Land use characteristics of subway catchment areas and their influence on subway ridership in Seoul
Introduction
The factors affecting transit ridership have long been a research topic of interest among scholars and policy-makers in the field of transportation and urban planning. Existing literature on this subject has been carried out with different research motivations. Some studies investigated how both internal factors, such as transit service quality and operation strategies, and external factors, including connectivity with other transport modes, land use, and public policy, affect transit ridership (Fielding, 1995, Pucher, 2004, TCRP, 2007). The research objectives of these studies were to understand determinants of transit patronage under circumstances of declining demand for public transit, and in the face of criticism of transit investment and subsidies and to suggest policy options to improve transit ridership.
Another avenue of research emphasizes the relationship between land use patterns of the transit catchment area and transit ridership. These studies suggest a focus on transit-oriented development (TOD), a dense and mixed-use development with a pedestrian-friendly design, as a means of encouraging transit ridership by maximizing accessibility to public transit. Since Calthorpe (1993) has developed the concept of TOD, there has been a large volume of literature on TOD research, mainly focusing on the US cases. For example, TCRP (2002) made an extensive review on the existing TOD studies in terms of four main topics: 1) institutional issues; 2) evaluation of impacts and benefits; 3) implementation; and 4) urban design. One-half mile radius of a transit station has become the standard catchment area for designing TODs.
Third, in an effort to better forecast future station-level ridership by understanding the factors that influence transit ridership, a direct ridership model was developed as an alternative to the traditional four-step travel forecasting model (Walters and Cervero, 2003, Gutiérrez et al., 2011, Zhao et al., 2013). These studies assert that direct ridership models have several advantages over the traditional model as a result of capturing the influences of the characteristics of the stations and their service catchment areas on transit ridership with relatively low expense and fast response.
Given our research focus on analyzing determinants of transit ridership, this study has dual research objectives: 1) to evaluate the land use characteristics of subway catchment areas in the Seoul metropolitan area (SMA) in terms of TOD principles, and 2) to investigate how the land use patterns of a pedestrian catchment area (hereafter PCA) influence station-level ridership. In doing so, we employ a multinomial logit model for analyzing the land use characteristics of PCAs, while we use stepwise regression models and mixed geographically weighted regression (MGWR) models to examine the relationship between the land use characteristics of PCAs and subway ridership.
This paper consists of three sections. The first section reviews the existing literature, focusing on theoretical and empirical studies of factors affecting transit ridership. The second section explains the results of our multinomial logit models for characterizing land use patterns of the PCAs. The third section presents results of stepwise regression and mixed geographically weighted regression (MGWR) models, followed by conclusions and policy implications.
Section snippets
Literature review
A considerable body of empirical studies has investigated determinants of transit ridership over the last several decades. Some studies have suggested a positive correlation between population density and transit ridership (Pushkarev and Zupan, 1977, Frank and Pivo, 1994, Nelson/Nygaard Consulting Associates, 1995, Seskin et al., 1996, Cevero, 1996, Messenger and Ewing, 1996, Cervero and Kockelman, 1997;), while others examined the roles of specific land use features, such as density, design
Data and PCA definition
Seoul is known for its extensive and high-quality subway system, which consists of 17 lines covering about 600 miles and includes 493 stations, as shown in Fig. 1. Subway service began in 1974 following the completion of line 1. The system carries a daily average of 9.8 million passengers, making it one of the busiest subway systems in the world.
The primary data source for this research is the ‘jipgyegu’ dataset for the SMA from the Korea Statistical Office, which contains land use,
Land use and socio-economic characteristics of PCA
The multinomial logit model was employed to analyze land use and the socio-economic characteristics of the PCAs. As the dependent variable of the logit model, four alternative choices were used in this study: core PCA (within 300 m), primary PCA (300–600 m), secondary PCA (600–900 m), and the rest of the region.
On the basis of the explanatory variables identified through our literature review, four types of variables were selected for characterizing built environments: land use,
Determinants of subway ridership in the SMA
To achieve the second objective of this research, investigation of the relationship between built environments and subway ridership, we conducted two statistical analyses: stepwise regressions and MGWR approaches. In doing so, we first built a station-level dataset for each PCA. We used the natural log of daily station-level ridership, which is the sum of boardings and deboardings, as the dependent variable. In addition to the land use and socio-economic variables used in the multinomial logit
Conclusion: implications for policy
This study aimed to evaluate built environments surrounding Seoul's subway stations in terms of TOD principles and to examine their influence on subway ridership. The major findings can be summarized as follows. First, the built environments of subway PCAs in Seoul were compatible with TOD principles in terms of density and diversity. They have declining density gradients of population and employment extending outward from stations and a higher level of mixed land use in the core and primary
Funding
This work is supported by the National Research Foundation of Korea grant funded by the Korea government (MSIP) (NRF-2010-0028693)
References (43)
- et al.
A comprehensive analysis of built environment characteristics on household residential choice and auto ownership levels
Transp. Res. B
(2007) - et al.
The influence of land use on travel behavior: specification and estimation strategies
Transp. Res. A Policy Pract.
(2001) - et al.
Application of geographically weighted regression to the direct forecasting of transit ridership at station-level
Appl. Geogr.
(2012) - et al.
Travel demand and the 3Ds: density, diversity, and design
Transp. Res. Part D: Transp. Environ.
(1997) - et al.
The relationship between urban form and station boardings for Bogota's BRT
Transp. Res. A Policy Pract.
(2008) Congestion pricing and the future of transit
J. Transp. Geogr.
(1995)- et al.
Transit ridership forecasting at station level: an approach based on distance-decay weighted regression
J. Transp. Geogr.
(2011) - et al.
Walk the line: station context, corridor type and bus rapid transit walk access in Jinan, China
J. Transp. Geogr.
(2012) - et al.
Factors influencing light-rail station boardings in the United States
Transp. Res. A Policy Pract.
(2004) - et al.
Rail-based transit-oriented development: lessons from New York City and Hong Kong
Landsc. Urban Plan.
(2010)
Factors generating boardings at metro stations in the Seoul metropolitan area
Cities
Transit-oriented development in a high-density city: identifying its association with transit ridership in Seoul, Korea
Cities
What influences Metro station ridership in China? Insights from Nanjing
Cities
Logistic regression using the SAS system: theory and application cary
Public finance and transit-oriented planning: new evidence from southern California
J. Plan. Educ. Res.
The Next American metropolis: ecology, community, and the American Dream
Induced travel demand: research design, empirical evidence, and normative policies
J. Plan. Lit.
Transit-based housing in the San Francisco Bay area: market profiles and rent premiums
Transp. Q.
Transit station access trips and factors affecting propensity to walk to transit stations in Bangkok, Thailand
J. Eastern Asia Soc. Transp. Stud.
An analysis of Metro ridership at the station-to-station level in Seoul
Transportation
Ridership models at the stop level
Cited by (212)
Causality between multi-scale built environment and rail transit ridership in Beijing and Tokyo
2024, Transportation Research Part D: Transport and EnvironmentExploring the long-term threshold effects of density and diversity on metro ridership
2024, Transportation Research Part D: Transport and EnvironmentAnalyzing spatiotemporal distribution patterns of metro ridership: Comparison between common-class and business-class carriage service
2024, Journal of Transport GeographyAnalysis of the relationship between metro ridership and built environment: A machine learning method considering combinational features
2024, Tunnelling and Underground Space TechnologyImpacts of characteristics of service facilities in metro station area on housing prices
2024, International Journal of Transportation Science and TechnologyModeling the resilience of interdependent networks: The role of function dependency in metro and bus systems
2024, Transportation Research Part A: Policy and Practice