Assessing airport ground access by public transport in Chinese cities

Abstract

Chinese

This article assesses airport ground access by public transport in China. Recent literature has highlighted the economic, environmental and social significance of airport ground access. Existing studies on airport ground access have predominately centred on North America and Europe and, to date, limited attempts have been made to assess the emerging Chinese market. Studies of urban and transport geography have detailed the shifting air connectivity of Chinese cities and the economic impacts, but have paid little attention to ground access to airports. We, therefore, assess the ground accessibility to major Chinese airports based on online map services. Specifically, we characterise airport ground access across entire cities, as well as comparing time and monetary costs for travelling between airports and city centres by private car and public transport. We conclude with suggestions for future research, and call for more systematic data collection related to airport ground access.

Keywords

airport–city link airport rail link China ground access public transport

Introduction

The success of airports and air-based economies depends on the connectivity between airports within air transport networks (i.e. the airside transport at airports), as well as ground access from cities to airports (i.e. the landside transport; Coogan et al., 2008; Matisziw and Grubesic, 2010; Yang et al., 2016). While air transport between airports is expanding at an unprecedented speed, the journey between cities and their airports might be changing at a slower rate. For example, Murakami et al. (2016: 89) observe that ‘ground transportation between airport terminals and cities are [sic] sometimes more burdensome for travelers … than line-haul air transportation between airports’. Furthermore, recent academic and policy debates have highlighted the importance of landside airport access by public transport, often citing its positive economic and environmental effects.

Managing airport ground transportation poses an imminent challenge to Chinese cities and airports. Against the backdrop of increasing global air travel and its rising economy, China has already become the worldwide second largest aviation market (Gibbons and Wu, 2017) and may take the top spot by 2020 according to recent forecasts (Civil Aviation Administration of China (CAAC), 2016a). Specifically, approximately 60 new airports were built in China in the past decade. Domestic air passenger numbers grew by double digits and reached 394 million in 2015. (CAAC, 2016b). Furthermore, a growing proportion of air travel is for non-business purposes (Qin, 2009). For example, in surveys at several airports, 30–50% of the respondents were travelling for tourism (Huang et al., 2011; Xu, 2012). With the continuous growth of air transport in China, the issue of ground access will likely become more pertinent. As China is investing large sums in the country’s infrastructure, and China has undergone unprecedented urbanisation, the potential exists to (re)shape the link between airports and cities. For example, 50 new civil airports are planned for the thirteenth Five-Year Plan (2016-2020) (State Council of China, 2017). Nevertheless, developing and sustaining public transport access to airports often entail many planning and management challenges (Budd et al., 2016).

Despite its significance, limited research has critically examined landside accessibility to Chinese airports. Literature on airport management has predominately centred on airports in North America and Europe (Budd et al., 2014; Coogan, 2000; Ison et al., 2014) and, to date, few attempts have been made to evaluate airport ground access in the emerging Chinese market (e.g., Qin (2009) and Hirsch (2016)). Studies of urban and transport geography have detailed the shifting air connectivity of Chinese cities and its economic impacts (Gibbons and Wu, 2017), but have paid little attention to ground access to airports. Existing attempts at characterising airport ground access mostly employ accessibility measures that do not account for transit schedules and traffic conditions (Dai et al., 2013; Yang et al., 2016). A solid understanding of airport ground access is significant not only for those working in the transport sector but also for planners and other urban professionals (Freestone, 2009). For example, Ryerson (2016) has highlighted how air transportation posits planning challenges, and called for the integration of air transportation into planning education.

This article aims to compare landside accessibility to major Chinese airports by public transport as well as by private cars. The next section examines in more detail the significance of airport ground access by public transport, reviews how accessibility to airports has been measured and provides necessary background information regarding the planning and management of airport landside transportation in China. This information is followed by a summary of data and methodologies. We then characterise how accessibility to airports for transit and auto users varies within and across Chinese cities. Finally, the article concludes with major findings and suggestions for further studies.

Literature review

The significance of airport ground access by public transport

Because ground or landside access by public transport to airports is often associated with reduced energy use and carbon emissions, it has been placed high on policy agendas in recent drives towards sustainable air transport (Ison et al., 2014). Specifically, airports often lead to large volumes of local trips being made, which generate energy and environmental impacts (Budd et al., 2011). For instance, Coogan et al. (2008) link airport size with ground access demands, estimating that airports with 45 million and 5 million passengers per annum may generate up lead to as many as 5 million and 500,000 vehicle miles, respectively, of ground access on a daily basis. Most of these trips are by car, and a shift towards public transport is believed to reduce the environmental footprint of the aviation industry (Budd et al., 2016; Ison et al., 2014). Growing traffic between airports and urban areas has become a major urban mobility and environmental issue in North American and European cities, and is increasingly evident in China (Qin, 2009; Zheng, 2010).

In addition to its environmental benefits, landside accessibility to airports can have significant urban-economic impacts and can shift local land use–transport dynamics (Appold and Kasarda, 2013; Loo and Chow, 2011). Kasarda and Lindsay (2011) have proposed a new urban-economic developmental model that capitalises on easy ground access to airports. One particular form of public transport, airport rail link, has received much attention. Niedzielski and Malecki (2012) observed that leading producer service centres are often well-equipped with airport rail connections, citing that an airport rail link can help capture time-sensitive business travellers and boost the image of the city. By contrast, airport rail link projects are often criticised for their high construction and maintenance costs as well as their relatively small ridership (de Neufvilles, 2006; Guerra and Cervero, 2011; see also Hirsch (2016) for a detailed discussion on the business-oriented design of Airport Express and the increasing number of non-business travelers in Hong Kong). Furthermore, de Neufvilles (2006: 353) suggests that the expansive urban form in the US is the culprit, as ‘few final destinations are at central points served by fixed rail even in dense cities like New York or London’. There is also a social equity dimension because public transit in general is believed to provide services to a larger proportion of the population than private cars (Saghapour et al., 2016). Still, airport ground access entails the complicated interactions between airport-related infrastructure and emerging mobility and sociospatial issues (Hirsch, 2016).

