Spatiotemporal Dynamics and Driving Factors of Urban Expansion in the Chongqing Metropolitan Area Based on Nighttime Light Remote Sensing

Abstract

This study investigated the spatiotemporal dynamics and driving mechanisms of urban expansion in the Chongqing Metropolitan Area by integrating multi-source big data and employing a suite of quantitative analytical methods. Drawing upon high-resolution remote sensing imagery, land-use datasets, socioeconomic statistics, and transportation network data spanning 2019 to 2023, the research revealed pronounced spatial and temporal heterogeneity in urban growth. Specifically, expansion manifested through a core-periphery spatial structure and temporal imbalances. The findings underscore a growing economic interconnectedness between core urban districts and peripheral cities such as Guang’an and Luzhou, giving rise to a multilayered and increasingly networked spatial-economic system. Moreover, urban expansion is shown to be tightly coupled with industrial distribution, transportation optimization, and regional integration strategies. In particular, the implementation of the Chengdu-Chongqing Twin-City Economic Circle has significantly facilitated cross-regional factor mobility and spatial restructuring, thereby accelerating coordinated development across the metropolitan area. Looking forward, urban expansion in the Chongqing Metropolitan Region is expected to continue leveraging transportation infrastructure and strategic industrial placement to advance regional economic integration.

1. Introduction

Urban expansion in metropolitan areas has become a defining feature of global urbanization, presenting both development opportunities and governance challenges. While major cities around the world have undergone extensive spatial restructuring, problems such as fragmented urban form, weak coordination between core and peripheral regions, and unbalanced growth remain unresolved [1]. These issues are particularly pronounced in inland metropolitan regions, where administrative boundaries, industrial capacities, and infrastructure systems often misalign with functional urban agglomerations [2].

The Chongqing Metropolitan Area, as the first trans-provincial metropolitan zone in western China, encompasses a central urban core, surrounding municipal districts, and Guang’an City in Sichuan Province [3]. With its vast geographical coverage, population size, and economic potential, the region represents a critical node within the national spatial strategy. However, it also faces several structural bottlenecks including insufficient agglomeration capacity in the core city, weak economic momentum in satellite towns, and limited regional integration [4]. These conditions underscore the need for a more holistic understanding of metropolitan dynamics and cross-boundary development models.

Despite the growing body of literature on metropolitan expansion, existing studies often rely on single-source data, focus narrowly on core urban zones, or overlook the functional heterogeneity and governance complexity of integrated metropolitan regions [5]. Moreover, the coupling mechanisms between spatial expansion and land-use function remain insufficiently explored [6]. To address these gaps, this study proposes a comprehensive analytical framework that integrates multi-source big data—such as nighttime light imagery, land-use remote sensing, POI functional classification, and socioeconomic statistics—to examine the spatiotemporal evolution and driving forces of urban expansion in the Chongqing Metropolitan Area [7].

This study contributes in three important ways. First, it bridges the gap between physical expansion and functional restructuring by identifying how spatial growth corresponds with land-use transformations and inter-city coordination [8]. Second, it applies a cross-regional lens to explore the developmental interplay between Chongqing and Guang’an, providing insights into strategic integration at the metropolitan scale [9]. Third, it proposes a replicable model for inland metropolitan governance in the context of national spatial strategies [10].

The period from 2019 to 2023 was selected for three key reasons. First, this timeframe corresponds to the operationalization phase of major national policies, including the Chengdu–Chongqing Twin-City Economic Circle [11], thereby offering a policy-relevant context. Second, comprehensive, high-resolution spatial and statistical datasets for this period are methodologically consistent and readily available [12], allowing for robust and replicable analysis. Third, the selected years are key milestones in urban infrastructure construction, industrial development, and regional collaboration efforts within the metropolitan area [13]. This temporal window thus provides both analytical rigor and strategic value for understanding the recent patterns and mechanisms of urban expansion, as shown in Figure 1.

2. Materials and Methods

2.1. Study Area

The study area encompasses the main metropolitan regions and adjacent districts of Chongqing Municipality. This includes the central urban districts (Yuzhong, Dadukou, Jiangbei, Shapingba, Jiulongpo, Nan’an, Beibei, Yubei, and Banan) as well as the extended urban districts (Fuling, Changshou, Jiangjin, Hechuan, Yongchuan, Nanchuan, Qijiang–Wansheng Economic Development Zone, Dazu, Bishan, Tongliang, Tongnan, and Rongchang). Additionally, it covers Guang’an City in Sichuan Province (Guang’an District, Qianfeng District, Yuechi County, Wusheng County, Linshui County, and Huaying City) as well as parts of the neighboring provinces including Suining, Ziyang, Neijiang, Luzhou, and Dazhou in Sichuan Province, and Zunyi in Guizhou Province. The total study area covers approximately 10,700 km² based on statistical data, as shown in Figure 2.

