Contents lists available at ScienceDirect Journal of Transport Geography journal homepage: www.elsevier.com/locate/jtrangeo A multi-decade longitudinal analysis of transportation and land use co- evolution in the Greater Toronto-Hamilton Area Dena Kasraian ⁎ , Shivani Raghav, Eric J. Miller Department of Civil & Mineral Engineering, University of Toronto, 35 St. George St., Toronto, Ontario M5S 1A4, Canada ARTICLE INFO Keywords: Transport accessibility Urbanization Panel analysis The Greater Toronto-Hamilton Area ABSTRACT Transportation infrastructure networks (TINs), and the accessibility that they provide, are argued to be major determinants of land-use change. However, the spatio-temporal extent of their impact on urbanization is unclear and calls for further investigation. This paper empirically investigates the impact of both road and transit net- works on urbanization in the Greater Toronto-Hamilton Area (GTHA) from 1971 to 2016, using a traffic analysis zone (TAZ) level panel database of 17,160 observations. It first documents and measures the magnitude, location and rate of urbanization and TINs' development in the region and in relation to each other using descriptive and buffer analyses. It then estimates the share of transport accessibility in urbanization over time and space using fixed effects panel regression models. Our findings demonstrate the significant extent of the ongoing sprawl, the substantial historical and current difference between the potential transit and car travel times, and the stability of the potential car travel times despite improvements to the road network. They also highlight the importance of urban proximity in prompting urbanization and the decreasing impact of transit improvement on urban development in urban areas and over time. A policy implication is that to encourage transit-oriented develop- ment, it is more effective to introduce or reinforce connections to activity centers rather than invest in reducing travel times to reach them. 1. Introduction Access to transportation infrastructure networks (TINs) is known as one of the drivers of urbanization, along with factors such as proximity of urban land, demographic and economic growth, and spatial policies (Knight and Trygg, 1977). Improvements in TINs increase accessibility, which results in the redistribution of land use, i.e., the location of ac- tivities. Consequently, change in activity locations and the way people use TINs to access them (i.e., their travel behaviour), induces the need for TIN improvements. This dynamic interaction rests on the relation- ship between urban structure, travel demand and transport accessi- bility. Areas around road transportation have been shown to become attractive for decentralized population and employment centers, leading to expansion of urban areas. Urban sprawl, coupled with re- duced driving costs and the exponential increase in car ownership and use, has potential direct and indirect negative effects on the environ- ment, such as degradation of air quality, greenhouse gas emissions, increased threat of global climate change, degradation of water re- sources, noise and habitat loss (Demirel et al., 2008; Forman et al., 2003). While highways are known to promote urban sprawl, transit stations are used as a tool to encourage compact urban development in their vicinity, though the success of rapid transit is shown to be context- specific and dependent on supporting pro-development public policies (Knight and Trygg, 1977). Developments in land-use patterns and, to a higher degree, TINs are almost irreversible, due to their high sunk costs and costs of replace- ment. This durability has serious policy implications, as TINs not only last for a long time, but could also affect urban development patterns for decades to come. It is critical to measure the impact of TINs on the magnitude, rate and location of land-use change in general, and (sub) urbanization in specific, to evaluate past TIN investments and strate- gically guide future urban growth through integrated land use and transport plans. Such evidence can shed light on where to expect urban growth and to what degree TIN investments can actually be used as a tool to guide it. Yet the extent of TINs' influence on land-use change is unclear and calls for further investigation (Badoe and Miller, 2000; Iacono and Levinson, 2009; Kasraian et al., 2016a). Such investigations need to be longitudinal as the impact of transportation on land use is a slow pro- cess and only observable in the long run. However, long-term in- vestigations spanning over several decades are limited. Previous long- term studies of transportation and land use have mainly evaluated https://doi.org/10.1016/j.jtrangeo.2020.102696 Received 27 June 2019; Received in revised form 19 March 2020; Accepted 22 March 2020 ⁎ Corresponding author at: Department of the Built Environment, Eindhoven University of Technology, 6 De Groene Loper, 5612 AE, Eindhoven, The Netherlands. E-mail addresses: d.kasraian@tue.nl (D. Kasraian), shivani.raghav@mail.utoronto.ca (S. Raghav), eric.miller@utoronto.ca (E.J. Miller). Journal of Transport Geography 84 (2020) 102696 0966-6923/ © 2020 Elsevier Ltd. All rights reserved. T