Despite the productive line of research on airport ground access, significant gaps exist in the literature. As mentioned, while existing literature largely draws upon experiences in North American and European contexts (Budd et al., 2016), less is known about access to and public transport at Chinese airports, which entails different relations among planning, management and markets (Wang et al., 2016). Furthermore, many studies are case-based and focus on individual airports. These studies may need to be complemented by more systematic comparisons across airports (see, for example, Coogan, 2000). A first step towards addressing these issues is therefore systematically measuring landside accessibility by public transport at Chinese airports.

Measuring airport ground access

Literature is growing on air-transport geography in China, with studies examining the temporal evolution of air-transport networks (e.g. Wang et al., 2016), the impacts of airports on local economic development (e.g. Gibbons and Wu, 2017) and the relationship between cities’ connectivity within the air-transport network and their positions within urban hierarchies (Ma and Timberlake, 2013). Many of these studies concentrate at the intercity level without exploring the intracity patterns of transport and land use (Liu et al., 2016). Nevertheless, while few studies have explicitly focused on ground access to airports (e.g. Dai et al., 2013), landside accessibility to airports is usually calculated through GIS network analysis and based on digital road networks and assumed speed limits (Yang et al., 2016).

Overall, accessibility studies based on GIS network analysis have several key drawbacks (Salonen and Toivonen, 2013). First, most analyses are implicitly conducted for private vehicles and assume free-flowing traffic. Specifically, travel time on individual segments is calculated based on road length and the corresponding speed limit, while the optimal route is selected with shortest path algorithms. The ways traffic conditions vary across cities and through time are not considered (Kawabata and Shen, 2007; Liang and Zhang, 2018). Second and relatedly, less attention has been paid to other modes of transport, especially public transport. Mavoa et al. (2012) argue that more attention should be paid to analysing accessibility via public transport because it is in line with reducing car dependence as well as promoting social equity for the transit-reliant population. Most relevant to our purpose here, a main type of public transport accessibility studies concerns the overall access to a selected location (Lei and Church, 2010). Third, as GIS software in general has been weak at integrating time, network analysis has limited capacity to incorporate temporal considerations, such as transit schedules and peak hours. Recent attempts have been made to use online map and journey planning services and to characterise public transport accessibility within cities (Saghapour et al., 2016; Salonen and Toivonen, 2013). Murakami et al. (2016) have attempted to gather airport access information based on online map services; however, their measures cannot capture how access disparities between different transport modes vary within individual cities. Overall, the intracity impacts of air transportation and ground access to airports in particular are less well-researched in the Chinese context. Our study therefore aims to fill some of these voids and characterize the accessibility disparities between and within individual Chinese cities.

Planning and managing airport ground access in Chinese cities

The governance structure of Chinese cities and regions is described as ‘fragmented authoritarianism’ (Xu and Yeh, 2013; Wu, 2016). Following Xu and Yeh (2013), we focus on three dimensions in the activities of different functional and territorial state agents involved in planning and managing airport ground access in China. A first dimension concerns interministry interactions; as a socialist legacy, airport development in general and ground transportation in particular involve multiple functional branches of the government. As Xu and Yeh (2013: 134) state:

each ministry stipulates a sectoral plan for the infrastructure sector in charge … these ministries are constitutionally at the same administrative level and no one ministry has authority over another. This leads to interministerial rivalries and complex political maneuvering in practice.

Specifically, on the ‘hard’ infrastructure side, the Civil Aviation Administration of China (CAAC), a semi-ministry authority under the Ministry of Transport, is, among others, in charge of airport infrastructure development as well as regulating and ‘overseeing’ airport services (Qin, 2009). At the same time, key infrastructure projects such as airports, state expressways and urban rails (including urban rail links to airports) need be approved by the National Development and Reform Commission (NDRC). Nevertheless, the development of expressway and rail links to airports is within the policy portfolio of the Ministry of Transport and the Ministry of Housing and Urban-Rural Development (Zhang, 2014). For the provision of transit service, local bus and subway companies are often overseen by and sometimes directly managed by local branches of the Ministry of Transport, while shuttle buses are often provided or franchised by airport companies, which entail complicated ownership structures (Yang et al., 2008) and point to a second dimension regarding the state-market relationship. While a new Ministry of Natural Resources was established in March 2018 to consolidate other government roles in spatial and urban planning, it may be too early to judge the effects and effectiveness of this consolidation.

The second state–market dimension is associated with the liberalisation, privatisation and commercialisation of China’s aviation sector (Yang et al., 2008). Since the 1980s, China’s airport governance and management have followed the national trend of decentralisation, with local governments assuming the role of managing and operating most of the airports. The CAAC, which used to ‘own’ and operate all airports in China, is now largely removed from the day-to-day operations and focuses on the overall planning, supervision and regulation. Another key change is the introduction of private and foreign capital into the aviation sector, with some airports even becoming public companies and being listed on stock exchanges (Yuen and Zhang, 2009). Most relevant to our study here, these and other related policy reforms provide airports with more freedom in the provision of non-aeronautical services such as ground transport. For example, express buses at airports that provide direct service to selected locations within the city are often run by chartered individuals and companies, sometimes airport companies themselves. Similarly, in many Chinese cities, transit services such as local buses and urban rails have been transformed into state-owned enterprises (Qin, 2009). Further complicating the situation, different transport modes are often run by different companies, many of them being SOEs. For example, both Shanghai Metro and Shanghai Maglev are, through a web of ownership linkages, supervised by the Shanghai Municipal Governance and are providing urban rail service to Shanghai Pudong airport. Regardless of whether the services are provided by airport companies or by local bus/rail companies, service providers must balance between offering transit access to airports as a basic public service (i.e. the equity dimension) and delivering transit with cost-benefit considerations (i.e. the efficiency dimension; Saghapour et al., 2016).

