajectories analysis is
depicted in Fig. 6. The ordinate of the graph represents the anthropogenic factor, which
explains 61% of the total variance while the abscissa is the acidityalkalinity factor,
accounted for 17% of the total variance. The north sector is well distinguished from the
other sectors along the acidityalkalinity axis.
Better results are taken out by the application of DA on the factor scores extracted by the
FA. Discriminant function 1 accounted for 75.5% of the total variance is strongly correlated
with acidityalkalinity, discriminant function 2 accounted for 18.2% of the total variance is
strongly correlated with the ions due to anthropogenic activities and discriminant function 3
accounted for 6.3% is also strongly correlated with ions influenced by marine origin.
Figure 7 exhibits the results regarding the first two out of three discriminant functions
extracted. The first function (75.5%) is the abscissa and represents the acidityalkalinity
while the second function (18.2%) is the ordinate and stand for the anthropogenic activities.

178

J Atmos Chem (2007) 58:167179

Four groups distinguished by the same symbols as in Fig. 6 correspond to the sectors
obtained by the air masses trajectory analysis. The east sector is not appeared in the graphs
because of its small contribution (only two rain samples). The northern and southern sectors
present the greater dispersion along the axis of alkalinityacidity. The southern and local
sectors are depicted between the northern and the western sector, as they were expected,
accordingly to the trajectory analysis of the air masses. (Table 4).

4 Conclusions
The application of multivariate analysis to the concentrations of major anions (Cl, NO

3,
+
+
+
2+
2+
2

and
SO
)
and
cations
(H
,
Na
,
K
,
Ca
and
Mg
)
showed
that
the
chemical
HCO

3
4
composition of rain in Athens is determined by three main factors, which explain 90% of
+


the total variance. The first factor (61%) is associated with the ions SO2

4 , NO3 , K and
2+
Mg , and implies mainly the anthropogenic contribution. The second factor (17%) is
2+
and the pH and for this reason is the so-called
correlated with the ions HCO

3 , Ca
alkalinityacidity. The third factor (12%) is correlated with the ions Cl and Na+, as well as
Mg2+ and represents the marine origin of pollutants. Another finding of our research is that
the frequency distribution reveals a shift towards alkalinity in the rain sample from the
previous year to the year of our study. Finally, the back trajectories analysis showed that the
air masses coming from NWNE Europe (northern sector) are associated with less ion
concentrations in the rain water, while Western Mediterranean (western sector) is
responsible for more ion concentrations.
Acknowledgements This research is financially supported by the Special Account for Research Grants of
the University of Athens.

References
Avila, A., Alarcon, M.: Relationship between precipitation chemistry and meteorological situations at a rural
site in NE Spain. Atmos. Environ. 33, 16631677 (1999)
Bartzokas, A., Metaxas, D.A.: Covariability and climatic changes of the lower-troposhere temperatures over
the Northern Hemisphere. Nuovo Cim. 16, 359373 (1993)
Beilke, P.: Acid deposition. The present situation in Europe. Technical Report EUR 8307, Comm. Eur.
Communities, 330 (1983)
Camarero, L., Catalan, I.: Chemistry of bulk precipitation in the central and eastern Pyrennes, northeast
Spain. Atmos. Environ. 27A, 8394 (1993)
CEC: Report on the actions of the Commission of the EEC on acid deposition. EUR 9985 EN., 157 (1985)
Cheng, R.I., Hwu, I.R., Kim, J.T., Leu, P.M.: Deterioration of marble structures. The role of acid rain. Anal.
Chem. 59(2), 104A106A (1987)
Dikaiakos, J.G., Nastos, P.T.: Chemical analysis of rain in Athens. Publication of the Laboratory of
Climatology and Atmospheric Environment, University of Athens, No 23 (1987)
Dikaiakos, I.G., Tsitouris, C.G., Siskos, P.A, Melissos, D.A., Nastos, P.T.: Rainwater composition in Athens,
Greece. Atmos. Environ. 24B(1), 171176 (1990)
Draxler, R.R., Hess, G.D.: Description of the hysplit-modelling system, p. 24. NOAA Technical
Memorandum ERL Alp-224 (December 1997)
Draxler, R.R., Hess, G.D.: An overview of the hysplit-modelling system for trajectories, dispersion and
deposition. Aust. Meteorol. Mag. 47(4), 295308 (1998)
Galloway, I.N., Knap, A.X., Church, T.M.: The composition of western Atlantic precipitation using
shipboard collectors. J. Geophys. Res. 88, 1085910864 (1983)

