Urban areas are highly dynamic and diverse systems and the interactions and networks
within an urban area are, at present, only partly understood, although some of the most
important impacts on human health occur in these areas. It is essential, therefore, to develop
a deeper understanding of these urban dynamics and processes especially with regard to
exposure and health risk assessment. This thesis describes the development and application
of an urban SImulation for ENvironmental health Analysis (SIENA). SIENA provides a
controlled, simplified urban environment to develop and test spatial epidemiological
concepts and models, to simulate processes and interactions relating to environmental
exposure and to explore theoretical and methodological problems in the spatial analysis of
environmental health.
The development of the simulated urban system focuses on identifying and quantifying
fundamental processes and inter-dependencies in the structure of urban areas in Great
Britain. Twelve cities are chosen as sample cities and their spatial data structure
(topography, transport network, land cover) and relationships between these structures and
the urban population are statistically analysed. Based on the results of the statistical analysis
SIENA is developed within a Geographic Information System (GIS) using probabilistic
models and spatial analysis tools. Beside the identified core structure, topography,
transport network, land cover and population, additional data such as traffic flow, air
pollution monitoring networks or emissions from industrial sources amongst others are
modelled and incorporated into SIENA.
To demonstrate the potential of the simulation, SIENA is applied in two case studies
both focusing on the misclassification of human exposure to urban air pollution. The first
case study explores the representativeness of various air pollution monitoring networks and
the resulting implications for exposure assessments. For the second case study, personal
exposure is simulated within SIENA and then compared to the use of a location-based
exposure proxy and the potential exposure misclassification spatially analysed.