With the decentralisation of state functions, a third dimension arises regarding interscalar relations. This dimension is most evident in the development of airport rail links. For example, in recent years, local governments have often made grand proposals about urban rails as an urban development strategy (Xu and Yeh, 2005). On the other hand, the NDRC – which is in charge of approving major infrastructure projects and must consider financial viability, the local debt situation and overall national policies – has recently raised the bar for urban rail project applications (State Council of China, 2018). Lastly but just as importantly, like other transport infrastructure projects, the construction of expressways and rail links for airport ground access may also face social issues such as land rights and environmental concerns (See for example, Chen et al., 2007; Zheng, 2010). These interactions and dynamics therefore produce diverse public transport systems between Chinese airports and their cities, and one of this article’s main objectives is to develop and apply techniques to compare airport access alternatives at Chinese cities.

Data and methods

Data collection

Following Salonen and Toivonen (2013), our measure of airport landside accessibility hinges on a series of routing analyses for public transport and private cars.

Determination of route origin and destinations: We divide individual cities into grid cells and treat cell centroids as route origins. Chinese cities are similar to metropolitan areas in North American and European contexts in that they consist of urban cores, outlying urbanised settlements such as towns and county seats, as well as rural areas. Airports are often located in the outlying rural areas, and most of the public transport services focus on connecting airports with the corresponding core urban areas. We used the grid cell alignments of the LandScan global population distribution dataset for the year 2015. The LandScan dataset captures the worldwide population distribution in approximately 1 km by 1 km grids, and has been applied as a ‘ground truth’ measure in population studies (Liu and Wang, 2016). The main terminals of individual airports are chosen as route destinations. For example, Terminal 3 of Beijing Capital International Airport is chosen as one destination. Specifically, destinations are set at terminal entrances so that our analysis does not include walking distance within airports.

Gathering route information from online map services: For an airport, we gather route information between all grid cells in the corresponding city and the airport terminal. Because transit information is not available in standard data formats (e.g. general transit feed specifications (GTFS)), we rely on online map services (such as Gaode Map, a Chinese equivalent of Google Map) for route information. These online services have accounted for both public transport schedules as well as real time traffic conditions. We have developed a Python-based web crawling tool that systematically queries and stores information about the quickest routes between cell grids and airport terminals. Specifically, for routes by public transport and private cars, we gather travel times. We also record the walking distance, fares and number of transfers for public transport. Walking distances within airport premises are not included by the online service. Our search is conducted under peak travel scenarios (i.e. 6–8 pm on weekdays), thus conservatively estimating the time advantage of private cars. Route information is collected during the last week of August 2018.

Landside accessibility across cities

Extending from Gutiérrez (2001), we have chosen three standard indicators to characterise the landside accessibility to individual airports. A first indicator evaluates the weighted minimum travel time between all grid cells and the airport. This indicator is calculated by using population in individual grids as weight.

T_{a} = \frac{\sum_{i = 1}^{n} t_{ai} \times po p_{i}}{\sum_{i = 1}^{n} po p_{i}}

where $T_{a}$ is the weighted minimum travel time of airport a, $t_{ai}$ represents the quickest or minimum travel time between airport a and grid i, $po p_{i}$ indicates the population of grid i and n denotes the total number of population cells in the corresponding city. This indicator measures the overall ease of accessing the airport from all grids within the corresponding city. It shows the length on average of a trip from a random grid to the city’s airport. It is calculated for all grids for which we can gather access information for transit and for car.

A second gravity-based indicator captures the ‘economic potential’ of airports. It measures the number of people who can travel to/from an airport, after accounting for the costs of travel. In our analysis, it is operationalised as follows:

G_{a} = \sum_{i = 1}^{n} \frac{po p_{i}}{{t_{ai}}^{b}}

where $G_{a}$ indicates the ‘economic potential’ of airport a, $t_{ai}$ represents the quickest or minimum travel time between airport a and grid i, $po p_{i}$ indicates the population of grid i, n denotes the total number of population cells in the corresponding city and b is the distance decay parameter. Following Gutiérrez (2001), b is set to 1 and the travel time for grids without transit/car access is set to a sufficiently large number (i.e. 999,999,999). While we use population to approximate the importance of individual grids, the variable can be substituted by gridded measures of economic activities, such as night-time remote sensing images. An airport with a larger ‘economic potential’ usually has a greater number of people with easy access to airports. A third indicator assesses the proportion of the population that can travel to/from the airport within 60 minutes. The second and third indicators are calculated for all grids in individual cities. We also calculate these indicators for both public transport and private cars.

Connections between city centres and airports

We pay special attention to the link between city centres and airports. Because most Chinese cities have a monocentric spatial structure for their core urban areas (Liu and Wang, 2016), identifying central business districts (CBDs) in individual cities is relatively straightforward. For example, Tiananmen Square is used to denote the city centre for Beijing, while for smaller cities we record the location of municipal governments because CBDs in these cities tend to collocate with them. To allow for errors in choosing CBDs, for each city centre–airport pair, we examine the nine grids around the chosen CBD and identify the grid with the median travel time. We subsequently record the corresponding travel time and estimated monetary costs by transit and taxi. Following Coogan (2000), we use estimated taxi fares as a surrogate measure for the costs of using private cars.

Study areas

The CAAC has officially designated (airports in) three cities as national gateways (Beijing, Shanghai and Guangzhou), eight as regional/primary hubs (Chongqing, Chengdu, Shenyang, Kunming, Wuhan, Xi’an, Urumqi and Zhengzhou) and 12 as secondary hubs, while Beijing and Shanghai are the only two cities in China with two operating civil airports. Our analysis includes all of these hub and gateway airports as well as those handling more than 5 million passengers in 2015 (CAAC, 2016b). We collect data for all population grids within the jurisdictions of individual cities, except two cities with very large administrative areas (Chongqing and Harbin), where we focus on the districts under direct municipal control. We envision this to be less an issue, since our focus is on the (relative) comparison between airport accessibility by transit and by cars within individual cities. In total, information is collected for 520,312 grids through more than 1 million callings to the map service website. The final sample includes 35 Chinese airports.

