Transit-Oriented Development in the Sun Belt

Abstract

Although existing transit-oriented development manuals are chided for lack of empirical foundation and what is known from the empirical literature about the relationship between land use-urban design and transit ridership is reviewed. More is needed to be known about this relationship. What was done in the first of two studies of transit ridership in Dade County is reported and what was learned of relevance to transit-oriented development is summarized. Bus mode share by place of residence proved primarily dependent on automobile ownership and secondarily on jobs-housing balance and bus service frequency. Automobile ownership, in turn, proved dependent on household income, overall density, and transit access to downtown. Thus, three types of variables—sociodemographic, land use, and transit service—were found to affect bus use through a web of interrelationships. Likewise, bus mode share by place of work proved dependent on the cost of parking, transit access to downtown, and overall density, again through a web of interrelationships. By using equations thus derived, densities required to support different levels of transit service and productivity were computed. It takes 8.4 dwellings per acre to support 25-min headways at the transit operator's minimum productivity level and 19.4 dwellings per acre at the systemwide average productivity. As household incomes, parking costs, or other local conditions vary from the county average, required densities rise or fall accordingly. In any event, densities are likely to be well above the old industry standard of seven units per acre.