Use this URL to cite or link to this record in EThOS:
Title: Development and evaluation of portable passive and real-time measurement systems, and dispersion models, to estimate exposure to traffic-related air pollutants
Author: Masey, Nicola
ISNI:       0000 0004 7226 3600
Awarding Body: University of Strathclyde
Current Institution: University of Strathclyde
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
This research developed efficient applications of portable measurement systems to assess human exposure to traffic-related air pollution through direct measurement, and evaluation of exposure models. Passive NO2 samplers are deployed at large numbers of sites in epidemiological studies to estimate typical concentrations over 1-4 weeks. I found that deployment time could be reduced to 2 days with limited impact on the accuracy and precision of exposure estimates. This shorter measurement time enabled observation of wind-speed effects leading to overestimation of ambient concentrations by passive samplers. Through development of a post-processing technique and/or inclusion of a membrane I improved sampler accuracy. Portable sensors can provide detailed estimates of personal exposures to air pollution. Many sensor-based monitors have not been subject to rigorous testing procedures to quantify their accuracy. I observed that the most accurate estimates of concentrations from NO2 and O3 sensor-based monitors required regular, intermittent calibration against reference analysers under similar environmental conditions to field measurements. I also found deterioration in BC monitor accuracy and precison when the attenuation of the collection filter exceeded 40 and no improvement in monitor accuracy was observed when filter darkness correction algorithms were applied. Portable sensors can be used to identify locations with higher concentrations, which may require more detailed monitoring. I established that repeated 6-minute measurements of BC and particle number concentrations estimated similar spatial trends to 1-week NO2 measurements using passive samplers. Dispersion models can be used to estimate pollution exposure at multiple locations over a study area. I found that initial user parameterisation in a weather model had limited effect on pollution estimates from a dispersion model. I evaluated a new GIS-based dispersion model (5 x 5 m NO2 estimates for a 3,500 km2 area, with model run times of under 10 minutes). I demonstrated that inclusion of discrete street canyon models and geospatial surrogates (accounting for urban morphology) improved model accuracy. The measurement and modelling evaluation research in this thesis complimented each other by providing efficient ways to directly measure population exposures.
Supervisor: Heal, Mathew ; Hamilton, Scott ; Beverland, Iain Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral