Trace metal regional sources in East Baltic region and
processes governing trace metal budget
D. Čeburnis
Atmospheric Pollution Research Laboratory, Institute of Physics, A.Goštauto 12, 2600 Vilnius, Lithuania, e-mail: ceburnis@ktl.mii.lt
abstract
Trace metal concentrations were
investigated in atmospheric air and precipitation at two stations in Lithuania,
located in a distance of 300 km downwind. Distribution of trace metals on
coarse and fine aerosols was investigated using virtual impactor (2.5 μm
cut-off size). Air mass trajectories calculated on the NOAA Air Resources
Laboratory’s web site showed no principal difference in air mass origin at both
sites. Concentrations of most metals were considerably higher at the coastal
station, both in air and precipitation. The major exception was Mn, which
showed a vegetation influence at the continental site. It is concluded that
East Baltic region acts mainly as a sink for long-range transported heavy
metals. Significantly lower lead concentrations, both in air and precipitation,
at the continental station allow us to conclude that local lead emissions in
Lithuania are not significant on a regional scale.
Introduction
Today the occurrence of the
long-range transport is well known and established fact. Nevertheless it is
still need to gain more knowledge about transport, transformation and
deposition of airborne pollutants including trace metals. The recent decade is
distinctive one in Central and Eastern Europe due to major economical and
structural changes which induced environmental changes as well. After sharp
decrease in environmental pollution in the beginning of decade, there was a
gradual increase in industrial activities afterwards, the most important (ten
fold) being connected with transport traffic. Atmospheric trace metal pollution
in Lithuania was studied since 1975 (Šakalys and Kvietkus,
1989) and extended in 1990 when
GFAAS method was introduced for all kind of trace metal analysis. The sampling
and sample preparation methods were intensively developed and unified during
the decade. This study summarizes the most recent results obtained during field
campaigns or within monitoring programs. The main purpose of this study was to
evaluate regional trace metal sources with the emphasis to contributions from
smaller or larger scale local sources.
Materials and methods
During the decade atmospheric trace metals were investigated at two main stations, one located at the Baltic sea coast, Preila (most western part of Lithuania), and second one 300 km apart, Rugsteliskes (most eastern part of Lithuania). Air samples were collected using either total filters (Whatman 40) or virtual impactors, using the same filters (Ulevičius et al., 1999), on daily basis during campaigns or weekly basis within monitoring program. Filters were extracted in 1 % nitric acid by shaking in an ultrasonic bath (1 hour, 680 W). The estimated uncertainty of the collection efficiency was very low (1-2 %). The reproducibility of individual trace metal concentration was examined involving a parallel collection of the air samples (1.5 m distance) as well as an independent extraction and analysis procedure. This estimation was made of 7 samples (a 2-4 days collection period with a flow rate of 1 m3/h). The uncertainties obtained were as follows: Pb, 10%; Cd, 5%; Cu, 30%; V, 7%; Zn, 15% and Mn, 10%. Thus, the overall uncertainty was approximately 10%. The collection efficiency of the virtual impactor developed by Ulevičius et al. (1999) was estimated comparing with Whatman 40 filter measurements and was found to be 90 % or higher.
Precipitation was collected on weekly basis at the coastal station and
on monthly basis at the continental
site. The precipitation samples were collected in parallel acid washed
polyethylene collectors. After collection precipitation samples were acidified
to 1-2 pH in the collectors prior to transfer into the storage bottles to
prevent loss of metals on the collector walls. The uncertainty of trace metal
concentration in individual samples was estimated to be as follows: Pb, 34 %;
Cd, 46 %; Cu, 28 %; V, 11 %; Zn, 12 % and Mn, 23%.
Air mass trajectories were calculated on the NOAA Air Resources
Laboratory’s web site every day for the coastal site and every second day for
the continental site.
Results and discussions
Distribution of trace metals on fine
(<2.5 μm) and coarse (>2.5 μm) mode aerosols was studied during two summer and two winter months at the
coastal station. Results presented in Figure 1 clearly show a distinct
differences between anthropogenic elements like Pb, Cd, V, Zn and elements of
dual origin like Cu and Mn. It is important that distribution percentage did
not change for anthropogenic elements (only total concentration increased by
about 50 %), while for Cu and Mn concentration on fine aerosols did not change
during different seasons, but increase in concentration on coarse particles
during summer therefore increased total concentration of Cu and Mn and changed
the percentage respectively.
Figure 1. Percentage
of trace metal concentration distribution on fine and coarse aerosols during
summer and winter time at the coastal station.
Distribution of air masses according
to their origin was studied on the NOAA Air Resources Laboratory’s web site.
Table 1 shows the statistics of air mass distribution among four sectors for
two sites (300 km apart). It is clearly seen that during the year 1999, both
sites were affected by air masses of nearly exactly the same origin with about
65 % being of western origin and only 35 % originated from the east. Referring
to these data we made an assumption that air masses roughly enter East Baltic
region from the west and transport pollutants form the rest of Europe.
Table 1.
Distribution of air mass origin between the sectors for the sites Preila and
Rugsteliskes during 1999.
|
Sector |
Preila, % |
Rugsteliskes, % |
|
NW |
32 |
32 |
|
NE |
16 |
16 |
|
SE |
16 |
19 |
|
SW |
36 |
32 |
A principal component analysis (PCA)
performed on Preila fine aerosol trace metal concentration data showed no
difference between summer and winter data as presented in Table 2 and 3
(Čeburnis et al., 1999).
