The Chengdu-Chongqing Economic Circle (CCEC) was first officially proposed in the “Plan Outline for the Construction of the Chengdu-Chongqing Twin City Region” issued by China’s National Development and Reform Commission in 2020. The economic circle belongs to the concept of Economics and the scope of the CCEC (Figure 1) coincides with the Chengdu-Chongqing urban agglomeration (CCUA). CCUA is one of the four major urban agglomerations in China, and it is also the largest urban agglomeration in western China. CCUA takes the Chengdu metropolitan area and the Chongqing metropolitan area as the dual-core, including 15 cities of Sichuan and 27 districts in Chongqing, with a total area of about 185,000 square kilometers, a permanent population of 95 million (2020), and a GDP of 8.58 trillion (2022). It is positioned as a world-class urban agglomeration. CCUA occupies an important position in the development of Chinese cities and transportation systems. It is listed by China as the fourth pole of the national three-dimensional transportation network in 2021. CCAU has an obvious dual-core characteristic, densely covered with multi-node cities, and it is an important representative of China’s large urban agglomerations. In addition, CCUA is a typical basin-type urban agglomeration, with important rivers distributed, and an important part of the Yangtze River Economic Belt, with unique geographical patterns and location conditions.

We use 5 datasets in this study (Table 1), which are open street map (OSM), point of interest (POI), night light raster (NLR), Sentinel-2 image, and surveys and statistics, respectively. All datasets are available on their official website, except surveys and statistics.

OSM is a free and open geographic database updated and maintained by a community of volunteers via open collaboration (Zhao et al., 2019). We downloaded the latest map data of China from the OSM server through Geofabrik, including point, line, and polygon layers. The raw data is in Shapefile format, and geographic information system (GIS) software is used to clip, coordinate transform, and topologically process the data, and finally obtain a standardized road network in the CCTEC.

2) POI

POIs big data is useful to support an understanding of the complex human-land system element of human-economic geography (Xue et al., 2022). POI’s essence is a vector point and it is a record of a place on a map that we find useful or interesting. A POI is typically defined by its geographical coordinates and a few additional attributes like name and category. The POI used is from the Baidu Map server, the data information is more abundant than OSM and Google, especially in China.

3) Nighttime light

The nighttime light remote sensing data used in this paper is a new-type dataset (Wu et al., 2021) optimized based on DMSP-OL (Elvidge et al., 1997) and SNPP-VIIRS (Elvidge et al., 2021). The dataset has a spatial resolution of 1,000 m × 1,000 m and a time series of 1992–2021. In addition, we also used another night light data (130 m × 130 m) of Luojia-1 with higher resolution, which was mainly used for verification. The above two lighting data can be downloaded from Harvard Dataverse and Wuhan University databases respectively.

4) Sentinel-2

The Sentinel-2 images are derived from the Sentinel series satellites launched by the European Space Agency, which contain 13 bands and three types of spatial resolutions: 10, 20, and 60 m. The satellite revisit cycle is 5-day, suitable for monitoring earth surface changes and urban expansion (Wang et al., 2022b; Williams et al., 2022). The study area needs 12 scenes to cover all and the preprocessing (e.g., radiometric calibration, atmospheric correction, and band combination) was performed on the Google Earth Engine platform (Wang et al., 2022c). Finally, we obtain the natural color image with a spatial resolution of 10 m.

5) Surveys and Statistics

Excepting map and image data, we also used the TSP and special plan materials (2022), socio-economic statistics (1990–2021), basic geographic information data (latest version), and previous site survey data (2018–2022).

This research framework (Figure 2) includes three sections. Firstly, we analyze the urban system and traffic layout scheme in the provincial TSP and explain the possible problems. And then the spatial analysis methods are employed to analyze the urban expansion direction, multi-layer transportation network, and population economic spatial distribution. Moreover, we combined high-resolution remote sensing data and field survey data to verify the analysis results. Finally, an optimization and refinement scheme of the urban system layout is proposed. Specifically, the basic method of spatial analysis is as follows.