2.2. Methodology

2.2.1. Urban Expansion Intensity and Rate Indices

(1)

Urban Expansion Intensity Index (UEI)

To quantitatively assess the spatial dynamics of urban growth within the Chongqing Metropolitan Area, two widely used remote sensing-based indices were applied: the Urban Expansion Intensity Index (UEI) and the Urban Expansion Rate Index (URI). These metrics are commonly adopted in land-use change studies to capture both the scale and temporal speed of urban land expansion [14], particularly when analyzing large-scale or cross-regional systems.

The Urban Expansion Intensity Index (UEI) reflects the average annual rate of increase in urban construction land relative to the total area of the study region [15]. It is defined as:

$$\mathrm{U}\mathrm{E}\mathrm{I}={\displaystyle \frac{∆U}{\mathrm{A}\times ∆t}} \ast 100\%$$

where $∆U$ is the increase in urban construction land during the study period, $\mathrm{A}$ is the total area of the study region (e.g., metropolitan boundary), and $∆t$ is the number of years between the two observation points.

(2)

Urban Expansion Rate Index (URI)

The Urban Expansion Rate Index (URI), in contrast, measures the average annual relative growth rate of urban land based on its original area, and is given by [16]:

$$\mathrm{U}\mathrm{R}\mathrm{I}={\displaystyle \frac{∆U}{U_0\times ∆t}} \ast 100\%$$

where $U_0$ is the urban construction land area at the beginning of the study period, and other variables are as defined above.

While these formulas may appear similar, their conceptual bases differ. UEI captures the absolute expansion intensity relative to the entire study region, while URI normalizes the expansion relative to the original urban land size, thereby highlighting growth momentum in more urbanized areas.

These indicators serve as the foundational layer for further spatial heterogeneity analysis (Section 2.2.2) and are supported by extensive use in prior urban studies. Their values also help identify high-growth subregions where additional factors—such as transportation, economic clustering, or land policy—can be investigated in relation to urban morphology changes.

2.2.2. Nighttime Light Remote Sensing Metrics

(1)

Light Intensity Index (LII)

This index quantifies the overall radiance intensity of urban areas and is used to infer levels of economic activity and built-up density. It is computed as:

$$I={\displaystyle \frac{{\sum }_{i=1}^n{L}_i}{n}}$$

where $L_i$ is the digital number (brightness value) of the $i$, and $n$ is the total number of pixels in the region. High brightness values generally indicate denser urban construction and more intense economic activity [17].

(2)

Lit Area Index (LAI)

To measure the spatial extent of illuminated urban areas, the Lit Area Index calculates the total area of pixels with brightness values above a defined threshold $T$. The index is derived as:

$$LAI=N_T\times A_p$$

where $N_T$ is the number of pixels with $L_i\ge T$, and $A_p$ is the area of a single pixel. This metric provides a direct estimation of the horizontal expansion of urbanized zones. By comparing LAI values across different years, temporal changes in urban extent can be monitored [17].

2.2.3. POI-Based Spatial Functional Analysis

This study utilized point of interest (POI) data collected in June 2023 across the Chongqing Metropolitan Area. To ensure analytical reliability, a rigorous multi-step data preprocessing pipeline was implemented. Specifically, a POI entry was considered valid if it met the following criteria: (1) included complete geolocation (latitude/longitude) and semantic tags; (2) was not a duplicate (spatial and semantic duplication were removed based on string matching and spatial buffering); (3) did not belong to auxiliary or irrelevant categories (e.g., ATMs, streetlights, advertising structures, or temporary construction facilities). Entries failing to meet any of these conditions were excluded. After filtering, 842,139 valid POI records remained and were used for subsequent analysis.

For classification, POIs were systematically reclassified into six primary urban functional categories: (1) residential; (2) public administration and service; (3) commercial service; (4) industrial; (5) transportation and road facilities; and (6) green space and public squares.

This classification was not based solely on subjective public perception. Rather, it followed an objective, rule-based system adapted from existing frameworks in urban planning research. The classification principles considered: (1) standardized POI labels and tags from major platforms (e.g., Gaode, Baidu); (2) spatial context (e.g., proximity to residential clusters or industrial parks); and (3) function-specific keywords and hierarchical structure.

