1 VKT LEVEL AND DISTRIBUTION OF LAND USES SOLTANI, Ali1 Urban Planning Department, Faculty of Art and Architecture,

Shiraz University, Iran [email protected] Abstract There is considerable current interest in the effects of urban form and land use characteristics on the travel activities made by people. The underlying assumption is that built environment characteristics have an important influence on making shorter travel and a person's willingness to travel by public transport, ridesharing, walking or bicycling -modes other than driving alone. A moderate set of strategies have been proposed or implemented by both Federal and State Governments in US, Europe or in Australia. For instance, SA s Department for Transport, Energy and Infrastructure (DTEI) (2000) has made a connection between transport choice and urban design by providing a detailed set of guidelines covering microscale urban form attributes that may impact people in choosing their mode of travel. However, relatively little empirical work has been done to evaluate the real effects of land use and development strategies on travel behaviour in developing countries. This paper is organised as follows: the first section, reviews some previous empirical studies in the topic of the interaction between built environments and travel characteristics. The second section of this paper describes the methodological approach applied. Findings of the former studies are presented in the third section. The effects of various exogenous factors for work and non-work activities (i.e. shopping, recreation, medical and personal business) on VKT were discussed separately. Also, two distinct groups: inner suburbs versus outer suburbs were compared together. The final section describes the conclusions drawn from this study. 1 Review of past studies There is a vast amount of literature from around the world on the impacts of urban form on travel characteristics. Much of the work has originated in either North America or Western Europe. While a growing body of research have been undertaken on this topic, the findings are not consistent, thus it is hard to say overtly what we know and what don t know from the existing literature. These can be explained by variations in research design, spatial scale of the analysis, and geographical setting. The majority of these researchers conclude that urban form characteristics, ranging from regional to local in scale, have an effect on travel patterns and consequently the environmental impacts of transport such as emissions. Full reviews of the literature have been undertaken by Handy (1996); Badoe and Miller (2000); Crane (2000); and Stead and Marshall (2001). This review includes only empirical studies. The empirical studies deal with real examples and rely on fewer assumptions than simulation works, and also they are easy to understand and can be applied in planning and policy making. Most past studies attempt to find out the influence of urban form factors on more than one response variable. Also they tend to focus on the relative significance of the explanatory variables on travel behaviour. It is more difficult to extract the different effects of urban form variables on different travel response variables. The four response variable
Research paper presented at The 1st International Conference on Built Environment in Developing Countries (ICBEDC 2007), 1-2 December 2007, Universiti Teknologi Malaysia, Pulau Pinang, Malaysia

2 categories summarise the foci of most of the studies. These variables include trip frequency; modal split; distance travelled (or VKT); and energy consumption/emissions.
2 Comparative studies of development type

Development pattern is effectively a composite variable (or a set of variables) which is used to characterise areas of cities that are relatively homogenous to a range of attributes. These attributes could be the age of development (e.g. pre-war or post-war), the style of development (e.g. traditional, conventional) and the street network design (e.g. grid or curvilinear). Figure 1 provides an example:

Figure 1: Comparing two neighbourhood development paradigms: disconnected segregated modern development (Para Hills on left) versus connected mixed use traditional development (Norwood at right)

The reason for choosing this approach is that neighbourhood design features need to be considered as a balance of attributes. Otherwise considering them individually may not present a significant effect on travel behaviour. For instance, the analysis of four neighbourhoods in the San Francisco Bay Area found no evidence that individual residential design elements have an effect on travel behaviour, rather neighbourhood design as a whole is an important determinant of whether residents perceive walking as an option and affects pedestrians activity in a community (Handy 1993). Handy believed that in different areas, different types of design elements may be appropriate or fit into the neighbourhood environment differently. However, the importance of comprehensive design rather than identifying a specific design element works in all places. Furthermore, it is difficult to sort out the effects of land use mix and urban design because they are strongly correlated with density. Thus some studies tested the effects of density, diversity, and design (3D s), all together on travel behaviour. Table 1 details some examples of comparative studies based on the development pattern type. While countless studies are available, this table includes only samples varied in terms of geographical location, research methodology and empirical findings. From the studies summarised in this table, it can be claimed that suburban dispersed neighbourhoods generate a higher trip rates; higher share of car use; lower non-motorised travel; higher VKT and greater energy use/emissions comparing to traditional compact suburbs with more pedestrianfriendly environments. However, this ruling is not explicit to make a generalisation for other geographical areas. In

