The aim of this study was to explore the spatial variation in household greenhouse emissions from local transportation in Trondheim, the reasons for this variation, and explore scenarios of what bearing these variations might have on greenhouse gas emissions in the future. Data from a national travel survey was used together with modal emissions coefficients to model the average emissions per capita for 46 geographic zones in Trondheim. Linear regression was used to explain the variation in average emissions using a number of explanatory variables identified from the literature. The regression models explained around 75–80 per cent of the spatial variation in average emissions (0.75 ≤ adj r2 ≤ 0.79), with centre distance explaining the majority of variation. Using a regression function containing centre distance and access to public transport as explanatory variables, five scenarios were constructed for emissions in 2030, which suggest that centralisation of new residential building developments and improvement in the public transport network could limit the growth in annual greenhouse gas emissions to approximately 10 per cent in the presence of approximately 30 per cent population growth.