A novel study on the investigation of three very common atmospheric sulphur species
relevant to the Mpumalanga Highveld subregion was conducted. Long-term in situ
measurements were applied in the diurnal and seasonal evaluation of the observed sulphur
species. Ambient pollutant concentrations and surface meteorological data were collected
at an air quality monitoring station at Elandsfontein. Elandsfontein air quality monitoring
station was ideal for the observations due to its high elevation within the Mpumalanga
Province surrounded by few rolling hills and negligible windbreaks which easily allows for
extensive plume-contact with the surface during convective daytime mixing. The temporal
characteristics of the sulphur species have been assessed relative to one another with
varying meteorological conditions. The diurnal and seasonal concentration variations were
used to describe the physical characteristics exhibited by the compounds over
Elandsfontein. Pollution roses were used to target the source of the major release points
and areas of these sulphur species relative to the Elandsfontein monitoring station. Gas and
particle concentrations were analysed in relation to varying meteorological parameters with
a view to ascertaining the sulphur transformation and concentration distribution in the
planetary boundary layer. Particulate sulphate distribution has been modelled through
multivariate regression analyses in relation to three meteorological parameters, namely,
wind speed, relative humidity and ambient temperature for the various seasons observed
over southern Africa.
This study has shown that hydrogen sulphide, sulphur dioxide and sulphate species are
present throughout the year in the Mpumalanga Highveld at notably significant levels. The
presence of ambient particulate sulphate has been shown to result from the combination of
chemical interactions during long-range aerosol transport; atmospheric recirculation
processes shown from back trajectories over the southern Africa sub-region, as well as the
variation in the removal mechanisms and rates for the different seasons throughout the
year. These transport and removal processes all contribute to the overall sulphur mass
balance in the planetary boundary layer. Dosage of the three sulphur species was evaluated
to provide data for sulphur pollution loading that could form a basis for health and
vii
environmental impact assessments over the area. In view of the characteristic patterns
displayed by particulate sulphate, multivariate mathematical models have been developed
on a seasonal basis with variations in meteorological parameters. This was seen to predict
an accuracy of up to 70 % of the particulate sulphate loading for different seasons over the
South African Highveld.
In order to understand the chemical interactions of atmospheric sulphur species, it is
important to be able to predict the route taken and expected products of transformation on
any given condition. Theoretical analyses of the chemical thermodynamic properties of the
known reacting species and a well-established approach were used in predicting reaction
paths and establishing the possible and feasible products of chemical transformation in
relation to the ambient temperature. The determination of reaction paths and possible
products of chemical transformation provides a measure of the relative importance of the
reacting species and the mechanism of reaction. Gas-, aqueous-phase and radical reactions
involving sulphur (IV) were investigated with a view to establishing their relative
importances. Thermochemical properties of several sulphur-containing compounds not
available in the literature have been generated for evaluation of Gibbs free energy (ΔG)
and enthalpy (ΔH). An electronic energy structural approach has been applied to model for
ΔG and ΔH of 88 sulphur species in 90 chemical reactions comprising gas-phase, aqueousphase
and radical reactions. Modelling was evaluated for their relative importances over a
temperature range of –100 °C to +100 °C. The temperature range is well above the known
tropospheric temperature range to account for variations in the atmospheric environment.
To further comprehend the chemistry of sulphur with regards to distribution of the species
in the atmosphere, a kinetic model is developed and incorporated into a dispersion model.
The kinetic evaluation of the oxidation rate of SO2 to sulphate has been determined with
advection and dispersion over the Elandsfontein area. Gas-phase transformation with
advection and dispersion has been used to evaluate the extent of the distribution of SO2
relative to the major contributing sources. The dry deposition was considered to be the
dominant removal mechanism. It was assumed that the reaction rate was second order in
concentration and that the rate of deposition was first order. The oxidation rates obtained
for the seasons were 10.9 % h-1 for summer; 8.83 % h-1 for autumn; 6.56 % h-1 for winter;
viii
10.8 % h-1 for spring, while an overall rate of 9.6 % h-1 was obtained for the one year study
period. The transformation rate model produced a reaction constant and an activation
energy of 4.92 x 10-6 μg m-3 s-1 and 36.54 kJ kg-1 for summer; 3.939 x 10-6 μg m-3 s-1
and 43.89 kJ kg-1 for autumn; 2.90 x 10-6 μg m-3 s-1 and 115.69 kJ kg-1 for winter;
4.82 x 10-6 μg m-3 s-1 and 43.29 kJ kg-1 for spring, while for the year
4.29 x 10-6 μg m-3 s-1 and 34.31 kJ kg-1. A Gaussian puff unsteady state Lagrangian
dispersion model with reflection at the surface and inversion layer was applied for
concentration diffusion. The Lagrangian dispersion model with dry deposition was a better
prediction of the observed data than the models from previous studies using a first order
rate constant with or without deposition rate.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:wits/oai:wiredspace.wits.ac.za:10539/5587 |
| Date | 02 September 2008 |
| Creators | Igbafe, Anselm Iuebego |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Thesis |
| Format | application/pdf |
Page generated in 0.0026 seconds