J Atmos Chem (2007) 58:167179

179

Glavas, S.: A wet only precipitation study in a Mediterranean site, Patras, Greece. Atmos. Environ. 22,
15051507 (1988)
Glavas, S., Moschonas, N.: Origin of observed acidicalkaline rains in a wet-only precipitation study in a
Mediterranean coastal site, Patras, Greece. Atmos. Environ. 36, 30893099 (2002)
Guelsoy, G., Tayanc, M., Ertuerk, F.: Chemical analysis of the major ions in the precipitation of Instanbul,
Turkey. Environ. Pollut. 105, 273280 (1999)
Hontoria, C., Saa, A., Almorox, J., Cuadra, L., Sanchez, A., Gasco, J.M.: The chemical composition of
precipitation in Madrid. Water Air Soil Pollut. 146(14), 3554 (2003)
Jolliffe, I.T.: Principal Component Analysis. Springer-Verlag, New York (1986)
Kelepertsis, A.E., Nastos, P.T., Alexakis, D.E., Kanellopoulou, E.A.: Chemical analysis of rain in Athens. In:
Proceedings of the 6th Hellenic Congress of Meteorology, pp. 2528. Climatology and Atmospheric
Physics, Ioannina, Greece (September 2002)
Kita, I., Sato, T., Kase, Y., Mitropoulos, P.: Neutral rains at Athens, Greece: a natural safeguard against
acidification of rains. Sci. Total Environ. 327, 285294 (2004)
LeBolloch, O., Guerzoni P.: Acid and alkaline deposition in precipitation on the western coast of Sardinia,
central Mediterranean. Water Air Soil Pollut. 85, 21552160 (1995)
Loye-Pilo, M.D., Morelli, J.: Fluctuations of ionic composition of precipitations collected in Corsica related
to changes in the origin of incoming aerosols. J. Aerosol Sci. 19, 577585 (1988)
Mamane, Y., Gottlieb, J.: Ten years of precipitation in Haifa, Israel. Water Air Soil Poll. 82, 549558 (1995)
Manly, B.F.J.: Multivariate Statistical Methods: A Primer. Chapman & Hall, London (1986)
Mihalopoulos, N., Stephanou, E., Kanakidou, M., Pilitsis, P., Bousquet, P.: Tropospheric aerosol ionic
composition in the Eastern Mediterranean region. Tellus 49B, 314326 (1997)
Moody, J.R., Lindstrom, R.M.: Selection and cleaning of plastic containers for storage of trace element
samples. Anal. Chem. 49, 22642267 (1977)
Nastos, P.T., Alexakis, D., Kanellopoulou, E.A., Kelepertsis, A.E.: Chemical composition of rainwater in
Athens, in relation to the pollutants sources. In: Proceedings of the 7th Hellenic Scientific Congress in
MeteorologyClimatology and Atmospheric Physics, Nicosia, Cyprus (September 2004)
OSullivan, D.A.: Environmental concerns gain prominence in Europe. Chem. Eng. News 67(13), 715
(1989)
Pierson, W.R., Chang, T.Y.: Acid rain in Western Europe and Northeastern United States technical
appraisal. Crit. Rev. Environ. Control. 16(2), 167192 (1986)
Richman, M.B.: Rotation of principal components. J. Climatol. 6, 293335 (1986)
Samara, C., Tsitouridou, R., Balafoutis, Ch.: Chemical composition of rain in Thessaloniki, Greece, in
relation to meteorological conditions. Atmos. Environ. 26B(3), 359367 (1992)
Skoulikidis, T.N.: Effects of primary and secondary air pollutants and acid depositions on (ancient and
modern) buildings and monuments. In: Karlsruhe, F.R.G. (ed.) Proceedings of the EEC Symposium:
Acid deposition a Challenge for Europe (1983)
Smirnioudi, V.N., Siskos, P.A.: Chemical composition of wet and dust deposition in Athens, Greece. Atmos.
Environ. 26B(4), 483490 (1992)
Tuncel, P.G., Ungoer, P.: Rain water chemistry in Ankara, Turkey. Atmos. Environ. 26B, 483490 (1996)
Tsikritsis, G.E.: Precipitation chemistry in central Greece. Fresenius Environ. Bull. 15(12a), 14991505
(2006)
Vocom, I.E.: Air pollution damage to buildings on the Acropolis. J. Air Pollut. Control Ass. 29, 333338
(1979)