Novel Spatial Analysis Of Residential Resource Consumption Via The Melbourne Train Network

Urban data volumes are increasing and becoming more accessible at a rapid rate. Therefore, novel approaches are required to synthesise and analyse the data in a meaningful way. The Australian Urban Research Infrastructure Network (AURIN) is providing tools and is negotiating data access to key datasets about cities to facilitate research on urban issues. Through AURIN, a number of datasets around urban settlements have been made available to researchers and policy analysts, including on demographics, housing, and resource consumption. These datasets are aggregated to geographic regions such as Australian Bureau of Statistics statistical areas, local government localities or postcodes in order to cater for the purpose of these regions. However, this can be a challenge when performing contextual analysis over a given region, as the data is not easily compared between spatial aggregations. Although this is a wealth of information, we argue that these regions are less intuitive to the general public compared with regions located around a familiar local landmark, such as a train station. Therefore, using landmarks like these as the basis for performing spatial analysis of its surrounding region may provide improved understanding of urban issues. In this paper, we explore the use of Melbourne metropolitan train stations as landmarks to explore the functional relationship between key co-variates such as population, house size, income and dependent variables water and electricity consumption for areas in Greater Melbourne. To achieve this, Thiessen polygons are used to define catchment areas and data is aggregated to these regions surrounding each train station. By presenting the data spatially for these regions in Melbourne, we demonstrate the use of a novel approach to spatial visualisation data. Results show strong associations between income and electricity consumption, and between measures of urban density with resource consumption.