Linking expected mobility production to sustainable residential location planning: some evidence from Flanders

Based on a set of spatial proximity characteristics this paper develops a model that estimates for every neighbourhood in Flanders (Belgium) the amount of traffic that would be generated by an additional residential unit when socio-economic variables are held constant. The results show that residential density, land use diversity and proximity of facilities influence daily travelled distances when these variables are measured in the immediate vicinity of the residential location of the respondent (within a radius of 1 km). When aggregating these variables at a larger geographical scale, in most cases the impact proves no longer significant. Variables based on the spatial distribution of jobs, or on the global accessibility of the entire population in the study area, do not show any significant effects on the travel distance. Despite the statistical significance only a fraction of the observed variance in reported distances is explained by characteristics of spatial proximity. However, we can assume that the importance of spatial structure in the genesis of mobility patterns will increase in case the cost of transport would rise (cf. peak oil). For this reason, the application of the mapped results of the proposed model could contribute to the practice of sustainable spatial planning. (C) 2010 Elsevier Ltd. All rights reserved.