3 fact, the relationship between urban form and travel behaviour is more complex than it may seem on the surface. It is sometimes difficult to distinguish between suburban and traditional suburbs. In fact, if suburban development encompasses pedestrian-friendly elements such as coordinated urban design, and close proximity to facilities then there will not be much difference between two development paradigms. In such case, it might be better to discover the impacts of individual elements of physical form on travel behaviour. This would help to find out the relative importance of variations between neighbourhoods as well as variations within neighbourhoods. On the other hand, the presence of non-physical factors such as socioeconomics or personal attitudes may work better in influencing travel patterns. Moreover, the impact of regional factors such as regional accessibility to job may significantly contribute to change travel patterns. Therefore, it is required to conduct an analytical investigation taking into account most potential factors to find out the real impact of physical features in affecting travel behaviour.
3Travel distance studies

The estimation of travel distance from urban form has received generally some attention in the literature, although most attention having been given to estimating travel distance from travel-related factors. However, a number of studies can be found on discovering the association of land use variables and travel cost factors such as travel time and distance. White (1988) empirically demonstrated that, given the distribution of workplaces and residences, households and workers locate to minimise commuting. Small and Song (1992) showed that White s results are dependent on the degree of disaggregating origin and destination zones. When zones are finely disaggregated, they show that about two-thirds of all commuting in Los Angeles is surplus. Kockelman (1998) through regressing travel distance with measures of urban form in San Francisco, found that the number of choices a trip-maker has, increases with the distance travelled. The evidence suggests that people travel farther than they need to; this may very well be because they wish to expand their choice set of activity sites and thereby increase the expected quality of the activity they do engage in, at their chosen sites. For example, while one probably will travel only to the closest of a very specific activity type (such as eating out at a fast-food restaurant), one will not often travel to the closest dining establishment. The author suggested that as long as the marginal value of travel time plus the monetary cost of travel remains below the marginal value of increased opportunities brought about by travelling further, a trip-maker (as the utility maximising traveller) can be expected to lengthen their journeys. Naess (2000) using multivariate analysis of the influence of various urban form variables on the total distance travelled discovered that residential location is a significant factor in commuting behaviour. The location of the dwelling relative to the town centre of the study area (Danish Frederikshavn) is the factor that exerts the strongest influence on both the total distance travelled during the week of investigation, and the travel distance by car. Redmond and Mokhtarian (2001) using data from randomly-selected households in three neighbourhoods of the San Francisco Bay Area found that the suburban explanatory variable was highly significant. As indicated in studies of land use and travel and, in particular, travel patterns associated with suburban development, commutes tended to be longer for those living in suburbs. Schwanen et al. (2002) showed that workers living in decentralised regions commute longer than residents of centralised and selfcontained region. In addition, commuting distance is affected inversely by job density in the area.

