Semi-empirical
modeling of trace metal deposition close to the point pollution source
D. Ceburnis, J. Sakalys, D. Valiulis and K. Kvietkus
Atmospheric Pollution Research Laboratory,
Institute of Physics, A.Goštauto 12, 2600 Vilnius, Lithuania, e-mail: ceburnis@ktl.mii.lt
Abstract
For the estimation of heavy metal deposition close to the point pollution source, data from the analysis of moss growing close to the thermal power stations and snowpack have been used. Snow and moss samples were analyzed using an atomic absorption spectrophotometer Perkin-Elmer Zeeman/3030. A semi-empirical model was proposed to describe atmospheric trace metal deposition close to the thermal power station. Model parameters were derived from experimental data and amounts of nickel and vanadium washed out with snow and rain were calculated. Using data of long-term meteorological observations about duration of rain and snow events and metal uptake efficiencies in moss it was calculated the average emissions of vanadium and nickel from the stack. The coincidence between data from emission inventory and model results was quite sufficient.
INTRODUCTION
Mosses are bioorganisms
sensitive to pollution, especially to heavy metals. Mosses receive nutrients
needed for vital processes from the atmosphere, while the relationship with the
substrate beneath is slight. The methods of using moss for heavy metal
monitoring have been developed in Sweden (Ruhling and Tyler, 1968) and are
widely used all over the world at present. The moss methods are especially well
suited to the estimation of heavy metal deposition from the atmosphere on a
larger time scale. The largest amount of metals from the atmosphere is received
by moss with precipitation and especially with rain. Therefore, moss was used
for the estimation of pollutant washout with rain (also partly with snow after
subsequent spring snowmelt). Snow analysis if it is possible to sample snowpack
of all winter period enables to
distinquish pollution during cold period. If the experimental data can provide
with distinctive plume profiles close to the point pollution source, than a
certain calculations can be performed to describe the washout of heavy metals
from the atmosphere. There are several very distinctive thermal electric power
stations in Lithuania situated in rural areas or small cities and two of them
were selected for this study.
Materials
and methods
Snow and moss samples were
collected in the vicinity of
two thermal electric power stations. Elektrenai thermal power station is
situated in the southern part of Lithuania, in the small city Elektrenai.
Production of electricity in this
station depends on the production rate in nuclear power station. About 150-200
mln. tones of residual oil and up to 100 mln. m3 of natural gas in average
are burned every year in this station. The other thermal power station serves
as energy source for oil refinery station in northern part of Lithuania, rural area, close to the small
city Mazeikiai. Moss samples were collected at different distance from the
source varying from as close as 0.5 km to as far as 15-20 km. Sampling grid
followed more or less prevailing wind direction (more samples in favourable
direction than opposite). Snow samples were collected only close to Elektrenai
station during the exceptional winter 1995-1996 when collection of snow of all
winter period was possible. There were collected up to 20 moss samples (around
every station) and 53 snowpack samples in the first case. The moss and snow
samples were analysed using an atomic absorption spectrophotometer Perkin-Elmer
Zeeman/3030 according to the methods described in Čeburnis 1997.
Description of the calculation method
The emission from the pollution source in the absence of washout is expressed:
E = 2 p x v q, (1)
where x is the distance from the stack; v is the wind speed; q is the amount of pollutants in the whole air column, the base of which is the area unit.
As the precipitation fall speed at which pollutants are washed out is w, pollutants will reach the ground surface at a distance ro = h v/w, where h is the effective height of the stack.
At the distance ro the emission variation can be expressed by equation:
dE = - E m dr (2)
where μ is the coefficient of washout, i.e. the part of pollutants washed out when air mass had passed the distance of the length unit. the speed of both precipitation and the wind changes concurrently with ro.
A special attempt was made to estimate dry deposition close to the point pollution source. For this purpose meteorological data from Vilnius air quality station were used. As Vilnius and Elektrenai are not very apart (50 km distance), meteorological data does not differ much (Kaušyla and Šver, 1983). Pollutant flux to the Earth surface can be calculated according Zanetti (1996) using recurrence of atmospheric stability classes and wind speed data in case of 100 % aerosol uptake by surface. The maximum flux than would be in the distance of 4-5 km from the stack and within this distance about 6 % of all emitted pollutants would deposit. Unfortunately due to the limited data it was impossible to calculate an uptake value, but it is obvious that 100 % uptake is very critical, so 6 % value is highly overestimated. According to the results below we can conclude that dry deposition in the close vicinity (up to 6 km) does not play an important role at least in areas with regular precipitation pattern.
After appropriate mathematical transformations the final equation which describes the surface concentration in case of snow events was obtained:
(3)
A much more simple equation can be obtained for the rain events when taking into account the velocities of rain and snowfall. The velocity of snowflakes is about 1 - 2 m/s, while that of raindrops is about 10 - 15 m/s. Taking into account that the average wind speed is about 5-10 m/s, the effective stack height is about 300 m, we calculate r0 from ro = h v/w. In the case of rain it is about 100 - 200 m, and in the case of snow, about 0.7 - 3 km. The closest samples of snow and moss were taken at 0.5 - 1 km.
Therefore in the case of rain ro ® 0 and σ ® 0. Than:
(4)
Metal concentration distribution in snow was calculated according to Eq. 3 and that in moss - according to Eq. 4.
Results and Discussion
Using the method of least squares, parameters μ and A were calculated for the case of moss and m , s, A and r0 - for the case of snow. The main parameters m , s, A ir r0 are presented in Table 1.
Table 1. Calculated values of the model parameters.
|
Metal |
Rain |
|
Snow |
|
|
|
m |
m |
ro, km |
s, km |
|
V |
0.061 |
0.046 |
1.89 |
0.77 |
|
Ni |
0.059 |
0.023 |
2.20 |
0.88 |
Measurement and model data are presented in Figs. 1 and 2.
,