Aerosol Remote Sensing and Modelling: Estimation of Vehicular Emission Impact on Air Pollution in Nairobi, Kenya

Abstract

In cities experiencing rapid urbanisation and motorisation like Nairobi, deterioration of air quality is bound to occur due to vehicular emissions. Exposure to ambient aerosols is a leading risk factor which is attributed to respiratory diseases, cardiovascular diseases, lung cancer and stroke. Regular monitoring of air quality is therefore important for constant assessment of the abundance of air pollutants in the air we breathe. The model for Highway Development and Management (HDM-4) was calibrated against Edgar-DICE emissions. Calibration of vehicle Engine Output Models (EOM) involved adjustment of default emission coefficients till the model output emissions was in agreement with the Edgar-DICE emissions. Calibrated model was used to predict annual exhaust emissions within Nairobi. The effect of traffic growth on exhaust emissions was investigated for a twenty-year period. Sentinel 5P satellite measurements of total NO2 tropospheric column was used to map spatial distribution of NO2 in the study area. Spearman’s rank correlation coefficient was used to rank the degree of sensitivity of the model input variables and then visualized as Tornado plots. According to this test, vehicle life, vehicle operating weight, and road geometry had the highest influence on the exhaust emissions. The HDM-4 annual modelled quantities of CO, NOx, and PM2.5 were 60324, 4582, and 4273 tonnes respectively. For the same pollutants, the Edgar-DICE emissions predicted 42420 tonnes, 1458 tonnes, and 3188 tonnes respectively. The improvement in HDM-4 predictions was attributed to its bottom-up approach unlike the Edgar-DICE emissions which were generated through a top-down approach. Exhaust emissions forecasting showed possible increment of PM2.5 and CO by 11.5% and 2.20% respectively for the high traffic growth scenario while NOx reduced by 0.22%. In low traffic growth scenario, PM2.5 and CO recorded a reduction of 1.33% and 0.1% respectively, contrary to NOx which increased by 0.5%. This behaviour was attributed to the fact that traffic congestion has negative influence on NOx emissions. Observation from Tropospheric Ozone Monitoring Instrument (TROPOMI) satellite showed patterns of high NO2 concentration along busy highways connecting the city. The study has demonstrated alternative tools which can not only provide information on air quality in areas with limited coverage of ground monitoring but also form new frontier in air quality research through modelling and satellite remote sensing. Adequate understanding of the of source contribution to overall emissions is key to formulation of target control measures.