4 Criticisms of urban form-travel link appeared after the publication of Newman and Kenworthy s leading researches (1989; 1999) on the negative correlation between density and car dependency/gasoline consumption. Gordon et al. (1991) argue that individual households prefer to avoid the time penalties caused by the extensive congestion in mono-centric urban areas by systematically changing their job or residence. This residential and workplace changing provides them opportunities to travel shorter distance or make use of less congested routes. In addition, employment centres seek ways to escape the disadvantages of highdensity locations such as congestion, high land values, and limited choices for spatial expansion. In aggregate, the dispersal of activities across the metropolitan area leads to decline in travel distances and time. In a literature review, Ewing (1997) discusses the characteristics, causes, and costs of sprawl, refuting pro-sprawl arguments by Gordon and Richardson (1997). There are explicit costs of sprawl include longer commutes, energy consumption, air pollution and loss of open space, among other things. For example, "households living in the most accessible locations spend about 40 minutes less per day travelling by vehicle than do household living in the least accessible locations". It can be argued that the studies by Gordon et al. (1991; 1997) limited the influence of metropolitan structure on travel distance to the impact of mono-centrism and poly-centrism, thus the importance of other dimensions of spatial structure such as population and employment density were ignored. This neglect has been found in other studies e.g. Giuliano and Small (1993) resulted in claim that travel distance and time (in a mono-centric spatial structure) cannot be explained by the distribution of housing and jobs alone. Giuliano and Dargay (2003) investigated travel trends and land use characteristics in both US and British cities and suggested that improved income, demographics and economics explain travel trends. Therefore, urban spatial characteristics have little impact on travel behaviour. However, Giuliano and Dargay (2003) considered aggregated attributes and measures of land use and travel, but failed to consider: the finer characteristics of neighbourhoods (such as provisions for various forms of travel, neighbourhood character and its conduciveness for non-motorised forms of travel); the socio-demographics of the population (such as the difference in travel patterns of individuals and households); and the type of travel being undertaken. Understanding how self-selection and the built environment affect travel behaviour is important because they relate to land use policies intended to reduce the VKT level. Residents choose localities to go with their desire for walking, bicycling or transit use to those neighbourhoods more beneficial to such behaviour. This suggests that differences in VKT level among households with different neighbourhood design should not be attributed to the neighbourhood design alone. Rather, the differences should be accredited to self-selection . In other words, individuals who are likely to have lesser VKT, select a place in where they have a better opportunity of engaging in non-motorised travel. The study by Handy et al. (2006) confirmed the self-selection issue; people do self-select; their attitudes and preferences affect their activity level and their choice of neighbourhood. On a more complex level, however, the study found that the built environment had an impact on walking behaviour even after attitudes and preferences were accounted for. This short review of literature serves to illustrate that the effects of spatial patterns on travel distance are not undisputed. It seems that the relationship between urban form and travel behaviour is more complicated than originally thought. While some scholars have argued that the relationship is marginal (Brunton and Brundle 1998), others suggest finding better ways to study on urban form-travel interactions (Handy 1996). However, the role of several

5 potentially important factors, such as the degree of centralisation needs to be investigated further.
Table 1: Studies comparing neighbourhood development designs.

Source, Location

Development Patterns and Attributes

Key Findings
-Standard neighbourhoods generated a 25% higher average number of trips - Traditional neighbourhoods generated a lower proportion of car trips but higher transit and other modes

Friedman et al. (1994), San Francisco, US

PBQD (Parsons Brinckerhoff Quade & Douglas) (1996), San Francisco Bay Area, US Cervero and Gorham (1995), San Francisco, US

Traditional neighbourhoods

Non-traditional neighbourhoods

- Density is the most significant physical factor in explaining modal choice; -The residents of traditional neighbourhoods were more interested in no auto-based modes for non-work trips ; - There was not any explicit relation between urban variables and mode choices for non-work trips. - Transit neighbourhoods showed a higher transit use - The mode split depends on the character of both neighbourhood and its larger region

Transit neighbourhood

Car neighbourhood

Handy (1992), Austin, US

Pre-WWII/high local accessibility -Rectilinear grid network - Smaller blocks and more intersections - More direct routes - Traditional downtown area

Post-WWII/low local accessibility - Curve and irregular network more cul-de-sacs - Very few connections to arterial roads - Car oriented commercial centres

- For travel characteristics, the variation between the neighbourhoods was significantly greater than the variation within neighbourhoods - Not any effect on strolling activity by neighbourhood type - The regional context played an important role in influencing modal split

