Abstract
Housing has long been thought to play a significant role in population exposure to environmental hazards such as high temperatures and air pollution. However, there is sparse data describing how housing may modify heat and air pollution exposure such that housing's role in poor health and mortality from these hazards may be estimated. This paper describes the development of individual-address level indoor overheating and air pollution risk modifiers for Great Britain, for use alongside historical weather, outdoor air pollution, population socio-economic data, and mortality data in a large-scale epidemiological investigation. A geographically-referenced housing stock database was developed using the Homes Energy Efficiency Database (HEED) and the English Housing Survey (EHS). Simulations of unique combinations of building, fabric, occupation, and environment were run using a modelling framework developed for EnergyPlus 8.0, estimating indoor temperature metrics, indoor/outdoor ratio of pollution from outdoor sources, and indoor air pollution from multiple indoor sources. Results were compiled, matched back to individual properties in HEED, and mapped using Geographical Information Systems (GIS). Results indicate urban areas had higher numbers of buildings prone to overheating, reduced levels indoor air pollution from outdoor sources, and higher air pollution from indoor sources relative to rural areas, driven largely by variations in building types. The results provide the first national-scale quantitative estimate of heat and indoor air pollution modification by dwellings, aggregated at levels suitable for inclusion in health analysis.
•A spatially-referenced building stock model was developed for housing in Britain.•EnergyPlus simulations estimated indoor overheating and air pollution risks.•Spatial variation of housing modification of risk was examined using GIS.•Housing stock variations led to greater overheating in urban areas.•Variation in housing stock and terrain caused higher I/O ratios in rural areas.