Indoor and outdoor exposure to PM10 in properties in the vicinity of urban streets

Abstract

Occupants of buildings are exposed to indoor pollution from cooking and smoking and infiltrated outdoor pollution. The fabric of a building (doors, windows, ventilation etc.) has an influence on the infiltration of outdoor pollution into the building. In some studies, personal exposure has been investigated within homes and different transport modes. However, there is a lack of knowledge about pollution level variations along congested, busy and quiet roads in urban areas and its infiltration into the buildings located some distance from or along the roads. Only a few studies have investigated dynamic and static indoor/outdoor monitoring simultaneously in the same urban area to establish relative levels of exposure in different microenvironments. The aim of this study was to investigate PM10 exposure to indoor and outdoor air pollution simultaneously as a function of activity patterns in urban streets/areas. This thesis describes the research carried out to investigate indoor and outdoor monitoring of PM10 exposure within and outside the air quality management area (AQMA), in Gosforth, Newcastle upon Tyne, UK. It examined the results of several days (at a sampling rate of one second or one minute) of monitoring of particulate matter (PM10) levels simultaneously indoors (static monitoring) and outdoors (static and dynamic monitoring). The static monitoring was conducted in a number of houses and commercial premises in Gosforth and Jesmond areas in Newcastle whilst dynamic monitoring was conducted along the High Street in Gosforth. For static monitoring, PM10 monitors were installed in the lounge and kitchen in houses and the reception areas of the commercial properties. The property owners were asked to record activity (such as cooking, vacuum cleaning, door opening etc.) in a diary for at least one day during the week and a day at weekends. For dynamic monitoring along the High Street Gosforth, the observer carried a portable PM10 monitor and a GPS monitor in a back pack and walked on the pavement alongside the street. The observer also noted the traffic condition, passing of HGV and buses, crossing of junctions and other activities, such as street cleaning, construction, cigarette smoking, all of which influence PM levels. Arc GIS software and statistical techniques were used to map spatial and temporal variations in PM10 levels recorded during several dynamic monitoring campaigns. Similarly, temporal variations in PM10 levels in houses were also plotted. Statistical techniques were used to fit distributions to the temporal variations in PM10 ii concentrations. Timestamps of traffic activities and events aligned with the time series for the dynamic monitoring have helped to identify their influence on PM10 levels. This research applied the basic theory of the statistical technique known as ‘decomposition’ to reveal features in the probability density functions (pdfs) derived from static measurements (indoor/outdoor) as well dynamic. The decomposition technique was used to characterise the influence of various sources and events on indoor and outdoor PM10 levels, to provide a richer understanding of whether exposure is influenced by the traffic flow regimes in the vicinity of properties. The decomposition technique was used to characterize pollution measured indoors disaggregating the contributions to the total pdfs of sources such as cleaning, cooking, sleeping as well as from outdoors with sources mainly traffic activity, street works. The dynamic second by second averaged to one minute PM10 levels were also decomposed to map onto sources associated with traffic condition. Component distributions fitted by the decomposition technique were lognormal for both static and dynamic monitoring. The results of the time series analysis have shown that monitored exposures vary substantially and are unique to the location and temporal variation of the measured microenvironment whether indoors in a kitchen or lounge, inside a commercial property or whether out of doors at the facade of a building or dynamically on a pavement alongside a road. The application of the decomposition technique was demonstrated to be promising. Static indoor and outdoor pdfs were mainly characterised by three or more log-normal distributions whilst the dynamically monitored data were fitted with three. Activities such as cooking, those associated with doors and windows opened or closed, use of extractor fan in the kitchen and vacuum cleaning were found to have a strong influence on indoor PM10 concentrations. Also, outdoor PM10 levels were governed more by the stop-start and idling characteristics of traffic rather than level of flow and traffic has little influence on temporal variations in indoor PM10 over time of the day. Instead it is the indoor activity that mainly governs the temporal variations in measured indoor concentrations of PM10. Multi-lognormal distributions explained typically 83% to 98% of the measured variance in the total pdfs. Finally, the author is not aware of any studies which have used the decomposition statistical technique to analyse dynamic and static indoor/outdoor monitoring in the same urban area to develop a fundamental understanding of the relative importance of the different sources of pollution in different microenvironments on personal exposure levels.