The technical workflow for functional POI classification and zoning is shown in Figure 3. It includes reclassification, weighted scoring, grid-based aggregation (1 km × 1 km), and functional type assignment based on proportional thresholds. The final functional category for each grid cell was defined using dominant-type rules (e.g., single-function if >50%, mixed-use otherwise) [18].

Figure 4 presents the distribution and proportion of different POI types. Among them, commercial service facilities accounted for the largest share, totaling 400,759 records (48%), while green spaces and public squares were the least represented, with 6422 POIs (0.76%). POIs were spatially clustered in Chongqing’s central urban districts and a few secondary growth centers such as Yongchuan and Qijiang. This clear core–periphery divergence highlights functional and economic centrality in the metropolitan core and lays a spatial foundation for the driving factor analysis of urban expansion patterns in Section 3.

2.3. Data Sources and Processing

This study integrated multi-source datasets spanning the years 2019 to 2023, selected to ensure consistency in spatial resolution and temporal comparability. All datasets were processed and harmonized at a 30-m resolution using Google Earth Engine (GEE) and ArcGIS Pro 3.2 [19].

(1)

Land-Use Remote Sensing (2019–2023)

Land-use classification was derived from two high-quality datasets: (1) China Land-Use/Cover Dataset (CLCD v2.0) with 30 m resolution, for 2019 and 2023, provided by the Chinese Academy of Sciences (https://www.webmap.cn/commres.do?method=globeIndex, accessed on 1 June 2024); and (2) ESA WorldCover Dataset: 10 m resolution, for 2020–2022, resampled to 30 m (https://esa-worldcover.org).

Land-use categories were reclassified into built-up and non-built-up classes based on the “Urban” or “Artificial Surface” labels. Urban areas were extracted by combining rule-based filtering and supervised visual interpretation using GEE [20].

(2)

DMSP/OLS Nighttime Light Data

NTL data from the Defense Meteorological Satellite Program’s Operational Linescan System (DMSP/OLS) and Suomi NPP’s Visible Infrared Imaging Radiometer Suite (VIIRS-DNB) were used to measure the nighttime radiance intensity [21]. Brightness thresholds were calibrated using histogram analysis, visual reference, and urban land masks to generate the Lit Area Index (LAI) and Light Intensity Index (LII) across years. All DMSP/OLS data were sourced from NOAA’s NGDC archive.

(3)

Landsat Satellite Imagery

Landsat 8 OLI and Landsat 9 OLI-2 imagery from 2019 to 2023 were retrieved from the USGS EarthExplorer platform (https://earthexplorer.usgs.gov, accessed on 1 June 2024). Though the Landsat program started in 1972, this study restricted its analysis to 2019–2023 to ensure alignment with other datasets.

(4)

POI Data

Point of interest data (842,139 entries) were collected and reclassified as described in Section 2.2.3. The dataset was cleaned, geocoded, and standardized for integration with grid-level analysis.

(5)

Administrative Boundary Data

Administrative divisions were obtained from the National Geomatics Center of China (https://www.ngcc.cn) with WGS-84 coordinates, ensuring spatial consistency across spatial overlays.

(6)

Socioeconomic Statistics

Annual data on the built-up land area and GDP were retrieved from (1) the China Urban Statistical Yearbook (2019–2023) and (2) National Bureau of Statistics of China (http://www.stats.gov.cn)

These data were used to correlate urban spatial expansion with socialeconomic development at the city and district levels.

3. Results

3.1. Spatiotemporal Characteristics of Urban Expansion in the Chongqing Metropolitan Area

3.1.1. Temporal Characteristics

The spatial expansion patterns reflect the strength and direction of economic linkages both within and between cities in the metropolitan area [22]. In Guang’an, brightened regions were primarily concentrated in Guang’an District, while Qianfeng District and Huaying City emerged as major growth areas. Guang’an District is dominated by green chemical industries, Qianfeng focuses on light textile manufacturing, and Huaying specializes in electronics and basalt fiber industries. In recent years, the Guang’an Economic Development Zone has emphasized green agricultural protection industries [23], while Guang’an District has focused on the smart manufacturing of power transmission and distribution equipment. Major industrial projects are under construction in both Qianfeng and Huaying. Relying on the industrial foundation of the Chengdu–Chongqing region, Guang’an is actively building itself into a supporting base for advanced manufacturing and a hub for specialized industries, thereby jointly establishing a cooperative manufacturing development zone with Chongqing [24].

