Planning nodes, places, and pedestrian experiences in mountainous cities: an empirical transit station assessment tool

Abstract

Introduction:

In the context of ongoing discussions in Chongqing (China) about urban development strategies for the city’s transit system, this paper introduces an empirical framework for assessing the development of urban transit stations in mountainous cities. Cities in mountainous areas possess unique natural topography, development patterns, cultures, and natural resources, leading to distinct urban development characteristics compared to cities built on plains.

Methods:

Drawing on the node-place modelling literature, we develop a multidimensional station assessment methodology adapted for mountainous cities. By adding the dimension of pedestrian experience, we propose indicators that represent the unique challenges of accessing stations in such terrains that are not typically reflected in conventional node-place analysis.

Results:

Our findings reveal station-specific development opportunities in greater detail and can guide more targeted planning for land use around stations.

Discussion:

Our assessment method is particularly useful for cities facing terrain challenges that impact pedestrian experience.

1 Introduction

The municipal government of Chongqing has outlined their 14th Five-Year Plan (Chongqing Housing and Urban-Rural Development Commission, 2022), aiming to transform Chongqing into a metropolis by fully developing a “rail-based metropolitan area” with a “one-hour commute circle” (Chongqing Municipal Government, 2020). Existing transit lines will be extended and connected, with new lines being planned to stimulate the growth of specific urban areas. The plan is driven by the region’s rapid development and strongly associated with Chongqing’s mountainous spatial structure. Chongqing is characterized by a “multi-center, multi-cluster” city layout largely due to the scarcity of flat land (Gao et al., 2023). Dispersed urbanization requires and will be enhanced by strong connections between clusters, with public transport often seen as a key solution. However, effectively coordinating land use with transit development remains a challenge. This is particularly evident in newly constructed stations, where land use planning often fails to keep up with or meet the development needs at these stations.

Future transit development in Chongqing will focus on identifying stations with the potential for further land use development. This potential is determined by two main factors: (1) the accessibility of the location by transit, buses, urban traffic, and pedestrians, and (2) the employment opportunities, resident population, and development intensity within the station’s coverage area. Existing “node-place” models (Bertolini and Spit, 1998; Zweedijk and Serlie, 1998; Bertolini, 1999) describe how well land use and transit development are coordinated within station areas. These models often classify stations into empirically informed typologies (Peek, 2006; Chorus and Bertolini, 2011), helping to identify stations that are ready for development or revitalization. This approach guides urban transportation planning to promote more balanced, integrated, and transit-oriented sites (Bertolini, 2008; Chorus and Bertolini, 2011).

In the context of mountainous cities, the conventional node-place model fails to adequately consider the importance of pedestrian experience to and from the node and the place. While subsequent studies have incorporated walkability dimensions into the original model (Schlossberg and Brown, 2004; Vale, 2015; Higgins and Kanaroglou, 2016; Lyu et al., 2016; Jeffrey et al., 2019), they still fall short of capturing pedestrian access conditions in cities built on mountainous terrains. In mountainous cities, like Chongqing, the topography features hilly landscapes with steep slopes, frequent elevation changes, fragmented sidewalks, and winding walking paths adapted to the terrain (Gao et al., 2023). These factors pose challenges to pedestrian movement, often resulting in longer walking times for equivalent distances compared to flat cities. This necessitates new indicators and contextualized empirical models to accurately reflect the dynamics.

This paper introduces new indicators and empirical typologies specifically adapted from node-place models for mountainous cities. Building on existing node-place models, some of which include additional assessment dimensions beyond the node and the place, we propose the dimension of pedestrian experience. This dimension reflects pedestrians’ experiences in terms of comfort, effort, and aesthetic elements, and we have devised corresponding indicators to represent them. Based on our analyses, we refine the existing station categories in local transportation plans, which are broad at the strategic level, by introducing a more detailed classification that highlights the development priorities for each station.

The rest of this paper is organized as follows. Section 2 reviews the literature on node-place models and previous studies on pedestrian accessibility. Section 3 introduces our study area, research methods, and explains our indicators in detail. Section 4 presents our station typologies based on cluster analysis, summarizes station-specific accessibility profiles, and elaborates on model improvements using four exemplary cases. Sections 5 and 6 provide discussions and conclusions, illustrating the implications of our findings for other mountainous cities and their planning practices.

2 Literature review

The node-place model integrates land use and transportation planning by analyzing the relationship between a station’s transport node function (connectivity and accessibility) and its place function (local land use and activities; Bertolini, 1999; Reusser et al., 2008; Chorus and Bertolini, 2011; Zemp et al., 2011; Cummings and Mahmassani, 2022). By comparing node and place values, it distinguishes typical situations for a station’s area - balance, stress, dependence, unbalanced node, and unbalanced place, to guide urban development and transit-oriented planning. Previous applications of the node-place model are usually based on specific case studies and serve two primary functions: First, they generate station typologies for proposing tailored station development plans. For instance, Bertolini’s (1999) study of nearly 1700 stations found most in the “unbalanced place” category, suggesting limited opportunities for new development due to high density and diversity (Reusser et al., 2008; Zemp et al., 2011). In Rio de Janeiro, the model classified new train line stations primarily as “balanced” or “dependent,” which invites additional land use development to maintain pace (Gonçalves and Portugal, 2008). In Tokyo, most stations showed good equilibrium, with few “unbalanced” nodes or places that require investments in pedestrian space and easier transfer from trains to highway buses (Chorus and Bertolini, 2011). Second, they visualize the performance of stations to allow visual comparisons between stations. Such visualizations usually take the shape of scatter plots where node values are on one axis while place values on the other, or polar graphs where additional dimensions are included and plotted along scaled axes with a common origin. Caset et al. (2019) provide an overview of different types of polar graphs that include a 5-dimensional kite model (Stadsregio Arnhem Nijmegen, 2011), an eight-dimensional web diagram (Singh et al., 2017) and a rose diagram for directional representation of multi-dimensional data (Groenendijk et al., 2018; Caset et al., 2019).