Results and discussion

Airport ground access by public transport

Buses are the main transit type at Chinese airports. Specifically, 18 of the 35 airports in our study are connected by multi-stop bus services, while all 35 airports are serviced by an airport shuttle. Furthermore, 13 cities have extended their urban rail systems into airports, while four cities have built dedicated airport express links.

Airport ground access by public transport varies within individual cities. Figure 1 illustrates the minimum travel time to airports for four selected cities (Shanghai, Nanjing, Wuhan and Yinchuan). It suggests that city centres, areas along major transport corridors (rails links and expressways) and, unsurprisingly, areas close to the airports usually have shorter travel times. Not all areas within the cities are serviced by public transport because the travel time by public transport quickly exceeds 120 minutes (Figure 1) once we leave city centres and major transport corridors. Yinchuan is a case in point, because we could not gather transit information for grids outside the city centres.

Figure 1.

Airport access time by public transport in selected cities.

Table 1 summarises accessibility indicators for individual airports. Airports with shorter population-weighted minimum travel time are usually closer to city centres as for example the distance between Xiamen, Urumqi and Sanya airports and their corresponding city centres is less than 20 kilometres – while the average distance between airports and city centres in our sample is 28.9 kilometres. Another explanation is the relatively small coverage of public transport in these cities, as for example, grids with public transport information (from the online map service) account for 64.48% and 41.54% of the population in Kunming and Nanning, respectively. Airports located in large cities, especially those with airport rail links, top the rankings of economic potential. Shanghai Hongqiao ranks top in this category because it is located in the city centre (Figure 1). As our analysis only includes intra-city public transit services, the results may underestimate the accessible area from Hongqiao airport, as the Shanghai Hongqiao Comprehensive Transport Hub integrates high speed rail, urban rail, and air-transport services (Huang et al., 2011; Li and Loo, 2016; Wang and Duan, 2018). Airports ranking top in the proportion of population reached within 60 minutes are a mix of those located closer to city centres (Xiamen, Guiyang, Taiyuan and Dailian) and those located in major cities with rail links (Shenzhen, Guangzhou, Chengdu and Shanghai Hongqiao).

Table 1.

Airport accessibility by public transport at selected airports.

Rank	City	Population-weightedminimum traveltime (minutes)	City	Economicpotential	City	Proportion ofpopulationreachedwithin 60 minutes (%)
1	Xiamen	61.40	Shanghai Hongqiao	5512.51	Xiamen	57.80
2	Urumqi	73.40	Shanghai Pudong	4296.76	Shenzhen	32.85
3	Sanya	74.38	Beijing Nanyuan	3949.58	Guiyang	30.70
4	Guiyang	76.52	Beijing Capital	3827.74	Guangzhou	26.93
5	Shenzhen	77.73	Guangzhou	2961.52	Taiyuan	26.45
6	Haikou	79.57	Chengdu	2726.97	Dalian	26.13
7	Kunming	83.46	Shenzhen	2639.00	Sanya	26.09
8	Shanghai Hongqiao	84.69	Tianjin	2101.74	Chengdu	25.54
9	Nanning	85.25	Wuhan	1836.46	Shanghai Hongqiao	25.42
10	Taiyuan	85.97	Hangzhou	1597.27	Urumqi	25.10
25	Nanchang	105.29	Jinan	938.00	Qingdao	6.84
26	Hefei	107.96	Taiyuan	911.29	Shanghai Pudong	6.05
27	Changsha	108.08	Guiyang	878.96	Tianjin	5.73
28	Lanzhou	113.39	Nanchang	867.78	Nanning	5.67
29	Fuzhou	116.34	Changchun	845.93	Shijiazhuang	5.29
30	Jinan	116.65	Urumqi	679.63	Zhengzhou	4.41
31	Xian	116.68	Nanning	664.36	Shenyang	4.41
32	Qingdao	122.22	Lanzhou	502.10	Jinan	2.36
33	Zhengzhou	123.02	Haikou	442.69	Nanjing	2.18
34	Tianjin	127.45	Yinchuan	329.70	Fuzhou	1.24
35	Shijiazhuang	131.65	Sanya	168.63	Lanzhou	0.79

Comparison of airport ground access by transit and by car

Figure 2 depicts the minimum travel time between individual grids and airports in the four selected cities. To allow for better visual inspection, Figures 1 and 2 are presented with the same map scale and legend. Airport ground access by transit has a significantly longer travel time and smaller service area when compared with access by cars. Most notably, the areas that can be accessed by transit from/to airports within 60 minutes are significantly smaller than those by cars in the four selected cities. On average, 14.55% of the population in our sample cities have direct transit access to cities (i.e. without transfers). Specifically, the population weighted minimum travel time by transit is on average 49.35 minutes longer than that by car (Table 2). The economic potential of individual airports from public transport is on average 46% of that from cars across the 35 cities in our sample. Guangzhou, Shenzhen, Beijing Nanyuan, Shanghai Hongqiao and Beijing airports have the highest ratios between economic potential by transit and economic potential by car. This is despite the fact that the proportion of population reached within 60 minutes by transit is 22% of that by car. For comparison purpose, we have applied our methodology to Hong Kong, one of the world leading cities in public transit. While Hong Kong has fairly extensive public transit coverage - grids with public transit information host 99.8% of the population in Hong Kong- the population-weighted minimum travel time for Hong Kong airport is 68.69 minutes and 75.82% of the population can access the airports directly by transit.

Figure 2.

Airport access time by car in selected cities.

Table 2.

Difference between airport access by transit and by car.

	Mean	Max.	Min.	Std. deviation
Population weighted minimum travel time (minutes)
By transit	98.39	131.65	61.40	16.74
By private car	48.80	69.02	28.42	9.78
Difference (transit – car)	49.58	76.57	24.35	13.26
Ratio (transit/car)	2.06	2.73	1.45	0.34
	Economic potential
By transit	1591.55	5512.51	168.63	1219.71
By private car	3238.98	9714.96	476.94	2015.21
Difference (transit – car)	−1647.43	−308.31	−4202.45	887.14
Ratio (transit/car)	0.47	0.70	0.32	0.09
	Proportion of population reached within 60 minutes (%)
By transit	15.13	57.80	0.79	12.08
By private car	67.74	99.53	21.19	17.92
Difference (transit – car)	−52.61	−20.39	78.02	15.93
Ratio (transit/car)	0.22	0.58	0.02	0.15

Figure 3 further illustrates how accessibility disparities by transport modes vary across cities. Consistent with our previous findings, the differences between airport access by transit and by car are smallest in city centres, as well as along major transport corridors such as rail links and expressways. Notwithstanding the different sets of destinations (e.g., employments, health services, and food environment) and methods of characterizing transit accessibility, existing studies differ in that whether accessibility gaps between different modes of transportation are greater in city centers or suburbs (Salonen and Toivonen, 2013). This debate is also relevant for China, as in recent years multiple rounds of new town development have been experienced and more polycentric urban patterns are being pursued in many Chinese cities (Liu and Wang, 2016). As for international case studies, Benenson et al. (2011) conclude that the ratio between accessibility by transit and by car decreases from the city center of Tel Aviv to outskirts. Still, exploring the accessibility of supermarkets in Cincinnati, the US, Widener (2017) suggests that relatively good access for both transit and cars is seen in the city center if closest facility accessibility measure is employed. Our results are consistent with Kawabata (2009) and Salonen and Toivonen’s (2013) findings that accessibility gaps are smaller in city centres.

Figure 3.

Ratio between access time by transit and by car.

Connections between city centres and airports

The following section focuses specifically on the linkage between airports and their corresponding city centres. Many cities have used rapid transit access, especially rail links, as a way to promote city image and central business district development (Niedzielski and Malecki, 2012; Hirsch, 2016). Figure 4 compares the time and price ratios between public transport and taxis at individual airports. Individual airports are plotted as ‘bubbles’ on a scatterplot, with their time and price ratios as coordinates. By centring on the median values of time and price ratios, Figure 4 is divided into four quadrants.

Figure 4.

A typology of airport–city transport connections.

Lower-left quadrant

Airports in the lower-left quadrant are associated with below-average time and price ratios. Thus, the travel time between the airport and city centre by transit is not far from that by private car, while the monetary cost by transit is substantially lower than that by taxi. The fastest public transport options at these airports are usually urban or commuter rails, with Shanghai Hongqiao, Guangzhou Baiyun, Shenzhen Bao’an and Kunming Changshui airports in this category. Indeed, rail links to airports are often heralded for their reliability and speed. The effects of travel time reliability on travel mode choice have been observed in international cases (Kato et al., 2017). However, the ridership of these rail links seems to be less than ideal. For example, Zhang et al. (2013) suggest that rail links account for less than 10% of ground access at Shenzhen airport. Similarly, Xu (2012) estimates that only 3% of the airline passengers at Guangzhou Baiyun International Airport used the rail link in 2010.

Upper-left quadrant

Airports in the lower right corner may be characterised by below-average time ratio and above-average price ratio, i.e. relatively quick but expensive public transport. The above-average price ratio should be interpreted in relative terms, i.e. the price of a taxi is still higher than the price of public transport at these airports. Most airports in this category rely on airport shuttles, which are more expensive than multi-stop buses. The shuttle bus services usually use controlled-access expressways between these airports and the corresponding city centres.

The airport express at Shanghai Pudong is also in this category, with rail links to Beijing Capital being a marginal case. Similar to airports in the first category, ridership of rail links is comparatively low. According to our collected information, the shortest paths (as measured by travel time) between city centers and Shanghai Pudong airport involve the use of Maglev Train. Coogan et al. (2008) explain that in light of the relatively expensive price and indirect connections of the rail link in Shanghai, passengers with smaller budgets opt for the cheaper shuttle buses while business travellers prefer the direct taxi service. Specifically, the city-end terminals of these airport express rail links (not extended urban rails) are located in the periphery. Value-added services such as in-town check-in and luggage handling are not offered at the city terminals of airport rail links, making airport rail links less appealing, especially for business travelers. The development, construction, and operation of Shanghai Maglev Train entail the interministry, interscalar, and state-market interactions highlighted in the literature review section. As for politics and planning at the interministrial and interscalar levels, Shanghai Maglev Train was related to the debate on which technologies and systems should be used to develop high-speed ground transport systems between Beijing and Shanghai (Liu and Deng, 2004). The Ministry of Science and Technology and then Ministry of Railway were mainly involved in this debate, with High Speed Rail (HSR) and magnetic levitation (Maglev) as the two main options. Against this backdrop, Shanghai Municipal government was tasked to launch Shanghai Maglev Train as an experimental line for the Maglev system in 2003 (Gao, 2003; Zheng, 2010). As China has focused on HSR development in the last decade (Wang and Duan, 2018; Yang et al., 2018), the originally planned extensions to the Shanghai Maglev Line did not materialize, limiting the system’s connectivity, service areas, and subsequently ridership. Specifically, the Maglev extensions are not included in the third phase (2018-2023) development plans of Shanghai urban rail transit. Since 2010 Shanghai Maglev Train faced further competition, as Shanghai Metro Line 2 was extended to the Pudong airport. As mentioned, both Shanghai Metro and Shanghai Maglev are subsidiaries of Shanghai Shentong Metro Group, a state-owned enterprise supervised by the Shanghai Municipal Government. While Shanghai Shentong started as Shanghai Metro Company and has been running Shanghai’s metro system since early 1990s, Shanghai Maglev was started as a joint project between Shentong and a few other state-owned enterprises (see Wu (2018) for a more detailed discussion about urban governance and spatial development in China and in Shanghai in particular). The issue of inconvenient interchanges can also be seen in the Beijing Capital airport express. The city-end station is not well-integrated with city bus or urban rail systems. Passengers wishing to take the subway must walk for at least 500 meters, in addition to hassles involved in entering and exiting platforms and stations.

Upper-right quadrant

Public transport access at airports in the upper-right corner may be characterised by above-average time and price ratios, i.e. relatively slow but expensive transit. Although public transport at most airports in this category relies on airport shuttles rather than multi-stop buses, the travel speed is limited because seven out of the eight shuttle services recorded in our data collection do not use expressways. A recent change in domestic airfares can be a ‘wild card’ in nudging price-sensitive passengers into using public transport. In the past, the government, through the CAAC and the National Development and Reform Commission, has controlled domestic airfares by administrative orders. Such regulations were loosened in early 2018, potentially leading to surging domestic airfares and leaving a smaller budget for ground transportation to airports, everything else being equal. Relatedly, the impacts of low-cost carriers (LCCs; de Neufville, 2006) is less of a concern at the moment, because their market share is in the single digits in China. By contrast, the three major carriers – China Southern, Air China and China Eastern – accounted for two-thirds of the market share in 2016 (CAAC, 2016a). LCC users have relatively lower willingness-to-pay for savings in time and may choose public transport. However, we notice the emergence of LCCs in Asia in general and in China in particular (Fu et al., 2015).

Lower-right quadrant

For the lower-right corner, ground access for airports in this quadrant has an above-average time ratio and a below-average price ratio. Public transport at these airports mostly relies on the relatively cheap but slow multi-stop buses. The ratios should be interpreted in relative terms. For example, despite its quick public transport access in absolute terms (Table 1), Xiamen airport falls into this category because its public transport access is much slower than travel by car.

Conclusion

Recent literature has highlighted the importance of airports for urban planning development (Ryerson, 2016). The success of airports and air-based economies depends on both connections with the air-transport networks and ground access between airports and their hosting cities (i.e. landside transport of airports). Despite its significance, little research has critically examined ground access to airports in China. A few case studies have used GIS-based road network analysis to characterise airport ground access, without accounting for public transport, transfers and traffic conditions. In this context, this study takes the first step of exploring the overall provision of airport ground access by public transport in China. Extending from Salonen and Toivonen (2013), we have developed a new approach to using online map and journey planning services and characterising how airport landside access varies within individual cities and between different transport modes.

In terms of airport ground access by public transport, city centres and areas near major transport infrastructure usually result in shorter minimum travel times. Airport ground access by transit has significantly longer travel time and a smaller service area when compared with access by car. The empirical measures for airport ground access between public transport and car suggest substantially less ground access by public transport than by private car. Still, Benenson et al. (2011, p. 514) warns that such significant accessibility gap may result from “a more detailed representation of travel by transit”. The accessibility disparities between public transit and cars vary within and across cities, as suggested by our detailed accessibility mapping within individual cities as well as a typology of airport ground access. Consistent with previous studies, the disparity of airport ground access by cars and by public transit is relatively low in city centres.

Our exploratory analysis, review of academic and professional materials, and anecdotal familiarity with individual airports point to some ‘speculative’ ideas and recommendations. First, because substantive accessibility disparities exist between public transport and cars, the provision of transit at airports leaves much room for improvement. Airport ground access could be an overlooked area for the aviation industry to reduce its carbon emission and for urban professionals to remedy increasingly auto-orientated Chinese cities (Qin, 2009). Second, for cities that already have rail transit or are planning/building rail links, attention should be paid to reducing travel time and alleviating the hassles in making transfers. In this regard, Shanghai Hongqiao airport shows initial signs of integrating different transit modes (Li and Loo, 2016). Expressways seem to be important for airports that rely on shuttle and bus services. Third and more generally, to facilitate ground access to airports and develop airports as transportation hubs, airport and urban planning need to have better synergies. As we discussed in the literature review, spatial and infrastructure development in China involves multiple state agents under different functional ‘branches’ and territorial jurisdictions (Wu, 2018). As airport-led development strategies are increasingly adopted in Chinese cities, synthesis and collaboration in this fragmented governance landscape become a pressing issue (Zhang, 2014). Fourth, the dissemination of public transport information should be improved. Our experience with airport websites suggests that information about public transport is not very detailed and can be more often accompanied by interactive maps. While China is advancing rapidly in information technology and economies, it might not necessarily be true that map and ride-sharing mobile applications can replace public information or that all passengers are equally technologically savvy.

The current analysis can be enriched and improved on the following fronts. First, our exploratory analysis focuses on the ‘supply’ rather than the ‘demand’ side of public transport, paying little attention to how users with different trip purposes, socioeconomic backgrounds and geographic distributions are using the service. Similarly, the overall mapping may overlook how local economic, political, and geographic contexts shape airport ground access systems. A subsequent analysis can determine the needs and demands of different users, the interests of various stakeholders (including airport authorities and relevant government bodies), as well as the potential conflicts and interactions therein (see for example, Hirsh, 2016). Still, while the taxi fares used in the current study is estimated by the online mapping platform based on travel distance, a demand-side study may reveal the impacts of fare policies in individual cities (e.g., Castillo-Manzano and Sánchez-Braza, 2011). That being said, the population-weighted indices in our current analysis could shed some lights on the overall demands, as airport-related trips are not necessarily concentrated in the central business districts and may follow the overall population distribution (see also the usefulness of population density as proxies of freight landscape in Giuliano et al., 2018). Our empirical framework can also be applied to assess intracity accessibility of rail stations and other transport hubs, as case studies by Diao et al. (2017, p. 2249) suggest that “inter-city travel brought by HSR are largely offset by the prolonged intracity travel time to reach the stations”. Furthermore, as airports often serve wider regions, future studies should include intercity transport links with airports. Second, because the environmental and economic impacts of individual access modes, especially airport rail links, are much heralded, the next step will be relating airport ground access to measures of environmental efficiency and economic productivity (Murakami et al., 2016). Attempts should be made to characterise how airport ground access and airport development in general have spurred land use and transport impacts at the intracity level (Cidell, 2014). Lastly, because our analysis relies on information provided by online map services, the critiques about digital cities and/or web-based urban information can also be applied to our measures (e.g. Rabari and Storper, 2014). Still, as our analysis has demonstrated, no systematic database exists for ground access systems and their performance for Chinese airports. Specifically, very few public records describe passenger characteristics, modal choice patterns for passengers and employers or architectural details of ground access systems. Our study, therefore, serves as a first step towards systematic data collection for benchmarking and improving airport ground access.