Rutherford et al. (1996), Seattle, US

Mixed use suburb

Homogenous suburb

Diepen and Voogd (2001), Netherlands

Inner city

Outer city (traditionally built / sustainable built )

Perkins (2002), Adelaide

City compact infill development

Sprawl fringe development

- The residents of mixed use suburbs travelled fewer personal kilometres than people living in non-mixed use suburbs. - Walk share is higher in mixed use suburbs - Households in both outer suburbs have a higher energyintensive travel activity than households in the inner city suburbs. - The distinction of outer city neighbourhoods into traditionally and sustainable built ones does not generate remarkable differences. - The choice of infill development over fringe development produced more significant energy savings than the combination of a number of other demand reduction measures

4 Modelling distance travelled Conceptually, the issue of travel distance modelling can be considered as trying to answer the question of how many kilometres (miles) will be travelled by an individual trip-maker. A number of methods have been developed over the years to study travel distance as one aspect of travel behaviour. These include the use of simple or multivariate regressions at an aggregated level (zonal or regional) or disaggregated level (individual or household) thus regression models remain for the most part the state of current practice. Several examples can be found in the literature that applied multiple regression to explain distance travelled or personal VKT level (Farthing et al. 1996; Mokhtarian and Salomon 1996; Cervero and Kockelman 1997; Ibrahim 1997; Kockelman 1998; Miller and Ibrahim 1998; Giuliano and

6 Dargay 2003; Choo 2005; Kwon and Preston 2005). For a given VKT category (total or purpose), motorised trips undertaken by a sample individual were aggregated across the metropolitan region. Note that the dataset was filtered by eliminating trips of 65 kilometres or longer trips to and from other states or rural area. The natural logarithm of the total number of kilometres in the category is considered as a measure of distance. The VKT is considered for only sampled trip makers with a vehicle. The natural logarithm of the sum of kilometres was selected after trying several other transformations. This transformation expresses the diminishing marginal impact of distance, reducing the impact of longer distances (Giuliano and Dargay 2003). If ordinary distances are used, respondents having very long trips exert a stronger influence and the estimated travel distances increase by about 50 km in the centre as well as in peripheral areas. It is assumed that the VKT is influenced by three main factors or groups of factors: socio-economic; location (urban form); and level of service. This implies that the amount of distance travelled by the individual will depend on his/her own socioeconomic status (income, occupation, age and gender), and on that of the individual s household (overall household income, household size and structure), but also on factors external to the household. Local and regional urban form variables provide an approximate indicator of likely availability of facilities for which travel may be undertaken. Level of service provides a degree of convenience to achieve the desired facility. It is a combination of mode availability, accessibility to the mode, and characteristics of the mode such as frequency, speed, cost, comfort, etc. 5 Conclusions and further investigation While separation of land uses and market forces have tended to segregate activities and lengthen trips, bringing origin and destinations closer together is associated to decrease daily kilometres travelled for work and non-work activities. Better urban form would locate workplaces and activities linked as trip productions and attractions are as close together as possible. Therefore, providing more alternatives to automobile usage and improving the selfcontainment of the residential suburbs are necessary conditions that should be given more attention in policy and planning. An indirect policy implication of these results is that planners could possibly achieve better outcomes by relaxing locational constraints on jobs and residences (especially zoning restrictions on the location of businesses, where such restrictions do exist). However, restrictions such as large lot suburban zoning may achieve other efficiencies. But such regulations may lead to undue travel and congestion by causing jobs to remain too centralised (Anas 1998). The methodology and the data used could be refined by: including some real urban cases; taking a sample of household travel survey in addition to working towards a better operationalisation for the analysis of urban form. They would allow a finer grained analysis (to detect influences that this research did not discover). References Anas, A. and Xu, R. (1998). Congestion, Land Use, and Job Dispersion: A General Equilibrium Model . Journal of Urban Economics (45): 451-473. Australian Bureau of Statistics (ABS) (2001). Census of Population and Housing: Journey to Work datasheets. Australia. Badoe, D. A. and E. J. Miller (2000). "Transportation-land-use interaction: empirical findings in North America, and their implications for modeling." Transportation Research Part D