Urban land expansion shows strong spatial and industrial correlation [25]. Among peripheral areas of the Chongqing Metropolitan Area, Luzhou, Suining, and Dazhou ranked as the top three cities in terms of increased brightness areas, each exceeding 20,000 km². A comparative analysis of the urban expansion data from 2019 to 2023 revealed that Luzhou has experienced significant growth in recent years, particularly Luxian County, which added 268.19 km² of brightness area, as shown in

Table 1. Urban area statistics in KM² (2019–2023).

Region Type	City/District	2019	Increase (2019–2023)	2023
Core Circle (Chongqing Metro Area)	Chongqing	3054.21	749.45	3803.66
	Yuzhong District	23.1	0.0	23.1
	Dadukou District	97.13	0.0	97.13
	Jiangbei District	219.98	−1.57	218.4
	Shapingba District	396.11	10.5	406.62
	Jiulongpo District	435.23	0.0	435.23
	Nan’an District	262.24	0.0	262.24
	Beibei District	399.0	133.61	532.62
	Yubei District	713.74	251.74	965.48
	Banan District	507.68	355.17	862.85
Tight Circle	Chongqing	3028.48	2120.5	5148.98
	Fuling District	348.6	217.88	566.48
	Qijiang District	274.58	156.71	431.29
	Dazu District	319.2	285.87	605.07
	Changshou District	318.41	142.54	460.95
	Jiangjin District	468.3	166.17	634.47
	Hechuan District	292.69	200.03	492.72
	Yongchuan District	306.86	411.08	717.94
	Nanchuan District	135.19	123.64	258.83
	Bishan District	320.25	252.79	573.04
	Tongliang District	244.39	163.8	408.19
Radiation Circle (Adjacent Cities)	Luzhou	806.03	726.51	1751.29
	Jiangyang District	210.88	120.82	331.7
	Naxi District	76.24	41.04	117.28
	Longmatan District	180.03	69.58	249.6
	Luxian County	118.32	268.2	386.51
	Hejiang County	81.6	86.16	167.76

.

3.1.2. Spatial Characteristics

Between 2019 and 2023, brightness areas—defined as pixels with nighttime light intensity values exceeding 158.68 nW·cm⁻²·sr⁻¹ in the SNPP-VIIRS dataset—within the Chongqing Metropolitan Area expanded by a total of 2869.95 km². Growth was predominantly concentrated around the periphery of the core circle, exhibiting a planar expansion pattern [26]. The most substantial increase occurred in the tight-integration region, with 2120.50 km² of newly expanded area—mainly in the western districts of Yongchuan, Banan, Dazu, and Bishan. Yongchuan District showed the largest expansion overall. Both Chongqing Municipality and Guang’an recorded brightness area increases exceeding 250 km².

As illustrated in Figure 5, the tight-integration regions of Chongqing experienced the most rapid expansion during the last two years, followed by the core circle. Compared with 2019, spatial linkages between Chongqing and Guang’an became significantly more integrated by 2023, with expansion trends primarily oriented westward and northwestward [27].

The spatial distribution of brightness growth aligns closely with regional economic linkages [28]. In the central urban core, expansion toward Luzhou follows the Yangtze River Economic Belt and leverages transportation infrastructure developed under the Chengdu–Chongqing Economic Circle strategy [29]. In neighboring regions, expressways and railways have supported growth in green finance, logistics, health services, and advanced manufacturing sectors such as automotive and equipment production [30]. These complementary developments have strengthened regional integration and facilitated coordinated metropolitan growth [31]. Yellow colour denotes brightness areas in 2019, brown colour denotes brightness areas in 2023, and overlapping regions represent changes between 2019 and 2023. The maps illustrate the westward and northwestward expansion patterns and the enhanced spatial connectivity between Chongqing and Guang’an.

3.2. Drivers of Urban Expansion in the Chongqing Metropolitan Area

3.2.1. Identification of Influencing Factors

The Chongqing Metropolitan Area contains a large number of areas with no POI data, followed by single-function zones (approx. 40% of total land area), while mixed-function zones account for about 50% [32]. Spatially, single-function and mixed-function zones were interspersed within the built-up urban core. A relatively high proportion of mixed land-use was observed in the built-up areas of both Chongqing’s core and Guang’an, whereas no-POI zones were predominantly found between the city center and its outer boundary as shown in Figure 6.

Single-use land for public administration and public service facilities accounts for the largest share, followed by industrial land, residential land, and transport infrastructure [33]. The number of single-function units is influenced not only by the volume of POI data, but also by the distribution characteristics of each POI type. For instance, although commercial service POIs make up nearly half of the total entries, their land area is relatively small due to their spatial concentration in high-mix areas. In contrast, public administration and service facilities typically occupy large, stand-alone pixels, resulting in a higher share of single-use land, as shown in Figure 7 and Figure 8 [34].

3.2.2. Analysis of Influencing Factors

(1)

Single-Function Regions

The spatial distribution of single-function regions in the Chongqing Metropolitan Area revealed clear functional clustering and differentiation patterns, as shown in Figure 9 [35]. Residential single-function regions, as shown in Figure 9a, were densely concentrated in the central urban core, particularly within the main districts of Yuzhong, Jiangbei, Shapingba, and Nan’an. These regions radiated outward along major transport corridors, indicating a strong dependence on accessibility and commuting convenience [36]. Peripheral counties, such as Tongliang, Rongchang, and Qijiang, exhibited more scattered residential clusters, often adjacent to township-level centers, reflecting localized urbanization driven by population aggregation and housing demand.

Single industrial functional region, as shown in Figure 9b displayed a more peripheral distribution pattern, with notable agglomerations in Liangjiang New Area, Banan, Yongchuan, and Dazu [37]. These regions aligned with the locations of industrial parks, logistics hubs, and manufacturing clusters planned under the Chongqing industrial restructuring strategy. Their spatial layout closely followed expressway and railway networks, highlighting the pivotal role of transportation accessibility in industrial site selection [38]. The concentration of industrial regions in the western metropolitan fringe suggests an ongoing shift of manufacturing functions from the urban core to suburban and regions, alleviating core area congestion while fostering regional industrial integration.

When considering the full classification of functional regions, as shown in Figure 9c, a clear monocentric-to-polycentric transition emerges [39]. The urban core is dominated by residential, commercial, and public service functions, forming a compact mixed-use center. Industrial and green space functions are more prevalent in the outer rings, creating functional buffers between high-intensity residential/commercial regions and ecologically sensitive areas. The spatial juxtaposition of road-related functional regions with industrial and logistics nodes further reinforces the integration of transport and economic activities. This distribution pattern underscores the influence of planning policies, transportation infrastructure, and land-use zoning on shaping the spatial structure of the metropolitan functional system [40].

(2)

Mixed-Function Regions

Mixed-function regions are defined as grid units where no single POI category exceeds 50% of the total. The top three POI types by proportion were used to assign the functional label. In total [41], 20 distinct mixed-function types were identified across the Chongqing Metropolitan Area. These regions are mainly distributed in the central urban districts and the core areas of adjacent built-up regions—primary concentrations of urban functionality [42]. Outside of central Chongqing, Guang’an in Sichuan Provice also features a relatively large extent of mixed-use land, as shown in Figure 10.

The spatial distribution of mixed-function zones in the Chongqing Metropolitan Area reveals a clear structural differentiation between the urban core and peripheral regions. Among the 20 identified types, residential + public service + commercial, residential + public service + industrial, and residential + industrial + commercial zones are the dominant types, jointly accounting for over half of all mixed-function grids. These types are predominantly concentrated in core metropolitan districts such as Yuzhong, Jiangbei, and Shapingba, where high-density residential quarters coexist with commercial facilities, public services, and industrial belts, reflecting the metropolitan core’s multifunctional and compact urban fabric.

By contrast, combinations such as green space + residential + commercial, green space + residential + road, and public service + green space + residential are more frequently observed in suburban new towns and ecological transition zones (e.g., Beibei, Banan, and Yongchuan). These configurations highlight the role of ecological buffers and green infrastructure in shaping peri-urban mixed-use landscapes, often supporting a balance between residential livability and service accessibility.

Furthermore, road-oriented combinations (e.g., industrial + commercial + road, residential + commercial + road, and public service + commercial + road) demonstrate the strong influence of transport corridors in structuring mixed-use development. These types are widely distributed along expressways, airport economic zones (Yubei), and arterial commercial corridors (Nanping, Shiqiaopu), underscoring the corridor-driven nature of metropolitan expansion [43].

Finally, public service + industrial + commercial and public service + industrial + residential types cluster around industrial parks and transitional belts, where public infrastructure plays a complementary role in mediating employment-housing interactions. This pattern indicates the importance of service facilities in supporting industrial workforce settlements and enhancing the functional resilience of production-oriented zones.

Collectively, these results suggest that the coexistence of residential, industrial, commercial, and public service uses constitutes the structural foundation of Chongqing’s metropolitan development, while green space and transport corridors act as key organizing elements for peri-urban and linear mixed-function patterns, as shown in

Table 2. Summary of mixed-function zone types in the Chongqing Metropolitan Area.

No.	Mixed-Function Type (Top 3 POIs)	Main Functional Composition (%)	Typical Distribution Area	Notable Characteristics
1	Green space + Industrial + Residential	35–45/25–30/20–25	Northern Yongchuan, Tongliang	Dense residential core with high accessibility to services and retail
2	Green space + Industrial + Commercial	30–40/25–30/20–25	Shapingba, Dadukou	Residential clusters adjacent to industrial plants and service nodes
3	Green space + Industrial + Road	30–35/30–35/20–25	Jiulongpo, Banan	Mixed manufacturing and retail near housing zones
4	Green space + Residential + Commercial	30–40/30–35/20–25	Nan’an, core industrial parks	Service facilities embedded within industrial-commercial complexes
5	Green space + Residential + Road	35–40/30–35/20–25	Wanzhou, Beibei	Green infrastructure integrated with housing and public services
6	Green space + Commercial + Road	30–40/30–35/20–25	Suburban residential clusters	Balanced ecological–Service–Living pattern
7	Industrial + Residential + Commercial	30–35/30–35/25–30	Traditional manufacturing	Mixed employment–Housing–Retail clusters supporting industrial workforce
8	Industrial + Residential + Road	30–35/30–35/25–30	River-based industrial corridors	Worker settlements aligned with industrial axes and road accessibility
9	Industrial + Commercial + Road	35–40/30–35/20–25	Logistics and wholesale corridors	Road-proximate industrial and trade complexes forming transport-oriented nodes
10	Public service + Green space + Industrial	30–35/30–35/20–25	Peripheral industrial districts	Service facilities integrated into industrial zones with ecological buffering
11	Public service + Green space + Residential	35–40/30–35/20–25	Suburban new towns	Public services embedded within ecological–Residential environments to enhance livability
12	Public service + Green space + Commercial	35–40/30–35/20–25	Recreational sub-centers	Green leisure areas supported by retail and service functions
13	Public service + Green space + Road	30–40/30–35/20–25	Roadside ecological corridors	Public facilities concentrated at intersections of green infrastructure and transport axes
14	Public service + Industrial + Residential	35–40/30–35/20–25	Transitional belts	Industrial–Residential coexistence structured by public service supply
15	Public service + Industrial + Commercial	35–40/30–35/20–25	Industrial parks	Service and administrative hubs facilitating industrial and trade activities
16	Public service + Industrial + Road	35–40/30–35/20–25	Highway-based industrial areas	Industrial service facilities clustered along major road networks
17	Public service + Residential + Commercial	30–35/30–35/25–30	Core metropolitan districts	Public services, housing, and retail jointly reinforcing compact urban cores
18	Public service + Residential + Road	30–35/30–35/25–30	Residential corridors	Service facilities supporting linear residential expansion along roads
19	Public service + Commercial + Road	30–35/30–35/25–30	Secondary urban centers	Retail and service functions concentrated along high-accessibility corridors
20	Residential + Commercial + Road	35–40/30–35/20–25	Central business corridors	Residential–Retail co-location forming dense linear mixed-use strips

[44].

3.3. Urban Functional Pattern Evolution and Spatial Correlation Analysis

Figure 11 illustrates the spatial clustering of urban functions based on Local Moran’s I (LISA) analysis of POI data. The results reveal a pronounced spatial heterogeneity, with statistically significant clusters distributed in a clear core–periphery gradient [1,45].

High–High clusters (in red) are densely concentrated in the metropolitan core, representing zones of strong functional agglomeration characterized by high POI density and diversity. These areas correspond to multifunctional urban centers where commercial, service, and residential functions are highly integrated. The persistence of High–High clusters in the urban core underscores the consolidation of centrality in functional development and the reinforcement of spatial polarization.

Low–Low clusters (in blue) are predominantly located in peripheral and suburban districts, forming contiguous zones of weak functional intensity. This pattern indicates the prevalence of mono-functional or underdeveloped spaces at the urban fringe, where limited functional diversity constrains spatial interaction. The spatial continuity of Low–Low clusters suggests that functional underdevelopment is not randomly distributed but is structurally embedded in peripheral urban space [46].

High–Low clusters (in orange) are primarily observed along the transitional belts between the central city and suburban zones. These areas indicate strong local contrasts, where highly active functional units are juxtaposed with low-intensity surroundings. This reflects uneven functional diffusion during the expansion process and highlights spatial discontinuities in the functional landscape. Conversely, Low–High clusters (in cyan) occur sporadically within high-intensity zones, suggesting local functional weaknesses embedded in otherwise strong clusters. These anomalies may be associated with fragmented development, land-use inertia, or localized renewal processes [47].