The expansion fan map (Figure 3) of the main cities in CCEC was obtained by analyzing the remote sensing data of night light from 1992 to 2021. The fan map can present the characteristics of the urban system and the direction of urban expansion in the past 3 decades. Deyang, Meishan, and Jianyang are included in the development plan of the Chengdu metropolitan area in the provincial TSP plan. It can be seen (Figure 3A) that every 5 years since 1992, the speed of Chengdu’s outward expansion has been relatively high (UEI = 2.89). Especially in the north, south, and southeast, incorporating Deyang, Meishan, and Jianyang into the Chengdu metropolitan area. In the recent 10 years, night light changes show that Chengdu has a significant expansion direction, particularly in the east and south (UEI = 1.98). Leshan and Mianyang are two important cities adjacent to the Chengdu metropolitan area, but they are not included in the metropolitan area due to spatial distance. The fan map shows that the urban expansion of Leshan City is not fast, and the main body of the city extends along major rivers such as the Minjiang River, Qingyi River, and Dadu River. The expansion characteristics of Mianyang City and Leshan City are relatively similar, and the intensity of urban expansion is not high. The main body of the city is basically connected with Deyang City in the south and expands to Santai County along the Fujiang River. Generally, a north-south urban belt has been formed, with the Chengdu metropolitan area as the core, connecting Leshan to the south and Mianyang to the north.

Yibin City is a key part of the SSCW city group in the TSP plan, it expands eastward and northwestward along the Minjiang-Yangtze River valley (Figure 3B). Especially in recent decades, urban has sprawl rapidly, and the central urban region gradually connected with Nanxi District and expanded to Jiang’an County (UEI = 2.16). Luzhou City, which is also an important city in the SSCW city group, is mainly expanded to the north due to the restrictions of administrative divisions (Figure 3B). It connects with Chongqing City through Lu County. In addition, Zigong City is a node city of SSCW, and its expansion mainly presents the characteristics of northwest and southeast directions (Figure 3B). In recent years, Zigong has gradually integrated with Neijiang in the east. In contrast, it has not achieved coordinated expansion with Yibin and Luzhou in the south. Overall, Yibin and Luzhou basically developed along the Yangtze River, connected the counties and districts along the river, and gradually formed an urban belt along the upper reaches of the Yangtze River.

The CCC city group contains three cities, Nanchong, Suining, and Guang’an. As the core city of the CCC city group in the TSP plan, Nanchong mainly expands southward along the Jialing River and reaches Xichong County in the northwest (Figure 3C). Suining mainly expands along the Fujiang River in the northwest-southeast direction and connects with Tongnan District in the south (Figure 3C). The characteristics of Guang’an’s urban expansion are similar to those of Leshan City, and it mainly develops into a dual-core city with its county-level city (Huaying City) under its jurisdiction. The difference from Leshan’s expansion is that Guang’an has been included in the development plan of the Chongqing metropolitan area. In view of the trend of urban expansion, the connection between the main node cities of the CCC city group is not high, lacking the support of county-level node cities.

As shown in Figure 3D, the WDK city group has a smaller size than the two city groups mentioned above. Dazhou and Wanzhou are the two core cities in the WDK city group. Judging from the trend of urban expansion (Figure 3D), the two cities have not developed in a coordinated way, and the degree of connection is not high. Chongqing Municipality is the core of the Chongqing metropolitan area, located in the southeast of the study area. Since 2006, the main body of Chongqing has expanded to the west, northwest, and northeast (UEI = 1.85), gradually connecting with Dazu, Yongchuan, Changshou, and Fuling. Chongqing City has expanded to Fuling District, Dazu District, Yongchuan District, and Guang’an City in the past 5 years, but there is an interesting phenomenon that the expansion has no focus in terms of space compared with the Chengdu metropolitan area. Moreover, the scale and night light intensity of the Chongqing metropolitan area is lower, especially in the core urban area.