Existing studies have enriched the set of indicators for the node and place components of the node-place model, as detailed in Table 1. Node values typically include the number of train connections, the number and type of feeder transportation connections, parking capacity, and street connectivity. Place values are often presented through Ewing and Cervero’s “5Ds” framework, which includes Density, Diversity, Design, Destination accessibility, and Distance to transit (Ewing and Handy, 2009; Ewing and Cervero, 2010; Giles-Corti et al., 2016; Zhang et al., 2019). The shared objective of these models, regardless of their indicators and application context, is to empirically inform policy discussions. Considering that most node and place indicators focus on a limited set of physical dimensions related to transportation and land development, often missing nuanced details about people’s experiences and behaviors, recent studies have added more dimensions to highlight demand-side constraints and needs (Caset et al., 2019). These new dimensions include walkability (Jeffrey et al., 2019), accessibility (Cummings and Mahmassani, 2022), ridership (Cao et al., 2020), orientation (Lyu et al., 2016), transport network centrality (Dou et al., 2021), social contact (Zhou et al., 2023), and so on. Emerging big and open data sources such as OpenStreetMap (OSM), the online map Points of Interest (POI) were increasingly employed to describe the dynamic activity in the physical environment (Dou et al., 2021; Zhou et al., 2023).

Walkability can be measured as a function of various indicators describing the structure of the road network, the land use characteristics of a neighbourhood, proximity to jobs, services, and public transport (Giles-Corti et al., 2016), visual attractiveness, comfort, infrastructure quality (Gori et al., 2014), and pedestrian safety (Macioszek and Wyderka, 2021; Macioszek et al., 2023). Researchers may use half-mile or quarter-mile buffers around public transportation stations to describe accessible service areas, based on an acceptable walking distance (El-Geneidy et al., 2014; Vale, 2015; Singh et al., 2017). Network travel time is also used to describe reachable areas, such as estimating the time it takes to walk from one point to another within a given urban network, accounting for factors such as street layout, the presence of natural barriers, and walking speed (Monajem and Nosratian, 2015; Cong et al., 2022). Vale (2015) and Lyu et al. (2016) have emphasized the need to incorporate pedestrian walking environment of a neighbourhood into the node-place model. Jeffrey et al. (2019) isolated walkability from the node and place dimensions and derived a typology for evaluating access to metropolitan Melbourne’s train stations based upon a cluster analysis methodology using 14 walkability measures. Their analyses distinguished stations located in high walkable neighbourhoods and those in areas with only some walkable features.

Mountainous cities differ significantly from flat cities in terms of walkability due to their unique topography. Gao et al. (2023) identified several factors that discourage pedestrian activity in mountainous cities. The steep inclines and slopes of the local terrain make walking difficult by shortening comfortable walking distances. The mountainous features, like changes in altitude and small-scale slopes, limit the number of pedestrian pathways. Limited available land for construction leads to higher building density, narrower walking paths and more intersections (Gao et al., 2023). Sun et al. (2015) highlighted that people’s perception of hilly environments can discourage walking. Sun et al. (2020) found that significant detours can make some locations nearly inaccessible, reducing the actual walkable area around transit stations. Xiong et al. (2024) uncovered nonlinear effects of road network density and land use mix on walking, contrasting common assumptions based on flat cities.

As demonstrated in the literature above, pedestrian accessibility at transit stations is crucial for informing empirical policy discussions that coordinate transportation and land use for sustainable cities. However, current indicators may not adequately reflect the accessibility needs in cities with challenging terrains. Identifying this gap, we focus on evaluating pedestrian experience in mountainous cities, thereby broadening the scope of node-place analysis beyond existing metrics.

3 Methodology

3.1 Study area

Chongqing is a large sprawling city in western China of approximately 80,000 km² and is home to over 30 million people (Chongqing Municipal Government, 2023). By 2050, its population is projected to exceed 40 million. This rapid growth will require an additional 1.6 million residential dwellings to be built in the next 25 years (Li, 2007). Urban sprawl into suburbs has led to high rates of car dependency. Recent state government planning emphasizes the importance of preparing for future population growth in established areas, including through transit development (Chongqing Housing and Urban-Rural Development Commission, 2022). In 2005, Chongqing became the 9th city in mainland China and the first in western China to launch its metro system. By the end of 2023, the city had 11 metro lines in operation, spanning a total of 494,6 km and 253 stations (China Ministry of Transport, 2024). These lines cover major functional nodes, such as airports, high-speed rail stations, comprehensive hubs, and business districts, with a maximum daily ridership of more than 5 million passengers. New lines under construction, along with branches and extensions, total approximately 186 km², shaping the city’s future development directions and connecting peripherals in the metropolitan area (Chongqing Housing and Urban-Rural Development Commission, 2022).

Walking accounts for over 40% of journeys made by commuters and is vital for last-mile mobility (Chongqing Municipal Government, 2019). Establishing a comprehensive regional urban transit system requires investment in pedestrian-friendly pathways and innovative solutions to ensure transit hubs are easily accessible despite the challenging terrain constraints. In December 2020, the Chongqing Municipal Government issued an implementation plan for Transit Oriented Development (TOD), announcing that 98 metro stations would undergo TOD integration (Chongqing Municipal Government, 2021). The goal is to enhance the surrounding infrastructure of these stations, strengthen the structure and functionality of pedestrian systems, and improve accessibility to rail transit. Considering the rapid development of its rail transit system, which often precedes and influences land development, Chongqing serves as an exemplary case study for exploring the relationship between transit and land use development. Moreover, it provides an opportunity to identify stations at various stages of development, making it suitable for implementing differentiated development strategies. The study area of this paper is shown in Figure 1.

Figure 1

3.2 A modified station assessment model for mountainous cities

We conducted an analysis encompassing 22 metro station areas in Chongqing, delineated based on the metro map (see Figure 1). Our model construction involved the utilization of two types of datasets: (Bertolini, 1999) spatial data, including sidewalk information sourced from the OpenStreetMap website, POI data and building footprint data acquired from Gaode Maps Place API, and (Bertolini, 2008) quantitative attribute data for each station area, including neighborhood-scale census data including population and employment. Refer to Table 2 for a list of data sources used in this study.

Building on the conventional node-place model, we incorporated pedestrian accessibility metrics to evaluate in detail the walking environment in mountainous cities. Our assessment model takes the shape of a rose diagram (Caset et al., 2019) with 3 dimensions and 11 indicators. The pedestrian experience dimension reflects the ease, enjoyment, and directness of reaching a station for individuals. This includes considerations like directness of pathways, density of POIs along the route, availability of shading. Each indicator is normalized to vary between 0 and 100. We illustrate the model structure in Figure 2 and detail the indicators in the subsequent sections.

Figure 2

3.2.1 Node dimension

Following existing scholarship, we assess node values based on criteria such as the proximity of other transit stations, the quantity and coverage of bus stops, and the connectivity of the station to urban road networks (Gonçalves and Portugal, 2008; Vale, 2015; Zhang et al., 2019; Dou et al., 2021; Su et al., 2022). We exclude accessibility by bicycle in this calculation because commuting by bicycle is less than 1% of the modal share in Chongqing, a trend commonly observed in other mountainous cities as well. Node indicators are listed in Table 3.

3.2.2 Place dimension

Previous studies typically apply catchment areas for each station to measure the place index (Vale, 2015; Jeffrey et al., 2019). The distance threshold for the place index is not universally standardized and commonly ranges from 5 to 10 min of walking, roughly equivalent to a 400–800 m walking distance (Frank et al., 2005, 2006; Zemp et al., 2011; Jun et al., 2015; Lyu et al., 2016; Li et al., 2019). Walkability extends beyond straight distances, particularly in mountainous cities. Factors such as steep walking slopes and irregular pedestrian networks significantly influence the distances most people are willing to walk. This study applies a circular area with a radius of 800 m as the transit impact area (TIA) and a network catchment area within 10 min of walking as the pedestrian sheds (Ped-Shed) area, drawing from Vale (2015). We aim to account for the ease of access when evaluating the vibrancy, richness, and liveability of urban places by comparing the actual accessible area to the maximum theoretical coverage. The Ped-Shed is determined using the GaoDe Map Distance API for an accurate representation of real-world accessibility in the given location. Figure 3 illustrates the TIA with the actual accessible Ped-Shed area at Shapingba station.

Figure 3

We use four indicators to describe the place dimension: population density, employment density, development intensity and land use mix. Having access to high-density land uses and the opportunity to engage in a variety of activities provides favorable conditions for both land use and transport development (Dittmar and Poticha, 2004; Cao et al., 2006; De Vos et al., 2013). Place indicators are listed in Table 4.

3.2.3 Pedestrian experience dimension

Evaluating pedestrian accessibility in mountainous areas differs from that in flat cities due to terrain challenges. Areas theoretically accessible can become inaccessible due to significant detours, long stairs, or elevation differences that disrupt direct connections. To empirically represent pedestrian experiences in accessing transit stations, we introduced indicators including parcel connectivity, pedestrian directness, street activity, and street greenery (Table 5). Parcel connectivity measures the ratio of directly connected parcels within walking distance. Pedestrian directness measures the ratio of actual walking distance to Euclidean distance, with lower values indicating fewer detours (Stangl, 2019). The street activity index comprehensively measures the number and density of POIs along streets (Yue et al., 2017). Street greenery refers to the percentage of visible greenery from specific vantage points, collected from Baidu Streetview images, reflecting the amount of vegetation in the urban environment (Long and Liu, 2017). These metrics help us understand how pedestrians navigate around stations, reflecting their experiences in terms of comfort, effort, and aesthetics.

3.3 Analyze patterns and develop a typology of stations

After we obtain the 11 indicator values for all stations, we create a rose diagram for each station. The diagram visualizes the 11 indicators in a circular layout, with each indicator ranging from 0 to 100. We then develop our typology of stations based on their node, place and pedestrian experience profiles. Principal Components Analysis (PCA), a data reduction technique and multivariate statistical method, is used to extract synthetic variables or factors from the original set. This approach assumes that the original data can be represented as a linear combination of “artificial” variables corresponding to latent factors. We performed a Hierarchical Clustering on Principal Components (HCPC) to identify groups of similar observations in the dataset and enhance cluster interpretation by leveraging the reduced, informative components from PCA. This study conducted factor and cluster analysis using R, specifically with the factoextra and FactoMineR packages.

4 Results

4.1 Station assessment result

Our factor analysis (orthogonal, varimax rotation), results in 2 interpretable factors, with an eigenvalue larger than 1 and explaining 60% of total variance. Factor 1 has strong loadings for place indicators including population and employment density (P1, P2) and development intensity (P3). Factor 2 has strong loadings for indicators on connections including motorway access (N3) and pedestrian directness (W2). Based on these factors, an HCPC is conducted, resulting in 5 interpretable groups of stations, as shown in Figure 4.

Figure 4

Stations in Cluster 5 have high development density and well-connected road networks compared to overall means across all clusters. Place variables (P1, P2, P3) and motorway access (N3) are most significantly associated with Cluster 5. Cluster 4 is significantly associated with street greenness (W4), land use mix (P4), and street activity (W3), indicating a pedestrian-friendly built environment. Cluster 3 is strongly influenced by node values but are less influenced by land use diversity (P4), with stations in this category strong in transportation functions. Stations in Cluster 2 features low place variables and a lack of parcel connectivity (W1). These stations are relatively new, suggesting opportunities to enhance land use development in their surrounding areas. Cluster 1 only includes Chongqing West Station, a unique example with low N1 (transit coverage) and a pedestrian-unfriendly environment. This station, operational since 2018, primarily serves regional rails and has limited transportation and pedestrian connections. We summarize the values of 11 indicators in our NPW (node, place, pedestrian) model in Figure 5.

Figure 5

The Chongqing Comprehensive Transportation Plan (2021–2035) categorizes urban transit stations into four functional types—Urban Center Stations, General Urban Stations, Transport Hub Stations, and Special Control Stations—based on their locations and functions within the transportation system. In this study, we blend these typologies to determine the stations that fall into different intersections and to gain insights into station-specific development characteristics. Table 6 presents a cross-tabulation of the station classifications.

These crosstabulations provide detailed insights into station-specific accessibility characteristics, which are not fully captured by standard node-place analyses. There is a noticeable overlap among the strong stations: Type 5 stations almost exclusively fall under urban stations, while Types 4 and 3 are mostly general urban stations with moderate to strong network functions. Special control stations, though vaguely defined in the municipal plan, typically refer to stations that are not commercial hubs but are crucial for accessing new and emerging city landmarks like the new Expo Center or the Garden Expo Park. Interestingly, both major transit hub stations, Chongqing North and Chongqing West, exhibit low development and pedestrian experience, indicating that while they serve significant transit functions, they are not conducive to pedestrian activities.

We visualize our blended station topologies based on NPW model in Figure 6. We have five categories: urban stations with high development and average pedestrian experience, urban stations with average development and high pedestrian experience, suburban stations with average development and average pedestrian experience, suburban stations with average centrality, low development and low pedestrian experience, and transportation hubs with low development and low pedestrian environment. Each category diversifies the land use and pedestrian development opportunities for stations.

Figure 6

4.2 Practical implication on four stations

In order to clarify what the station-specific development opportunities may mean for planning practice, this section discusses the characteristics of four stations, each belonging to one category from our NPW typology. We describe how the additional pedestrian experience dimension sheds light on targeted and prioritized planning recommendations. We also relate our findings to the objectives of transportation and land use planning policies in mountainous cities.

4.2.1 Shapingba Station

Located in one of the central business districts (Sanxia Plaza) in Chongqing, Shapingba Station belongs to Cluster 5 in NPW typology. Loop Line, Line 1, and Line 9 intersect at this station. The area surrounding the station has been well-developed, with a relatively high level of land development and an integrated transportation system, including comprehensive bus connections. Additionally, the station has been further enhanced with the recent completion and operation of the Shapingba high-speed rail hub and its superstructure. The passenger traffic in and out of this station is expected to grow. Figure 7 depicts the population density, employment distribution, street POI distribution, and sidewalk distribution within the station’s impact area.

Figure 7

Table 7 presents the indicator values of Shapingba Station. The population and job density of Shapingba Station are the highest among all stations in this study. The indices for street activity and pedestrian directness are also on the higher end.

Shapingba Station has high development density and well-connected road networks, surpassing the average level across all clusters. Land use in the station’s vicinity has been maximized, and additional development could potentially lead to conflicts due to limited available space. However, implementing urban design strategies to upgrade the walking environment of existing spaces can mitigate these challenges and foster sustainable growth. This may involve creating pedestrian-friendly paths, improving park-and-ride connections, enhancing streetscapes, and integrating green infrastructure.

4.2.2 Guangdianyuan Station

Located in the rapidly developing Yubei District in the northern suburb of Chongqing, Guangdianyuan Station is a Line 6 station that was completed and put into operation in December 2014. Commercial offices and research use dominate the surrounding land use. Figure 8 depicts the population density, employment distribution, street POI distribution, and sidewalk distribution within the station’s impact area. Table 8 presents the indicator values.

Figure 8

Guangdianyuan Station falls within cluster 3 and primarily serves the Guangdian Industrial Park, which situates away from the main urban area of Chongqing. Consequently, the station is serviced by only one transit line, with limited bus routes available, resulting in lower feeder transport convenience and node values. While the area boasts a large catchment area, its functional diversity is limited, mainly comprising commercial offices and research facilities, with moderate population and employment density. Pedestrian paths are generally favorable, with no significant slopes or detours. Thus, fully leveraging transportation and land use opportunities will be critical for future station development, including enhancing land use diversity and intensity, and ensuring pedestrian connections.

4.2.3 Central Park station

Central Park station was initially conceived to serve long-distance commuters from two of Chongqing’s planned new towns—Airport New Town and Yuelai New Town in the northern part of Chongqing. Currently, Central Park station only serves Line 10. The area around the station is still under development, with minimal land development and limited pedestrian services. Additionally, the station does not provide transfers to any bus lines. The station is classified into Cluster 2 in our model. Analysis of population density, job distribution, POI and sidewalk distribution within the station’s influence area can be seen in Figure 9.

Figure 9

The construction of Central Park Station was primarily driven by TOD and occurred before the comprehensive development of the surrounding parcels. These parcels are yet to be fully developed, showing sparse population and job density, inadequate street activity, and the lowest feeder transport and development intensity among all 22 stations. However, this offers an excellent opportunity to shape a balanced, well-developed area and create the most suitable pedestrian system for the location from the outset. Given that Central Park Station is still in its early development stage, it is crucial to align subsequent infrastructure development with its intended purpose and support the ongoing parcel planning and development processes (Table 9).

4.2.4 Chongqing North Station

Chongqing North Station is adjacent to the Chongqing North Railway Station, which is the largest passenger transportation center in Southwest China and Chongqing’s primary high-speed rail hub. Consequently, this metro station serves as a transfer hub for multiple metro lines: Loop Line, Line 3, Line 4, and Line 10, making it a pivotal transportation node. The southern square area of the Chongqing North Railway Station is part of an existing developed area, with the surrounding land primarily used for transportation facilities such as bus terminals and long-distance bus stations, complemented by commercial and residential areas. The northern square area has recently expanded and is currently under development. Analysis of population density, job distribution, POI and sidewalk distribution within the station’s influence area can be seen in Figure 10.

Figure 10

Chongqing North Station exemplifies a location where prioritizing node functions is crucial. Its role as a transportation hub restricts extensive land development due to limited space and suitability. To enhance connectivity, we can improve the pedestrian system by adding greenery, establishing convenient connections with local buses and guidance systems, and enhancing street-level commercial activities (Table 10).

5 Discussion

This research aims to introduce an empirical framework tailored for assessing urban transit stations in mountainous cities. Mountainous cities present distinct challenges when it comes to transportation and land use development, particularly due to their intricate road network layouts and pedestrian detours resulting from terrain elevation. Insufficient pedestrian infrastructure can hinder the connection of certain land parcels, making them impractical or unattractive for development. These challenges can be reflected using the node-place model framework combined with pedestrian-specific indicators. Our new model allows us to better categorize different urban transit station areas according to their respective realized pedestrian benefits.

The significance of our typology lies in its ability to reflect the diverse development statuses and needs of stations within a city. Unlike the traditional urban and suburban classifications, our typology reveals that stations located within the same urban/suburban areas exhibit varied characteristics and requirements. For instance, while Gongmao, Ranjiaba, and Guangdianyuan are all classified as general urban stations, their strengths and priorities differ as shown by the rose diagram. Ranjiaba excels in pedestrian accessibility, Gongmao has fully developed parcels, and Guangdianyuan requires improvement in land use mix.

Furthermore, the additional dimension of pedestrian experience evaluation proves valuable in a mountainous city. Stations near the commercial core of the old city, such as Shapingba, Daping, Xiaoshizi, and Guanyinqiao, are characterized by high node and place values. However, Daping is more pedestrian-friendly, whereas Shapingba, despite being busy, has limited space and could benefit from pedestrian system improvements. Without considering this dimension, opportunities for enhancing the pedestrian experience at these stations might be overlooked.

We would also like to emphasize the importance of bus connections in station areas. Our indicator system includes feeder transport (N2), which evaluates the number and frequency of connected buses. The mean value of N2 across the 22 stations is 31 (out of 100), with 60% of the stations scoring below this mean. Many suburban stations lack immediate ground transport connections, placing them at the lower end of the station assessment. Central Park, Expo, and Chongqing West stations are notable examples of this issue. Enhancing bus services, particularly feeder buses to transit stations, can promote sustainable urban development and should be a key focus for Chongqing’s future growth and infrastructure planning.

This study introduces an empirical station assessment tool that can be adapted for use in other mountainous regions outside of China. For areas facing terrain challenges that influence pedestrian experience, this study provides a set of indicators for evaluation. However, when applying this assessment model to other locations, it is essential to consider the unique urban characteristics of these cities and make necessary adjustments to improve the model’s accuracy and relevance.

There are also several limitations that should be acknowledged when interpreting the study findings. First, the transportation and land use indicators used in this study are derived from previous research conducted in flat cities. The calculation methods, such as for land use mix, may differ in mountainous cities and require refinement. Second, more indicators such as perceptions of safety, the width and quality of sidewalks could be incorporated into the model. The indicators used in our empirical model are based on available data and the situations of Chongqing and thus may not include all. Therefore, this model is a contribution to the broader node-place literature rather than a comprehensive solution. Studies in evaluating transit stations provide a wide range of perspectives for land use planning practices. This study specifically contributes to the perspective of walking in mountainous cities.

6 Conclusion

In this paper, we leverage existing node-place modelling literature to develop a multi-dimensional station assessment methodology specifically suited for mountainous cities. We introduced new indicators including parcel connectivity, pedestrian directness, street activity, and street greenery, to capture the distinct challenges of mountainous urban environments, thus augmenting the existing model with the dimension of pedestrian experience. We structured our indicators in a rose diagram. This approach allows for differentiated station profiles that provide detailed insights into station-specific accessibility dynamics. By refining local classifications from urban-suburban to five more detailed categories describing the level of land use development and pedestrian friendliness, we gain a deeper understanding of the development opportunities of each station that standard node-place analyses often overlook. Finally, we present four case studies from Chongqing to demonstrate how our analyses inform planning practically and propose targeted development plans for these stations.

For the stations studied, core development strategies identified in this research include strategically improving street view and vitality, enhancing direct connections between transit station entrances and adjacent parcels, establishing a coherent, safe, convenient, and comfortable pedestrian system, and strengthening feeder bus systems. These findings are valuable for other mountainous cities and indicate an important direction for future transportation and land use research.

Building on prior studies that primarily focused on plain cities, this study makes three contributions (Bertolini, 1999). It explores the possibility of integrating pedestrian experience into the traditional node-place evaluation framework, thus expanding the scope of analysis beyond conventional metrics (Bertolini, 2008). It focuses on the unique challenges of walkability in mountainous cities and developed indicators to measure them, thereby broadening the application of node-place models (Bertolini and Spit, 1998). It compares stations situated in varied locations and at different development stages, shedding light on the subtle influence of development opportunities and priorities outcomes.

A next step in this research will consist of a qualitative validation of the usefulness of this model. We could conduct surveys, interviews, or field observations to evaluate the development opportunities surrounding station areas and compare stations across different categories. Given the increasing importance of integrating transport and land use in Chongqing, a strategic approach would be to foster collaboration among public transport operators, the Chongqing Government, and other stakeholders to advocate for healthy, liveable and sustainable city development.

References

• 1

  Baidu Streetview. (2023). Baidu Streetview [Chongqing]. Retrieved from. https://map.baidu.com/@11862160,3426989.5,14.85z

  • Google Scholar

• 2

  BertoliniL. (1999). Spatial development patterns and public transport: the application of an analytical model in the Netherlands. Plan. Pract. Res.14, 199–210. doi: 10.1080/02697459915724

  • CrossRef
  • Google Scholar

• 3

  BertoliniL. (2008). Station areas as nodes and places in urban networks: an analytical tool and alternative development strategies. In: BruinsmaF.PelsE.PriemusH.RietveldP.WeeB.Van (Eds.), Railway development: Impacts on urban dynamics. Physica-Verlag Heidelberg, Leipzig, pp. 35–57.

  • Google Scholar

• 4

  BertoliniL.SpitT. (1998). Cities on rails: The redevelopment of railway stations and their surroundings. London: Routledge.

  • Google Scholar

• 5

  Bureau of Statistics of Chongqing. (2022). Chongqing Statistical Yearbook 2022. Retrieved from. https://tjj.cq.gov.cn/zwgk_233/tjnj/2022/zk/indexeh.htm

  • Google Scholar

• 6

  CaoZ.AsakuraY.TanZ. (2020). Coordination between node, place, and ridership: comparing three transit operators in Tokyo. Transp. Res. Part D: Transp. Environ.87:102518:102518. doi: 10.1016/j.trd.2020.102518

  • CrossRef
  • Google Scholar

• 7

  CaoX.HandyS. L.MokhtarianP. L. (2006). The influences of the built environment and residential self-selection on pedestrian behavior: evidence from Austin, TX. Transportation33, 1–20. doi: 10.1007/s11116-005-7027-2

  • CrossRef
  • Google Scholar

• 8

  CasetF.TeixeiraF. M.DerudderB.BoussauwK.WitloxF. (2019). Planning for nodes, places and people in Flanders and Brussels: developing an empirical railway station assessment model for strategic decision-making. J. Transp. Land Use12, 811–837. doi: 10.5198/jtlu.2019.1483

  • CrossRef
  • Google Scholar

• 9

  Chongqing Transport Bureau. (2023). Chongqing transite network map. Retrieved February 23, 2023 from https://jtj.cq.gov.cn/zwgk_240/zfxxgkml/gggs/tzgg/202302/t20230223_11643527_wap.html

  • Google Scholar

• 10

  China Ministry of Transport. (2024). 2023 Urban Rail Transit Operation Data. Available at: https://mp.weixin.qq.com/s/A2SMRyIYkcNJ25J8CeQsTQ (Accessed June 10, 2024).

  • Google Scholar

• 11

  Chongqing Housing and Urban-Rural Development Commission. (2022). Chongqing Urban Rail Transit Construction "14th Five-Year" Plan. Available at: http://zfcxjw.cq.gov.cn/zwxx_166/gsgg/202201/P020220126659119065139.pdf (Accessed June 10, 2024).

  • Google Scholar

• 12

  Chongqing Municipal Government. (2019). Chongqing Mountain City Trail. Available at: https://www.cq.gov.cn/zwgk/zfxxgkml/zdlyxxgk/shgysy/ggwhty/whly/201904/t20190409_8807251.html (Accessed June 10, 2024).

  • Google Scholar

• 13

  Chongqing Municipal Government. (2020). Creation of a "one-day living circle" at railway stations, 98 railway stations in Chongqing will be developed by TOD. Available at: https://www.cq.gov.cn/ywdt/jrcq/202012/t20201212_8646928.html (Accessed December 10, 2023).

  • Google Scholar

• 14

  Chongqing Municipal Government. (2021). Chongqing Comprehensive Transportation Plan (2021–2035). Available at: https://www.cq.gov.cn/zwgk/zfxxgkml/szfwj/qtgw/202110/t20211013_9800569.html (Accessed June 10, 2024).

  • Google Scholar

• 15

  Chongqing Municipal Government. (2023). Chongqing Resident Population Data. Available at: https://www.cq.gov.cn/ywdt/jrcq/202303/t20230308_11723441.html (Accessed June 10, 2023).

  • Google Scholar

• 16

  ChorusP.BertoliniL. (2011). An application of the node-place model to explore the spatial development dynamics of station areas in Tokyo. J. Transp. Land Use4, 45–58. doi: 10.5198/jtlu.v4i1.145

  • CrossRef
  • Google Scholar

• 17

  CongC.KwakY.DealB. (2022). Incorporating active transportation modes in large scale urban modeling to inform sustainable urban development. Comput. Environ. Urban. Syst.91:101726. doi: 10.1016/j.compenvurbsys.2021.101726

  • CrossRef
  • Google Scholar

• 18

  CummingsC.MahmassaniH. (2022). Does intercity rail station placement matter? Expansion of the node-place model to identify station location impacts on Amtrak ridership. J. Transp. Geogr.99:103278. doi: 10.1016/j.jtrangeo.2022.103278

  • CrossRef
  • Google Scholar

• 19

  De VosJ.SchwanenT.AckerV.WitloxF. (2013). Travel and subjective well-being: a focus on findings, methods and future research needs. Transp. Rev.33, 421–442. doi: 10.1080/01441647.2013.815665

  • CrossRef
  • Google Scholar

• 20

  DittmarH.PotichaS. (2004). Defining transit-oriented development: the new regional building block. In: The new transit town: best practices in transit-oriented development. Geography, Economics, Engineering, Environmental Science, 20–40.