The predominance of “Not Significant” areas suggests that functional spatial autocorrelation is highly selective, confined to distinct clusters rather than being uniformly distributed. This pattern reflects the dual structure of functional evolution: the strengthening and persistence of core agglomerations, accompanied by the emergence of fragmented, transitional clusters at the urban periphery [48,49]. Such findings provide evidence of both the concentration and diffusion processes shaping urban functional reorganization [50].

4. Discussion

4.1. Industrial Specialization and Spatially Directed Urban Expansion

The observed spatial configurations in the Chongqing Metropolitan Area reveal that industrial specialization—particularly in high-tech manufacturing, advanced materials, and logistics—acts as a decisive driver of targeted urban expansion [3,24,37]. Rather than simply noting where growth has occurred, these patterns highlight how sector-specific comparative advantages shape the functional geography of metropolitan development [22,23]. Similar to the Ruhr area in Germany and the Pearl River Delta in China [5,28], specialized industrial clusters in Chongqing’s Qianfeng District and Huaying City have leveraged export-oriented production, supply chain integration, and skilled labor pools to catalyze urbanization in adjacent zones [29,30].

From a governance perspective, the challenge lies in steering this industrial momentum toward balanced growth [4,31]. In inner-core industrial belts, zoning reforms could prioritize the redevelopment of aging factory land into mixed-use innovation districts, integrating co-working spaces, light manufacturing, and residential amenities [46]. Peripheral clusters require a complementary infrastructure strategy—particularly in freight corridors and logistics parks—to ensure integration into metropolitan-scale transport and digital networks [38]. Future research could incorporate firm-level trade and productivity data to forecast the spatial ripple effects of industrial relocation [25], offering early-warning tools for planners before imbalances become entrenched.

4.2. Functional Diversification as a Marker of Metropolitan Maturity

The shift from single-function to mixed-function zones in Chongqing reflects a transition toward a more mature metropolitan morphology [45,51], echoing patterns seen in Tokyo’s Tama New Town redevelopment and Toronto’s waterfront renewal [48,49]. The increasing co-location of residential, commercial, and public services within 2–5 km transport corridors suggests that accessibility is now a critical determinant of land-use integration.

For the inner core—areas such as Yuzhong and Jiangbei—functional diversification must be carefully managed to avoid over-warming and the displacement of historical residential communities [46]. This could be utilize the mixed use of the space that priserves residential floor space while promoting street-level commercial activity, such as the examples in Barcelona’s superblock model [8]. In emerging sub-centers such as western Yongchuan and northern Banan, infrastructure investment should prioritize transit-oriented nodes, ensuring equitable access to both employment and amenities. Expanding the analysis to include socioeconomic indicators, such as income mix and housing affordability, would allow for a more nuanced understanding of whether functional diversification is translating into social equity and improved quality of life [43,44].

4.3. Spatial Clustering and the Reinforcement of Polycentric Development

The rise in spatial autocorrelation and the emergence of high–high functional clusters indicate a polycentric development trajectory similar to the Randstad in the Netherlands and the Kansai metropolitan region in Japan [27,40,50]. These secondary poles—western Yongchuan, northern Banan, and the Guang’an core—have the potential to alleviate pressure on Chongqing’s primary core, reduce commuting times, and foster localized economic ecosystems [9,42].

To maximize these benefits, governance should focus on connectivity-first investment: upgrading inter-node rapid transit links [29], harmonizing land-use regulations across jurisdictional boundaries [47], and ensuring digital infrastructure parity between the core and peripheral clusters. International experience from the Paris–Île-de-France region suggests that without synchronized governance, polycentric regions risk reinforcing spatial inequalities rather than reducing them. Incorporating spatial network modeling with environmental performance metrics (e.g., CO₂ emissions, urban heat island mitigation) could guide infrastructure prioritization toward nodes with the highest potential for sustainable growth [6,10].

4.4. Implications for New Challenges in Digital City Planning

The empirical findings from the Chongqing Metropolitan Area offer critical insights into how emerging challenges in digital city planning can be addressed through data-driven metropolitan governance [1,5,8]. First, the integration of multi-source geospatial big data—including nighttime light imagery, high-resolution land-use mapping, and POI-based functional classification—demonstrates the feasibility of constructing dynamic, fine-grained urban knowledge bases [17,18]. Such data infrastructures are essential for enabling the real-time monitoring of urban expansion and functional transformation, a capability increasingly demanded by cities navigating rapid structural change [7,8]. However, the study also revealed persistent data heterogeneity (e.g., resolution mismatches, temporal gaps) that can hinder cross-platform interoperability, indicating a need for standardized data governance frameworks in digital planning systems [19,20].