Footnotes

Acknowledgements

The author is very grateful for the insightful comments and suggestions made by the editor and anonymous reviewers. Sincere thanks to Xiande Li, Yuqi Liu, Xiang Luo, Lei Wang and Wangtu Xu for comments made on earlier versions of this paper, as well as to Xiuxin Ma and Nannan Wang for research support. All remaining errors are the author’s own. The usual disclaimers apply.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author is grateful for the financial support from the Hong Kong Research Grants Council [grant number 27604016 and 17600918].

ORCID iD

Xingjian Liu

References

Appold

Kasarda

(2013) The airport city phenomenon: Evidence from large US airports. Urban Studies 50(6): 1239–1259.

Benenson

Martens

Rofé

, et al. (2011) Public transport versus private car GIS-based estimation of accessibility applied to the Tel Aviv metropolitan area. The Annals of Regional Science 47(3): 499–515.

Budd

Ison

Budd

(2016) Improving the environmental performance of airport surface access in the UK: The role of public transport. Research in Transportation Economics 59: 185–195.

Budd

Ison

Ryley

(2011) Airport surface access in the UK: A management perspective. Research in Transportation Business & Management 1(1): 109–117.

Budd

Ryley

Ison

(2014) Airport ground access and private car use: A segmentation analysis. Journal of Transport Geography 36: 106–115.

Castillo-Manzano

J I

Sánchez-Braza

(2011) An evaluation of the establishment of a taxi flat rate from city to airport: the case of Seville. Urban Studies 48(9): 1909–1924.

Chen

Tang

Huang

, et al. (2007) High-speed maglev noise impacts on residents: A case study in Shanghai. Transportation Research Part D 12(6): 437–448.

Cidell

(2014) The role of major infrastructure in subregional economic development: An empirical study of airports and cities. Journal of Economic Geography 15(6): 1125–1144.

Civil Aviation Administration of China (CAAC) (2016a) 2015 Civil Aviation Industry Development Statistics Bulletin. Beijing: CAAC.

10.

Civil Aviation Administration of China (CAAC) (2016b) 2015 Statistical Data on Civil Aviation in China. Beijing: CAAC.

11.

Coogan

(2000) Improving public transportation access to large airports. Transit Cooperative Research Program Report 62. Washington, DC: Transportation Research Board.

12.

Coogan

MarketSense Consulting LLC and Jacobs Consultancy (2008) Ground access to major airports by public transportation. ACRP (Airport Cooperative Research Programme) Report 4. Washington, DC: Transportation Research Board.

13.

Dai

Zhang

Zhao

(2013) Using accessibility measures to assess spillover effect of civil airports in China. Acta Geographica Sinica 68: 1668–1677 [in Chinese].

14.

Diao

Zhu

(2017) Intra-city access to inter-city transport nodes: The implications of high-speed-rail station locations for the urban development of Chinese cities. Urban Studies 54(10): 2249–2267.

15.

de Neufville

(2006) Planning airport access in an era of low-cost airlines. Journal of the American Planning Association 72(3): 347–356.

16.

Freestone

(2009) Planning, sustainability and airport-led urban development. International Planning Studies 14(2): 161–176.

17.

Lei

Wang

, et al. (2015) Low cost carrier competition and route entry in an emerging but regulated aviation market: The case of China. Transportation Research Part A: Policy and Practice 79: 3–16.

18.

Gao

(2003) Planning and practice of the Electromagnetic suspension. Urban Planning Forum, 143, 33–36 [in Chinese].

19.

Gibbons

(2017) Airports, market access, and local economic performance: Evidence from China. SERC Discussion Paper 211. Available at: http://eprints.lse.ac.uk/83630/1/sercdp0211.pdf (accessed on 29 March 2019).

20.

Giuliano

Kang

Yuan

(2018) Using proxies to describe the metropolitan freight landscape. Urban Studies 55(6): 1346–1363.

21.

Guerra

Cervero

(2011) Cost of a ride: The effects of densities on fixed-guideway transit ridership and costs. Journal of the American Planning Association 77(3): 267–290.

22.

Gutiérrez

(2001) Location, economic potential and daily accessibility: an analysis of the accessibility impact of the high-speed line Madrid–Barcelona–French border. Journal of Transport Geography 9(4): 229–242.

23.

Hirsch

(2016) Airport Urbanism: Infrastructure and Mobility in Asia. Minneapolis, MN: University of Minnesota Press.

24.

Huang

Yang

(2011) Airport passenger flow characteristics: Analyzing Shanghai Hongqiao hub. Traffic and Transportation 2011: 133–136 [in Chinese].

25.

Ison

Merkert

Mulley

(2014) Policy approaches to public transport at airports—Some diverging evidence from the UK and Australia. Transport Policy 35: 265–274.

26.

Kasarda

Lindsay

(2011) Aerotropolis: The Way We’ll Live Next. New York: Farrar, Straus and Giroux.

27.

