Page 373 - 360.revista de Alta Velocidad - Nº 6

The configurations of Chinese national urban systems in both high-speed railway and airline networks

different city classes in both networks. Although the three cities as national cores have the
similar socio-economic performance in China, Beijing and Guangzhou are only in the first
class of airline networks but Shanghai in the first class of both airline and HSR networks.
This can be explained by two reasons:

Firstly, Beijing and Guangzhou’s average distances to other cities (828 km and 1034 km) are
larger than Shanghai’s (723 km) in HSR networks, making air travel more attractive than
HSR travel due to a shorter travel time.

Secondly, HSR networks in the densely populated YRD are much more developed (e.g. higher
density and frequency of HSR networks) than in the Bohai Rim and the PRD, leading to more
functional interactions of cities with Shanghai in the YRD than that with Beijing in the Bohai
Rim and Guangzhou in the PRD by HSR networks.

Meanwhile, regarding the geographical location of Nanjing in the YRD region with completely
developed HSR networks and its role of being a regional socio-economic core rather than a
national one, it is not surprising to find that Nanjing is in the second class of HSR networks
but the third class of air networks. Except to Shenzhen as a sub-provincial city in the east,
Chongqing as a municipality city, Chengdu, Kunming, and Xi’an as provincial capitals in
the west are in the second class of air networks, which reflects that major cities being as
regional socio-economic cores in the west are competitive for the air travel than in the
east.

In the third class of dominant cities, there are 25 HSR cities most of which are mainly
regional capitals and economic centers in the middle and the east (i.g. Wuhan in Hubei
province and Hangzhou in Zhejiang province) of the country. Only Chengdu and Chongqing
are located in the west of China offering only connections between each other and not with
the rest of the country. In the third class of dominant cities we also find 30 airline cities
most of which are provincial capitals and economic centers in the middle and east, but
include in comparison with HSR cities more provincial capital cities in the west, such as
Urumchi, Guiyang, Nanning, Lanzhou and Yinchuan and typical tourism cities such as Sanya
and Guilin.

In order to identify the different positions of a city in HSR and airline networks, we further
identified the HSR and airline advantage cities by comparing the differences of city strength
values (DIT) between HSR and airline networks among the 51 HSR-Airline cities. If one city’s
DIT value of HSR networks is larger than that of airline networks, it is considered as a HSR
advantage city or otherwise an airline advantage city in Figure 4. Furthermore, Pearson's r
(city strength correlation coefficient) and Spearman' rho (city rank correlation coefficient)
test are used to identifying whether there is a direct correlation for 51 HSRAirline cities
between HSR and airline networks.

The associations between the two networks for HSRAirline cities by the values of city strength
and related rankings are statistically significant: Pearson's correlation coefficient is 0.871
(p < 0.01) and Spearman's rho is 0.788 (p < 0.01). This means that a city is dominant and
highly ranked in one network should be the same in another network, thus it is necessary
to not only take into account the large absolute value (DIT_HSR minus DIT_Airline >1 or
DIT_Airline minus DIT_HSR>1) but more importantly the class changes to identify those
outlier cities between two transportation networks.

International Congress on High-speed Rail: Technologies and Long Term Impacts - Ciudad Real (Spain) - 25th anniversary Madrid-Sevilla corridor 371