Search

Traffic volumes (and hence energy consumption) from the transport sector continue to rise, yet the potential fundamental role of urban planning in helping to reduce transport energy consumption remains to be poorly understood and hugely underplayed. Current urban planning practice, particularly in suburban areas, tends to increase traffic volumes by dispersing activities and hence facilitates private car travel rather than travel by public transport, walking or cycling. Public transport orientated development as an evolving practice tends to be focused very much on urban areas. This thesis seeks to understand the logic behind travel and suggests that urban planning can be applied more fully, at the strategic and local levels, to reduce energy consumption in car use (at least in the journey to work). The detailed analysis assesses the extent to which the design of the urban environment affects travel behaviour. The research hypothesis is that: "Journey to work travel behaviour generated by new residential development is dependent on a number of land use and socio-economic variables. The strength, significance and range of interaction vary spatially and over time." Within the analysis, the journey to work is used as the dependent variable, and is measured in terms of journey length and time, mode share and composite energy consumption. The independent variables considered include: Land use: resident population density, resident employment density, workplace population density, workplace employment density, resident population size, workplace population size, distance from urban centres and strategic transport networks, jobs-housing balance, resident classification (relative to the urban area), type of journey to work, neighbourhood streetscape design, public transport accessibility, and resident location (relative to the green belt). Socio-economic: household tenure, house type, house size, number of children, car availability, company car ownership, household income, house value, respondent sex, respondent age, marital status, occupation, qualification, attitude to travel, attitude to home and home location, reason for moving home and choosing new home location, relative levels of mobility, and dual income households. The methodological approach is to systematically examine the study hypothesis and a series of related research questions using data from the county of Surrey, UK. The empirical analysis is based on two new household occupier surveys carried out in 1998 and 2001, together with additional, complementary data taken from local authority datasets and the Census 2001. The thesis's particular originality is in providing: An examination of the complexity of the land use and transport interaction field, using energy consumption as the dependent variable and an estimation of the strength and significance of a wide range of land use and socio-economic variables - both previously researched and under researched variables A segmentation of respondents into different groups, such as stayers, inmovers and outmovers, showing the different manifestation of the land use and transport relationship for different groups within society A systematic tracking of the impact of time on the land use and transport relationship, with temporality and adaptation (including "co-location" effects) noted as critical features in travel behaviour, with the analysis controlling for potential attrition factors Analysis of a seldom-studied London fringe/suburban county such as Surrey - much previous work is concentrated on the city or other urban areas. The key findings and recommendations are that each land use, socio-economic and attitudinal variable, when considered on its own or even in small groupings, offers limited explanatory power in explaining travel behaviour. When a number of variables are brought together, including some variables not usually considered in the literature, the explanatory power of the modelling begins to work. Linear regression analysis shows that the land use and socio-economic variables, when considered together, explain 60% of the variation in energy consumption in 1998 and 54% in 2001 and for the stayers data only explain 65% of the variation in energy consumption in 1998 and 54% in 2001. Land use variables by themselves contribute approximately 10% of the variation in transport energy contribution hence a major part of the logic behind travel. In terms of temporal change from 1998-2001, although aggregate distance co-location might occur, aggregate energy consumption is likely to increase due to increased car dependency. Also, focusing on the aggregate trends also hides several detailed kurtosis effects: households located at higher densities, closer to major strategic centres, in areas with good public transport accessibility and strong jobs-housing balance are all likely to reduce their commuting travel distance. Other groups are likely to increase their composite transport energy consumption, for example, the higher income cohorts. Integration thus requires action across a wide range of fields.