Kato

Fukuda

Yamashita

, et al. (2017) Latest urban rail demand forecast model system in the Tokyo Metropolitan Area. Transportation Research Record: Journal of the Transportation Research Board 2668: 60–77.

28.

Kawabata

Shen

(2007) Commuting inequality between cars and public transit: The case of the San Francisco Bay Area, 1990-2000. Urban Studies 44(9): 1759–1780.

29.

Lei

Church

(2010) Mapping transit-based access: Integrating GIS, routes and schedules. International Journal of Geographical Information Science 24(2): 283–304.

30.

Loo

(2016) Towards people-centered integrated transport: A case study of Shanghai Hongqiao Comprehensive Transport Hub. Cities 58: 50–58.

31.

Liang

Zhang

(2018) Assessing the public transport service to urban parks on the basis of spatial accessibility for citizens in the compact megacity of Shanghai, China. Urban Studies 55(9): 1983–1999.

32.

Liu

Deng

(2004) Comparing operating characteristics of high-speed rail and maglev systems: case study of Beijing-Shanghai corridor. Transportation Research Record (1863): 19–27.

33.

Liu

Wang

(2016) How polycentric is urban China and why? A case study of 318 cities. Landscape and Urban Planning 151: 10–20.

34.

Liu

Derudder

(2016) Measuring polycentric urban development in China: An intercity transportation network perspective. Regional Studies 50(8): 1302–1315.

35.

Loo

Chow

(2011) Spatial restructuring to facilitate shorter commuting: An example of the relocation of Hong Kong International Airport. Urban Studies 48(8): 1681–1694.

36.

Timberlake

(2013) World city typologies and national city system deterritorialisation: USA, China and Japan. Urban Studies 50(2): 255–275.

37.

Mavoa

Witten

McCreanor

, et al. (2012) GIS based destination accessibility via public transit and walking in Auckland, New Zealand. Journal of Transport Geography 20(1): 15–22.

38.

Murakami

Matsui

Kato

(2016) Airport rail links and economic productivity: Evidence from 82 cities with the world’s 100 busiest airports. Transport Policy 52: 89–99.

39.

Niedzielski

Malecki

(2012) Making tracks: Rail networks in world cities. Annals of the Association of American Geographers 102(6): 1409–1431.

40.

Qin

(2009) Linking airports and cities: Planning for mass transit at large airports. Beijing: China Architecture & Building Press [in Chinese].

41.

Rabari

Storper

(2014) The digital skin of cities: urban theory and research in the age of the sensored and metered city, ubiquitous computing and big data. Cambridge Journal of Regions, Economy and Society 8(1): 27–42.

42.

Ryerson

(2016) Planners take flight: Integrating air transportation into planning education. Journal of Planning Education and Research 36(4): 427–439.

43.

Saghapour

Moridpour

Thompson

(2016) Public transport accessibility in metropolitan areas: A new approach incorporating population density. Journal of Transport Geography 54: 273–285.

44.

Salonen

Toivonen

(2013) Modelling travel time in urban networks: Comparable measures for private car and public transport. Journal of Transport Geography 31: 143–153.

45.

State Council of China (2017) Notes on the modern transport system plan for the thirteenth ‘Five-Year’ Plan. Accessed via: http://www.gov.cn/zhengce/content/2017-02/28/content_5171345.htm [in Chinese].

46.

State Council of China (2018) Opinions of the General Office of the State Council on regulating urban rail transit planning and construction. Accessed via: http://www.gov.cn/zhengce/content/2018-07/13/content_5306202.htm [in Chinese].

47.

Wang

Duan

(2018) High-speed rail network development and winner and loser cities in megaregions: The case study of Yangtze River Delta, China. Cities 83: 71–82.

48.

Wang

Bonilla

Banister

(2016) Air deregulation in China and its impact on airline competition 1994–2012. Journal of Transport Geography 50: 12–23.

49.

Widener

(2017) Comparing measures of accessibility to urban supermarkets for transit and auto users. The Professional Geographer, 69(3): 362–371.

50.

(2016) China’s emergent city-region governance: A new form of state spatial selectivity through state-orchestrated rescaling. International Journal of Urban and Regional Research 40: 1134–1151.

51.

(2018) Planning centrality, market instruments: Governing Chinese urban transformation under state entrepreneurialism. Urban Studies 55(7): 1383–1399.

52.

Yeh

(2005) City repositioning and competitiveness building in regional development: New development strategies in Guangzhou, China. International Journal of Urban and Regional Research 29(2): 283–308.

53.

Yeh

(2013) Interjurisdictional cooperation through bargaining: The case of the Guangzhou–Zhuhai railway in the Pearl River Delta, China. The China Quarterly 213: 130–151.

54.

(2012) The analysis of the characteristics and problems on Chinese hub airport ground transportation organization. Huazhong Architecture 2012: 69–73. [In Chinese].

55.

Yang

Dijst

Witte

, et al. (2018) Comparing passenger flow and time schedule data to analyse High-Speed Railways and urban networks in China. Urban Studies 56(5): 1267–1287.

56.

Yang

Tok

(2008) The privatization and commercialization of China’s airports. Journal of Air Transport Management 14(5): 243–251.

57.

Yang

Notteboom

(2016) Airport location in multiple airport regions (MARs): The role of land and airside accessibility. Journal of Transport Geography 52: 98–110.

58.

Yuen

Zhang

(2009) Effects of competition and policy changes on Chinese airport productivity: An empirical investigation. Journal of Air Transport Management 15(4): 166–174.

59.

Zhang

(2014) The coordination of airport planning and urban planning. China Civil Aviation 187: 47–48 [in Chinese].

60.

Zhang

(2013) Optimizing landside transportation of Beijing Capital International airport. China Transportation Review 2013: 15–20 [in Chinese].

61.

Zheng

(2010) Dilemma in NIMBY facility planning: A case study of magnetically levitated train in Shanghai. City Planning Review 2011: 74–86 [in Chinese].