7 vol. 5: 235-263. Cervero, R. and R. Gorham (1995). "Commuting in Transit Versus Automobile Neighborhoods." Journal of American Planning Association(61): 210-225. Cervero, R. and K. Kockelman (1997). "Travel Demand and the 3D's: Density, Diversity, and Design." Transportation Research D 2(3). Choo, X. (2005). "Wanting to travel, more or less: Exploring the determinants of the deficit and surfeit of personal travel." Transportation 32: 135-164. Crane, R. (2000). "The influence of urban form on travel: An interpretive review." Journal of Planning Literature 15(1): 3-22. Diepen, A. V. and H. Voogd (2001). "Sustainability and planning: does urban form matter?" International Journal of Sustianable Development 4(1). Ewing, R. (1997). "Is Los Angeles-style sprawl desirable?" Journal of the American Planning Association 63(1): 107 126. Farthing, S., T. Winter, et al. (1996). Travel behaviour and local accessibility to services and facilities. The Compact City. M. Jenks, and E. Burton, K. Williams. Oxford, Oxford University Press: 181-189. Frank, L. and G. Pivo (1994). "The impacts of mixed use and density on the utilization of three modes of travel: the single occupant vehicle, transit and walking." Transportation Research Record 1466: 44-52. Frank , L. D., M. A. Andresen, et al. (2004). "Obesity Relationships with Community Design, Physical Activity, and Time Spent in Cars." American Journal of Preventive Medicine 27(2): 87 96. Friedman, B., S. Gordon, et al. (1994). "Effect of neotraditional neighbourhood design on travel charactristics." Transport Research Record 1466: 63-70. Giuliano, G. and N. Dargay (2003). "Another look at travel patterns and urban form: the US and Great Britain." Urban Studies, Oct, Vol. 40(11): 82-95. Giuliano, G. and K. A. Small (1993). "Is the journey to work explained by urban structure?" Urban Studies 30: 1485-1500. Gordon, P. and H. Richardson (1997). "Are compact cities a desirable planning goal?" Journal of American Planning Association 63: 95-106. Gordon, P., H. W. Richardson, et al. (1991). "The Commuting Paradox Evidence from the Top Twenty." Journal of the American Planning Association 57(4): 416-420. Handy, S. (1992). "How land use patterns affect travel patterns." Council of Planning Librarians, Chicago. Handy, S. (1993). "Regional versus local accessibility: neo-traditional development and its implications for non-work travel." Built Environments 18(4): 256-267. Handy, S. (1996). "Understanding the link between urban form and non work travel behaviour." Journal of Planning Education and Research 15: 183-198. Handy, S., X. Cao, et al. (2005). "Correlation or Causality Between the Built Environment and Travel Behavior? Evidence from Northern California." Transportation Research Part D 10(6): 427-444. Hanson, S. (1980). "The Importance of the Multi-Purpose Journey to Work in Urban Travel Behavior." Transportation 9: 229-248. Hillier, B. (1983). "Space Syntax." Architects Journal 178(48). Hodge, D. (1991). Development of Method of Analysis for Planning Transit Systems Components in and Around Major Activity Centers. Part 1: Trip Chaining. The Behavioral Basis for the Design of Circulation Systems for Major Activity Centers, Transportation Northwest, Seattle, University of Washington. Ibrahim, A. (1997). Investigation of the Relationship between Urban Spatial. Structure

