New insights into understanding urban traffic emissions using novel mobile air quality measurements in the Breathe London pilot study

Ma, Qingping 

Thumbnail Image
Change log

This thesis reports an investigation into the use of mobile air quality measurements to better understand urban traffic emissions. As part of the Breathe London project, two Google Street View Cars (based at the National Physical Laboratory) were fitted with air quality monitoring instrumentation and were driven across London over approximately a one year period. These mobile laboratories used a range of reference standard techniques to measure particulate matter and pollutant gases nitric oxide (NO), nitrogen dioxide (NO₂), ozone (O₃). Carbon dioxide (CO₂) was also measured as an atmospheric tracer linked to vehicle and other combustion related emissions.

The investigation begins with a characterisation and performance evaluation of the instrumentation, progressing to the development and deployment of advanced source apportionment techniques (scale separation and emission ratios) to extract the maximum information from the dataset. This study collected size-speciated particulate measurements using a light scattering device, black carbon using an aetholometer and Lung Deposited Surface Area (LDSA) using a diffusion-charging instrument. The performance of the instruments was compared using the mobile dataset. Significant amounts of close-to-source pollution were effectively measured using black carbon and LDSA (diffusion-charging) measurements. However, it was found that light scattering based particle counters were unable to measure significant amounts of pollution when measuring close to certain tailpipe emission sources.

Techniques to differentiate between local emissions and a well mixed background from time series measurement signals were compared as part of the development of techniques to extract emission ratios. Whilst two distinct distributions (a low distribution for the petrol fleet compared to a higher distribution for the diesel fleet) may have been expected in the NOx emission ratio, consistently only a single distribution was seen. A time series simulation showed that the ability to resolve two separate (petrol and diesel) distributions in the emission ratio distribution was lost with increasing fleet density, due to atmospheric mixing. Emission ratios calculated with mobile data showed downward trends in NOx, black carbon and LDSA over the sampling campaign period.

Jones, Roderic
Martin, Nick
air, quality, atmospheric, chemistry, pollution, mobile, sensor
Doctor of Philosophy (PhD)
Awarding Institution
University of Cambridge
EPSRC (1944587)
EPSRC iCASE Studentship, sponsored by the National Physical Laboratory