A transportation network system with intermodal transportation and effective connection is an important infrastructure for the spatial integration of urban groups. By analyzing the characteristics of CCEC’s multi-layer traffic network, the commuting mode and commuting time between cities can be accurately obtained, which is the basic index for judging the efficient interconnection between cities in a city group or metropolitan area. The multi-mode transportation system includes highways, high railways, waterways, and aviation in CCEC. We use OSM data and traffic planning data-based spatial analysis to calculate the commute time of the CCEC high railway and a high-speed way to analyze the characteristics of a multi-layer traffic network (Figure 4).

The Chengdu and Chongqing metropolitan areas have a dense highway network and have formed a unique geometric pattern and spatial structure. Taking Chengdu as the center to Deyang, Jianyang, and Meishan, forming a typical ring-shaped high-way network structure. After calculation, it can be seen that the commuting time to the three cities is 0.44, 0.35, and 0.45 h respectively. The Chengdu metropolitan area realizes the expressway for half an hour to travel, which provides convenient infrastructure support for the effective connection of node cities. Chongqing’s ring-shaped feature is not obvious, but it is connected to important node cities by a highway network, and the expressway transportation is convenient, forming a radial highway transportation system along the terrain with the central region of Chongqing as the core. Starting from the central urban area, it takes 0.74 and 0.64 h to reach Rongchang District and Changshou District respectively. It takes more time (0.91 and 0.85 h) to go to Guang’an City and Qianjiang District. It can be seen that the east-west travel time of the Chongqing metropolitan area is lower than the north-south travel time. Generally, the Chengdu metropolitan area is more convenient than the Chongqing metropolitan area for traveling by highway. The CCC city group and the SSCW city group have a relatively high expressway network density, basically forming an urban expressway with a ring-like structure. However, due to the relatively small scale of the city, it has not developed into a standard ring-shaped urban structure similar to Chengdu. The commuting time from Nanchong, the central city of the CCC city group, to Suining and Guang’an, is about 0.56 h. Similarly, the commuting time from Yibin, which is also a central city, to Zigong and Luzhou expressways is about 0.59 h in the SSCW city group. However, the highway network of the WDK city group is relatively sparse, and the distance between Wanzhou and Dazhou is less than 100km, but the high-speed commuting time between the two cities is 1.2 h (the maximum speed of Chinese expressways is 120 km/h).

Additionally, compared with expressways, high railways have a stronger radiating and driving effect on urban areas along the route in China. The connection of high railways between node cities will further strengthen the interconnection, personnel migration, and spatial integration of cities. There are eight high rail lines (two under construction) radiating from Chengdu and forming a 0.5-h high-speed rail metropolitan area. Seven high-speed rail lines have been extended from Chongqing, three of which are under construction, basically achieving a 0.66-h travel circle for high rail travel. The density of the high railway network in the SSCW city group is second only to the density of the Chengdu and Chongqing megalopolis. As a core city, Yibin is the intersection of the Chengdu-Zigong-Yibin high-speed railway, Chengdu-Guiyang high-speed railway, and Chongqing-Kunming high-speed railway, and it plays an important role as an important hub for the opening up of Chengdu and Chongqing to the south. We can achieve half an hour of high-speed railway commuting and reach the Chengdu and Chongqing metropolitan areas within 1 hour. In the CCC city group, Suining has a higher high-speed rail density, while Nanchong and Guang’an have a lower density. In particular, Guang’an is a blank area of the high-speed rail network, which is not conducive to the interconnection and coordinated development of the city group. In the WDK city group, Wanzhou and Dazhou have already opened high-speed rail to other cities, but the high-speed rail lines connecting them are still under construction. It can be seen from the pattern of the high-speed rail network that the cities in this city group have obvious administrative divisions.

Water transportation and civil aviation in the Chengdu-Chongqing area are important supplements to land transportation, especially the waterway, which covers national skeleton waterways (levels I and II). As a water transportation hub, Chongqing has extremely convenient river transportation, which is incomparable to Chengdu. Through the Yangtze River and its tributaries, the Minjiang River and the Jialing River, Chongqing can connect Leshan, Yibin, Luzhou, and Wanzhou from east to west, and Nanchong and Guangyuan from north to south (Figure 4C). On the urban agglomeration scale, civil aviation is one of the four transportation modes whose role has been weakened, especially in the field of passenger transportation. However, in special scenarios (natural disasters, traffic paralysis, etc.), air transport still plays a key role in passenger and cargo transport. Therefore, the number and level of airports in each city group are also very important for their connectivity. Specifically, Chengdu has five airports, followed by Chongqing with three airports, and each of the other prefecture-level node cities has one airport.

