Concentrations and deposition of atmospheric species at regional sites in southern Africa
Martins, Jacobus Johannes
MetadataShow full item record
Increasing atmospheric emissions of trace gases as a result of fast-growing economies like South Africa, especially those of SO2 and NO2, are a major environmental concern. The air quality and thus air chemistry of a region ultimately determines the composition of the deposition taking place. The air quality of a region is determined by many factors, including transportation processes of neighbouring regions that remove, disperse and transform gas to aerosol and vice versa. All chemical processes from emission to removal of all the trace species within a region and the rate thereof are key factors in understanding the air quality of a region. Air quality on regional and multi-regional level needs to be determined in order to ensure sustainable development. By determining atmospheric depositions using ambient concentrations data, the air quality of regions can be better understood, leading to improved decision-making processes for sustainable development. For this study, regional air quality was determined using dry deposition estimates calculated from the ambient concentrations of gases and aerosols. This was done by creating a long-term data set of inorganic trace gases of interest, as well as data sets of aerosol species. Data of both gas and aerosol species were measured to coincide with each other in different regions to enable the modelling of atmospheric processes. The measurements of monthly mean gaseous concentrations of sulphur dioxide (SO2), nitrogen dioxide (NO2), ammonia (NH3), nitric acid (HNO3) and ozone (O3) at four remote sites — Louis Trichardt (South Africa), Cape Point (South Africa), Amersfoort (South Africa) and Okaukuejo (Namibia) — in southern Africa, over a period of nine to 11 years, were done by using a diffusive (passive) sampling technique. The average ambient SO2 and NO2 concentrations at Amersfoort were by far higher than any of the other sites and are characteristic of the highly industrialised region it is situated in. The O3 concentrations were the most constant for all the sites during the sampling period, except at the Louis Trichardt site where it was slightly higher than at the other sites. The annual concentrations of all gaseous species measured decreased from 1995 to 2001, after which they increased somewhat. Strong inter-annual variations were observed, which proves the scientific value of decision-making based on long-term observations. The 10-year mean concentrations provided the highest mean ambient concentrations for SO2 (2.8 ppb), NO2 (2.5 ppb) and HN03 (0.9 ppb) at the Amersfoort site, while the highest ambient concentrations of O3 were found at Louis Trichardt (35 ppb) and for NH3 (1.5 ppb) at Cape Point. Deposition estimates were made for the gaseous species using results from the inferential technique. The organic and ionic compositions of ambient aerosol measured on a daily basis during seasonal campaigns at Amersfoort (AF) and Louis Trichardt (LT) in South Africa were determined for 2005-2007. The average aerosol measured during all four campaigns at Amersfoort consisted of a carbonaceous fraction of between 45-60% in winter and 33-36% for the summer season. The ionic fraction of the Amersfoort aerosol mainly consisted of sulphate, ammonia, nitrate and minerals/organic acids making up a total of 25%, 10%, 7% and 6% for winter and 44%, 15%, 3% and 4 % for summer, respectively. The aerosol measured at the Louis Trichardt site consisted of a carbonaceous fraction of between 31-56% in winter and 23-31% for summer. The ionic fraction of the Louis Trichardt aerosol species of sulphate, ammonia, nitrate and minerals/organic acids consisted of 23%, 7%, 6% and 16% for winter and 44%, 17%, 3% and 9 % for summer, respectively. High correlations were found between the inorganic species of SO4-, NH4+, and NO3- suggesting that they originate from the same sources. Wind trajectories were calculated using the HYPSPLIT model, which suggests the transport of aerosol species from the same sources over the Amersfoort and the Louis Trichardt regions. By using similar methods as Guazzotti et al, 2003 during the INDOEX experiment, possible regional sources were identified through the organic and ionic composition of the measured aerosols. It was concluded from these results that complete combustion processes of fossil fuels dominate the ionic composition of aerosols over both these regions as well as the transformation reactions during transport. Only the formation of nitrate has different influences. The carbonaceous composition over the Amersfoort region is mainly influenced by incomplete biofuel and fossil fuel combustion processes. The carbonaceous content over the Louis Trichardt region is mainly influenced by incomplete biofuel combustion processes, of which biomass burning is the main contributor. The deposition rates of the inorganic aerosol species were calculated using results from the inferential technique, and this, together with the deposition rates of the gaseous species, was used to determine the total dry deposition of nitrogen and sulphate. It was found that approximately 30% of the atmospheric sulphur get deposited in the Amersfoort region through dry deposition, while 45% are deposited at Louis Trichardt. For total atmospheric nitrogen, only 27% get deposited through dry deposition in the Amersfoort region, while 45% get deposited in the Louis Trichardt region. The ORISAM-TM4 model was used to simulate secondary aerosol production over both the Amersfoort and Louis Trichardt sites. The measured gas and aerosol concentrations were compared with the ORISAM-TM4 model results. The model's simulation for the winter periods in general at both sites seems to be consistent and good for all secondary aerosols when no real temperature influences are present. Well correlating simulations were made when optimum conditions for key parameters (temperature, humidity, and precursor gas and aerosol concentrations) are present, as for the winter of 2006 campaign. The model, as well as the experimental results, shows more aerosol production during winter at both sites for all aerosol species. The model simulations are not too accurate during summer periods for both sites. The experimental BC/OCtot ratios at the Amersfoort site compared well with those in literature. OCp/OCp + OCsec and BC/OCtot ratios of both sites in South African conditions again show the poor simulations of the model during summer. All the parameters for which the model were sensitive, as found in development testing, were found to be applicable for southern African conditions, which are as follows: temperature < gas concentrations < humidity < aerosol concentrations. Particle size does not seem to have any real effect on the secondary aerosol simulated by the model.