This project is developing a new methodology for modelling the distribution of population in time-space.
This project is funded by the Economic and Social Research Council and led by Professor David Martin, Samantha Cockings and Samuel Leung. The project commenced on 1 March 2009 and runs until 30 April 2011.
Many areas of social science research and policy are concerned with the exact locations of population, for example to measure exposure or target resources. Currently available representations effectively estimate population at home addresses, or 'workplace' populations which provide a partial view of the population at places of work. Neither of these satisfy the pressing need for representations of the ever-changing spatial distribution of population over time. The temporal dimension incorporates both historical and cyclical timescales in addition to unique events which influence the spatial distribution of population.
David's established population surface modelling method redistributed data from area centroids and population counts and the SurfaceBuilder and Surpop2 software tools have been widely used by other researchers. This present project is building on these concepts to establish a generalisable modelling framework, able to incorporate spatially referenced count data appropriate to any specific time and date. Census output areas, workplaces, educational establishments, hospitals and leisure locations can all be treated as 'centroids' around which counts can be redistributed over both space and time. The project is concentrating on data covering the period 2001-2006 for England and Wales.
The project encompasses methodological development, production of a new software tool, extensive dataset review and preparation and development of appropriate visualisation methods. Specialist programming support is being provided by Jason Sadler and Hugh Darrah in Geography's GeoData Institute.