Second, the observed shift toward polycentric, mixed-function development underscores the importance of scenario-based spatial simulations in digital planning platforms [40,47]. The functional clustering patterns detected here suggest that future growth is likely to concentrate along high-accessibility corridors [48,49]. Embedding these empirical spatial correlations into urban digital twins could allow planners to test alternative land-use, transport, and industrial strategies before physical implementation, reducing the risks of path dependency or resource misallocation [27].

Third, the cross-boundary coordination within the Chengdu–Chongqing Twin-City Economic Circle illustrates both the opportunities and governance complexities of multi-jurisdictional digital planning [3,4,9,11]. While shared infrastructure and harmonized policies can unlock regional synergies, differences in technical capacity, regulatory environments, and planning priorities can create interoperability barriers [31]. Addressing these will require federated digital governance models—where data, algorithms, and decision-support tools are co-developed but locally adaptable—to maintain both system-wide cohesion and local flexibility [10,27].

Finally, the methodological framework developed in this study—linking industrial specialization, transportation accessibility, and functional integration—provides a possible path for other metropolitan regions facing similar growth pressures [22,24]. Embedding such models into decision-support dashboards could enhance predictive planning capabilities, enabling stakeholders to anticipate spatial transformations under different economic, environmental, and policy scenarios [29,30]. In the context of the New Challenges in Digital City Planning theme, this approach moves beyond descriptive mapping toward actionable, simulation-ready intelligence, bridging the gap between urban science and operational planning [8].

5. Conclusions

This study revealed that urban expansion in the Chongqing Metropolitan Area (CMA) from 2019 to 2023 is shaped primarily by three interlinked drivers—industrial specialization, transportation infrastructure, and functional land-use integration—each exerting distinct but complementary influences on spatial growth patterns. Industrially driven expansion, exemplified by Guang’an’s light textile clusters and Huaying’s electronics and basalt fiber industries, has been the most decisive force. Concentrated industrial layouts not only attract labor and stimulate housing demand, but also catalyze the development of supporting the commercial and service sectors. This direct coupling between industrial location and spatial growth underscores the role of strategic industrial policy in guiding orderly expansion and avoiding fragmented or inefficient land-use.

Transportation infrastructure has operated as a structural enabler of expansion, determining both the direction and intensity of growth corridors. High-speed rail, expressways, and intercity transit have reduced the travel times between the CMA core and peripheral cities, reinforcing polycentric linkages and accelerating the spread of mixed-function zones. Transport-oriented growth has facilitated the co-location of residential, commercial, and industrial functions while also integrating peripheral areas into the metropolitan labor and consumer markets. Well-planned connectivity has thus amplified the benefits of industrial clustering and extended the spatial reach of economic agglomeration.

Functional land-use integration within mixed-function zones represents the spatial manifestation of these economic and infrastructural drivers. The concentration of “residential–public–commercial” and “residential–public–industrial” grids in core and near-core areas reflects both market demand and deliberate planning to balance accessibility, service provision, and employment proximity. This synergy between land-use diversity and infrastructure accessibility enhances urban resilience, reduces commuting costs, and supports more sustainable forms of growth. Importantly, functional integration in the CMA is not merely an outcome of expansion, but a mechanism that channels and stabilizes growth pressures.

Regionally, the CMA’s collaborative development with neighboring jurisdictions—facilitated by the Chengdu–Chongqing Twin-City Economic Circle—has institutionalized these drivers into a coherent metropolitan strategy. Cross-boundary industrial coordination, harmonized land-use regulation, and shared infrastructure investments have improved factor mobility and reduced inter-jurisdictional competition.

By distilling the interplay between industrial specialization, transportation connectivity, and functional integration, this study advances a mechanism-based understanding of metropolitan expansion. Such clarity is critical for translating empirical patterns into targeted policy: industrial land allocations should be synchronized with transport upgrades; mixed-use zoning should prioritize transit-accessible nodes; and cross-jurisdictional governance should institutionalize spatial coordination. While the study’s temporal scope was limited to five years and relied partly on nighttime light data with known limitations, the identified mechanisms provide a robust foundation for future scenario-based modeling and policy formulation toward balanced, polycentric, and networked metropolitan systems.