  • Google Scholar

• 21

  DouM.WangY.DongS. (2021). Integrating network centrality and node-place model to evaluate and classify station areas in Shanghai. ISPRS Int. J. Geo Inf.10:414. doi: 10.3390/ijgi10060414

  • CrossRef
  • Google Scholar

• 22

  El-GeneidyA.GrimsrudM.WasfiR.TétreaultP.Surprenant-LegaultJ. (2014). New evidence on walking distances to transit stops: identifying redundancies and gaps using variable service areas. Transportation41, 193–210. doi: 10.1007/s11116-013-9508-z

  • CrossRef
  • Google Scholar

• 23

  EwingR.CerveroR. (2010). Travel and the built environment. J. Am. Plan. Assoc.76, 265–294. doi: 10.1080/01944361003766766

  • CrossRef
  • Google Scholar

• 24

  EwingR.HandyS. (2009). Measuring the unmeasurable: urban design qualities related to walkability. J. Urb. Des.14, 65–84. doi: 10.1080/13574800802451155

  • CrossRef
  • Google Scholar

• 25

  FrankL. D.SallisJ. F.ConwayT. L.ChapmanJ. E.SaelensB. E.BachmanW. (2006). Many pathways from land use to health: associations between neighborhood walkability and active transportation, body mass index, and air quality. J. Am. Plan. Assoc.72, 75–87. doi: 10.1080/01944360608976725

  • CrossRef
  • Google Scholar

• 26

  FrankL. D.SchmidT. L.SallisJ. F.ChapmanJ.SaelensB. E. (2005). Linking objectively measured physical activity with objectively measured urban form: findings from SMARTRAQ. Am. J. Prev. Med.28, 117–125. doi: 10.1016/j.amepre.2004.11.001

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 27

  Gaode Map. (2023). Gaode Map [Chongqing]. Retrieved from https://ditu.amap.com/search?query=chongqing&city

  • Google Scholar

• 28

  Gaode Map Place API. (2023). Retrieved from https://lbs.amap.com/. Gaode Map. (2023). Gaode Map [Chongqing]. Retrieved from https://ditu.amap.com/search?query=chongqing&city

  • Google Scholar

• 29

  GaoL.ChongH.ZhangW.LiZ. (2023). Nonlinear effects of public transport accessibility on urban development: a case study of mountainous city. Cities138:104340. doi: 10.1016/j.cities.2023.104340

  • CrossRef
  • Google Scholar

• 30

  Giles-CortiB.Vernez-MoudonA.ReisR.TurrellG.DannenbergA. L.BadlandH.et al. (2016). City planning and population health: a global challenge. Lancet388, 2912–2924. doi: 10.1016/S0140-6736(16)30066-6

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 31

  GonçalvesJ.A.M.PortugalL.D.S. (2008). Classificando estações metro-ferroviárias como pólo promotor do desenvolvimento socioeconômico, 4 Concurso de Monografia CBTU 2008 – A Cidade nos Trilhos. Companhia Brasileira de Trens Urbanos, Rio de Janeiro.

  • Google Scholar

• 32

  GoriS.NigroM.PetrelliM. (2014). Walkability indicators for pedestrian-friendly design. Transp. Res. Rec.2464, 38–45. doi: 10.3141/2464-05

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 33

  GroenendijkL.RezaeiJ.de AlmeidaH.CorreiaG. (2018). Incorporating the travelers’ experience value in assessing the quality of transit nodes: a Rotterdam case study. Case Stud. Transp. Policy6, 564–576. doi: 10.1016/j.cstp.2018.07.007

  • CrossRef
  • Google Scholar

• 34

  HigginsC. D.KanaroglouP. S. (2016). A latent class method for classifying and evaluating the performance of station area transit-oriented development in the Toronto region. J. Transp. Geogr.52, 61–72. doi: 10.1016/j.jtrangeo.2016.02.012

  • CrossRef
  • Google Scholar

• 35

  IvanI.BorutaT.HorákJ. (2012). “Evaluation of railway surrounding areas: the case of Ostrava city” in Urban transport XVIII – Urban transport and the environment in the 21st century. eds. LonghurstJ. W. S.BrebbiaC. A. (Valencia, Spain: WIT Press), 141–152.

  • Google Scholar

• 36

  JeffreyD.BoulangéC.Giles-CortiB.WashingtonS.GunnL. (2019). Using walkability measures to identify train stations with the potential to become transit oriented developments located in walkable neighbourhoods. J. Transp. Geogr.76, 221–231. doi: 10.1016/j.jtrangeo.2019.03.009

  • CrossRef
  • Google Scholar

• 37

  JunM.-J.ChoiK.JeongJ.-E.KwonK.-H.KimH.-J. (2015). Land use characteristics of subway catchment areas and their influence on subway ridership in Seoul. J. Transp. Geogr.48, 30–40. doi: 10.1016/j.jtrangeo.2015.08.002

  • CrossRef
  • Google Scholar

• 38

  LiY. (2007). Research on Population Growth Situation and Population Policy of Chongqing[J]. Journal of Chongqing Institute of Technology (Social Science Edition), 27, 52–55.

  • Google Scholar

• 39

  LiZ.HanZ.XinJ.LuoX.SuS.WengM. (2019). Transit-oriented development among metro station areas in Shanghai, China: variations, typology, optimization and implications for land use planning. Land Use Policy82, 269–282. doi: 10.1016/j.landusepol.2018.12.003

  • CrossRef
  • Google Scholar

• 40

  LiaoC.ScheuerB. (2022). Evaluating the performance of transit-oriented development in Beijing metro station areas: integrating morphology and demand into the node place model. J. Transp. Geogr.100:1–11. doi: 10.1016/j.jtrangeo.2022.103333

  • CrossRef
  • Google Scholar

• 41

  LongY.LiuL. (2017). How green are the streets? An analysis for central areas of Chinese cities using Tencent street view. PLoS One12:e0171110. doi: 10.1371/journal.pone.0171110

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 42

  LyuG.BertoliniL.PfefferK. (2016). Developing a TOD typology for Beijing metro station areas. J. Transp. Geogr.55, 40–50. doi: 10.1016/j.jtrangeo.2016.07.002

