Assessment of environmental exposure to air pollution within four neighbourhoods of the Western Cape, South Africa

Abstract

Background: A recent review on the effects of ambient air pollution on human health in sub-Saharan Africa, specifically calls for an urgent need for more epidemiological studies in developing countries due to a lack of data in these countries. Air pollution information on exposure is important for understanding and addressing its public health impact in developing countries. In many African countries, the spatial distribution of air pollutants has not been quantified even though air pollution is a global public health risk. The main goal of the study was to quantify and compare the seasonal spatial variation of household air pollution in the 4 Western Cape neighbourhoods. Methods: Weekly indoor and outdoor measurements of Particulate Matter (PM2.5), Sulphur dioxide (SO2), Ozone (O3), Carbon monoxide (CO) and Nitrogen dioxide (NO2) were conducted at 127 households in four informal settlement areas (Khayelitsha, Marconi-Beam, Masiphumulele and Oudtshoorn) during one month each in summer and winter. PM2.5 measurements were conducted using Mesa Labs GK2.05 (KTL) cyclone with the GilAir Plus Air Sampling Pump, Gases were measured using Passam passive samplers. Statistical analyses were performed using Stata V12. Simple linear regression was used to evaluate the relationship between continuous exposure levels and the respective predictor variables. These include distance to major roads, bus routes, open grills and waste burning sites. Results: The highest average weekly outdoor PM2.5 and NO2 concentrations for summer were recorded in Milnerton (8.76 µg/m3 and 16.32 µg/m3 respectively). However, the highest average concentrations during winter for PM2.5 were recorded in Oudtshoorn (PM2.5: 16.07 µg/m3), whilst the highest NO2, was recorded in Khayelitsha (NO2: 35.69 µg/m3). SO2 levels were consistently low during both seasons. Noordhoek generally recorded the lowest average levels for all pollutants. Winter average weekly concentrations were generally higher than the levels recorded in summer for all pollutants. In a sub-sample of indoor and outdoor measurements, the results were comparable for PM2.5, NO2 and CO. However, the results of Ozone (O3) showed relatively higher (~10 times) outdoor compared to indoor levels. Linear regression modelling results revealed that significant predictors of elevated exposure to PM2.5 were proximity to construction activities and open grills. Analysis demonstrated a clear dose-response relationship with distance, with open grills within 1000m associated with a 0.33 µg/m3 increase in PM2.5 to 6.77 µg/m3 at a distance of 25 meters. Results from the linear regression modelling revealed that significant predictors of exposure to NO2 were proximity to rapid transport bus stops, bus routes, taxi routes and major routes. Distance to rapid transport bus stops demonstrated an increase in NO2 between 0.09 µg/m3 (at 1km) to 2.16 µg/m3 (at 50m) during summer. A similar pattern was observed for taxi routes and bus routes displaying an increase of 6.26 μg/m3and 6.82 μg/m3 respectively within the proximity of 1000 meters.