Economic growth and population agglomeration are the driving forces of new-type urbanization (Mahtta et al., 2019). Superior economic-demographic conditions can effectively guarantee the capital and labor needed for urban development. Therefore, by analyzing the spatial economic distribution and population changes of city groups in CCEC, we can effectively judge the feasibility of spatial integration of each city group. However, traditional statistical panel data can only reflect the average level of the regional economy (Zhu et al., 2022), and cannot represent the spatial differences in regional economic distribution. Evidence shows that the number of firms is positively correlated with the size of GDP (Erdin and Ozkaya, 2020; Gherghina et al., 2020). Consequently, we use the POI data of enterprise distribution in 2012 and 2022 to analyze the fishing net density to characterize the spatial distribution and changes in the regional economy. In 2012 (Figure 5), the distribution of CCEC enterprises mainly concentrated in Chengdu (D = 589/km²) and Chongqing (D = 560/km²), with the highest distribution density of 589/km², while the distribution density of enterprises in node cities was relatively low. Among the prefecture-level cities, Mianyang (D = 190/km²) and Yibin (D = 170/km²) rank high, but there is still a huge gap between Chengdu and Chongqing. The GDP scale (statistical data) of node cities in 2012 has a high consistency with the distribution density of enterprises. View of the POI’s distribution (Figure 5), in the SSCW city group Yibin-Luzhou, Zigong-Neijiang initially formed a network with a simple structure, and the other two city groups did not show this feature obviously. The distribution density of the CCEC’s enterprises has increased significantly in the past decade (Figure 5). The density of enterprises in prefecture-level cities has increased significantly, especially in Yibin (D = 300/km²), Mianyang (D = 330/km²), and Luzhou (D = 285/km²). Except for the two metropolitan areas, the SSCW city group has the largest distribution of enterprises, with a density of 185/km² and a GDP of 932.2 billion yuan. In the CCC and Wanda WDK city group, the number of enterprises has increased significantly compared with 2012, but it is still lower than that of the SCCW city group. Their GDPs are 566 billion and 423.9 billion respectively (Table 2). Generally, the urban belt from Yibin to Yuxi via Luzhou has a large number of enterprises and superior economic conditions. In contrast, the spatial distribution characteristics of enterprises in the other two urban agglomerations are not obvious.

Additionally, to explore the demographic elements of urbanization, we employed the Kriging interpolation method (Yadav, et al., 2018; Sharma, et al., 2021; Maurya, et al., 2023; Yadav, et al., 2023) to interpolate the population data (town level) based on the seventh census material (2020). Combining the locations of the two metropolitan areas and the three city groups, eight paths were selected (Figure 6A) to analyze the changing trend of the permanent population from Chengdu to Chongqing via different node cities (Figure 6) with a spatial perspective. It can be seen from the above that the 8 paths cover the main node cities of the CCEC. The analysis results show that Chengdu and Chongqing are the most populous regions in CCEC (Figure 6B), and the average population size of Chengdu and Chongqing is about 1.35 million/district and 1.15 million/district. As shown in Figures 5C–J, the average population size of the central district of Suining, Nanchong, and Guang’an in the CCC city group is 0.6 million—0.75 million (Figures 6C, D, E). In the WDK city group (Figure 6F), the population scale gradually increased from Dazhou to Wanzhou. The average population size of Yibin, Zigong, and Neijiang is similar, and the population size of Luzhou and West Chongqing is higher in the SSCW city group. The population changes from Chengdu to Deyang, Ziyang, and Meishan were 1.35 million per district to 0.8 million per district, 0.95 million per district, and 0.5 million per district, respectively. It can be seen that the population density in the north and southeast of the Chengdu metropolitan area is relatively high, while the population density in the south is relatively low.