8 and Travel Demand in the GTA. Civil Engineering. Toronto, University of Toronto. MSc. Kitamura, R., P. L. Mokhtarian, et al. (1997). "A micro-analysis of land use and travel in five neighbourhoods in the San Francisco Bay Area." Transportation 24: 125-158. Knaap, G. J., Y. Song, et al. (2004). Seeing the Elephant: Multi-disciplinary Measures of Urban Sprawl. Maryland, National Center for Smart Growth Research and Education at the University of Maryland. Kockelman, K. M. (1998). A Utility-Theory-Consistent System-of-Demand-Equations Approach to Household Travel Choice. Department of Civil and Environmental Engineering, University of California at Berkeley. PhD. Krizek, K. J. (2003). Operationalizing neighbourhood accessibility for land use-travel behaviour research and regional modelling. Urban Planning & Design. Minneapolis, University of Minneapolis. PhD. Kwon, T. H. and J. Preston (2005). "Driving Forces behind the Growth of Per-Capita Car Driving Distances in the UK, 1970-2000"." Transport Reviews 25(4): 467-490. Meurs, H. and R. Haaijer (2001). "Spatial structure and mobility." Transport Research Part D(6): 429-446. Miller, E. J. and A. Ibrahim (1998). "Urban form and vehicular travel: some empirical findings." Transport Research Record 1617: 18-27. Mokhtarian, P. L. and I. Salomon (1996). "Modelling the choice of telecommuting." Environment and Planning A(28): 1877-1894. Naess, P. (2000). Residential location and transport in a small Danish town: a contribution to the discussion on the influence of land use on travel behavior. Transport and tourism in sociology and geography, Roskilde. Newman, P. and J. Kenworthy (1989). Cities and Automobile Dependence: An International Sourcebook. Aldershot, Gower. Newman, P. and J. Kenworthy (1999). Sustainability and Cities: overcoming automobile dependence. Washington DC, Island Press. PBQD (Parsons Brinckerhoff Quade & Douglas, I., et. al), (1996). Transit and Urban Form. Transit Cooperative Research Program Report. Washington, D.C., Transportation Research Board. Perkins, A. (2002). The influence of urban form on life cycle transport and housing energy and greenhouse gas emissions. Planning. Adelaide, University of South Australia. PhD. Planning SA (2000). Transport Choice and Urban Design: Design Issues for Accessible Neighbourhoods. Adelaide, Government of South Australia. Primerano, F. (2004). Development of Accessibility Measures for Transport and Urban Planning. Transport Systems Centre (TSC). Adelaide, University of South Australia. PhD. Radford, N., A. Chiaradia, et al. (2005). Critical mass: Emergent cyclist route choice in central London. the 5th Space Syntax Symposium, Delft, Holland. Redmond, L. S. and P. L. Mokhtarian (2001). "The positive utility of the commute: Modeling ideal commute time and relative desired commute amount." Transportation 28(2): 179-205. Rutherford, G. S., E. McCormack, et al. (1996). Travel Impacts of Urban Form:Implications from an Analysis of Two Seattle Area Travel Diaries. theTravel Improvement Program Conference on Urban Design, Telecommuting and TravelBehavior, Williamsburg VA. Schwanen, T., F. M. Dieleman, et al. (2002). The Impact of Metropolitan Structure on Commute Behavior in the Netherlands. A Multilevel Approach The Fifth Symposium of the International Urban Planning and Environment Association, Oxford. Small, K. and S. Song (1992). "Wasteful Commuting: A Resolution." The Journal of Political Economy 100(4): 888-898.

9 Sommer, B. and R. Sommer (1997). A Practical Guide to Behavioral Research: Tools and Techniques. Oxford, Oxford University Press. Stead, D. and S. Marshall (2001). "The Relationships between Urban Form and Travel Patterns. An International Review and Evaluation." EJTIR 1(2). White, M. J. (1988). "Location choice and commuting behavior in cities with decentralized employment." Journal of Urban Economics 24: 129-152.

This document was created with Win2PDF available at http://www.daneprairie.com. The unregistered version of Win2PDF is for evaluation or non-commercial use only.