Assessment of outdoor air pollution in Adama, Ethiopia

Abstract

PDS 65: Exposure assessment: implications for epidemiology, Exhibition Hall (PDS), Ground floor, August 27, 2019, 1:30 PM - 3:00 PM Background/aim: It has been shown that the risk of preeclampsia increases with increasing levels of PM2.5. In Ethiopia, preeclampsia is the main reason for maternal mortality, and for 25% of the stillborn babies. The aim was to compare different methods to assess the outdoor air pollution in Adama, as part of a project in which the effects of air pollution on preeclampsia in Adama, Ethiopia will be evaluated. Methods: NOx and NO2 was sampled at 40 different sites during the dry (February) and wet (August) seasons using passive samplers (Ogawa). Sampling time was 5 days. The assumption that NO2 can be used as a proxy for vehicle emitted PM2.5 were tested by measuring PM2.5 with sensors (Alphasense) at 20 of the sites. At one urban traffic site, advanced measurements were conducted using DustTrak and UPAS for time-resolved PM1, PM2.5, and PM10, Filter sampling for Organic carbon/Elemental Carbon (OC/EC) and nitrates and sulphates. Time-resolved PM measurements were also conducted at an urban background site. Sampling of polyaromatic hydrocarbons (PAH) was conducted at 3 sites. Results: PM2.5 levels at the 20 sensor sites were on average 23 µg/m3, exceeding the 10 µg PM2.5/m3 as annual average WHO guideline. None of the sites had averages below this guideline. Concentration peaks exceeding 1 mg/m3, and occasionally 2 mg/m3, of PM2.5 could be identified at most of the sites. The measured PM2.5 values were on average over double the PM2.5 values estimated from NO2 sampling. Conclusion: The NO2 method for estimating traffic emitted PM2.5 is does not seem applicable in settings such as Adama, Ethiopia, possibly because the vehicle fleet is old. Good quality sensors with SD cards, run on battery, placed in strategic locations (based on emission sources, city plans and security) was the most feasible method for assessing PM2.5 levels.