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 43

  MacioszekE.GranaA.KrawiecS. (2023). Identification of factors increasing the risk of pedestrian death in road accidents involving a pedestrian with a motor vehicle. Arch. Transp.65, 7–25. doi: 10.5604/01.3001.0016.2474

  • CrossRef
  • Google Scholar

• 44

  MacioszekE.WyderkaA. (2021). Road traffic distribution on public holidays and workdays on selected road transport network elements. Transp. Prob.16, 127–138. doi: 10.21307/tp-2021-011

  • CrossRef
  • Google Scholar

• 45

  MonajemS.NosratianF. E. (2015). The evaluation of the spatial integration of station areas via the node place model; an application to subway station areas in Tehran. Transp. Res. D40, 14–27. doi: 10.1016/j.trd.2015.07.009

  • CrossRef
  • Google Scholar

• 46

  OpenStreetMap. (2023). OpenStreetMap [Chongqing]. Retrieved from https://www.openstreetmap.org/search?query=chongqing#map=19/22.29639/114.17284

  • Google Scholar

• 47

  PeekG. J. (2006). Locatiesynergie: Een participatieve start van de herontwikkeling van binnenstedelijke stationslocaties. Delft, NL: Eburon Academic Press.

  • Google Scholar

• 48

  ReusserD. E.LoukopoulosP.StauffacherM.ScholzR. W. (2008). Classifying railway stations for sustainable transitions – balancing node and place functions. J. Transp. Geogr.16, 191–202. doi: 10.1016/j.jtrangeo.2007.05.004

  • CrossRef
  • Google Scholar

• 49

  SchlossbergM.BrownN. (2004). Comparing transit-oriented development sites by walkability indicators. Transp. Res. Rec.1887, 34–42. doi: 10.3141/1887-05

  • CrossRef
  • Google Scholar

• 50

  SinghY.LukmanA.FlackeJ.ZuidgeestM.van MaarseveenM. F. A. M. (2017). Measuring TOD around transit nodes - towards TOD policy. Transp. Policy56, 96–111. doi: 10.1016/j.tranpol.2017.03.013

  • CrossRef
  • Google Scholar

• 51

  Stadsregio Arnhem Nijmegen (2011). Knooppunten! Bereikbaarheid en ruimtelijke ontwikkeling op knooppunten van openbaar vervoer. Arnhem: Stadsregio Arnhem-Nijmegen.

  • Google Scholar

• 52

  StanglP. (2019). Overcoming flaws in permeability measures: modified route directness. J. Urban.: Int. Res. Placemak. Urban Sustain.12, 1–14. doi: 10.1080/17549175.2017.1381143

  • CrossRef
  • Google Scholar

• 53

  SuS.WangZ.LiB.KangM. (2022). Deciphering the influence of TOD on metro ridership: an integrated approach of extended node-place model and interpretable machine learning with planning implications. J. Transp. Geogr.104:1–128. doi: 10.1016/j.jtrangeo.2022.103455

  • CrossRef
  • Google Scholar

• 54

  SunG.HainingR.LinH.OreskovicN. M.HeJ. (2015). Comparing the perception with the reality of walking in a hilly environment: an accessibility method applied to a university campus in Hong Kong. Geospat. Health10:340. doi: 10.4081/gh.2015.340

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 55

  SunG.WallaceD.WebsterC. (2020). Unravelling the impact of street network structure and gated community layout in development-oriented transit design. Land Use Policy90:104328. doi: 10.1016/j.landusepol.2019.104328

  • CrossRef
  • Google Scholar

• 56

  ValeD. S. (2015). Transit-oriented development, integration of land use and transport, and pedestrian accessibility: combining node-place model with pedestrian shed ratio to evaluate and classify station areas in Lisbon. J. Transp. Geogr.45, 70–80. doi: 10.1016/j.jtrangeo.2015.04.009

  • CrossRef
  • Google Scholar

• 57

  ValeD. S.VianaC. M.PereiraM. (2018). The extended node-place model at the local scale: evaluating the integration of land use and transport for Lisbon’s subway network. J. Transp. Geogr.69, 282–293. doi: 10.1016/j.jtrangeo.2018.05.004

  • CrossRef
  • Google Scholar

• 58

  XiongR.ZhaoH.HuangY. (2024). Spatial heterogeneity in the effects of built environments on walking distance for the elderly living in a mountainous cityOxfordElsevier Ltd.

  • Google Scholar

• 59

  YueY.ZhuangY.YehA.XieJ.MaC.LiQ. (2017). Measurements of POI-based mixed use and their relationships with neighbourhood vibrancy. Int. J. Geogr. Inf. Sci.31, 658–675. doi: 10.1080/13658816.2016.1220561

  • CrossRef
  • Google Scholar

• 60

  ZempS.StauffacherM.LangD. J.ScholzR. W. (2011). Classifying railway stations for strategic transport and land use planning: context matters!J. Transp. Geogr.19, 670–679. doi: 10.1016/j.jtrangeo.2010.08.008

  • CrossRef
  • Google Scholar

• 61

  ZhangY.MarshallS.ManleyE. (2019). Network criticality and the node-place-design model: classifying metro station areas in greater London. J. Transp. Geogr.79:102485. doi: 10.1016/j.jtrangeo.2019.102485

  • CrossRef
  • Google Scholar

• 62

  ZhouM.ZhouJ.ZhouJ.LeiS.ZhaoZ. (2023). Introducing social contacts into the node-place model: a case study of Hong Kong. J. Transp. Geogr.107:103532. doi: 10.1016/j.jtrangeo.2023.103532

  • CrossRef
  • Google Scholar

• 63

  ZweedijkA.SerlieZ. (1998). Een “knoop-plaats”-model voor stationslocaties. Aust. Geogr.7, 35–37.

  • Google Scholar