Urban Land Expansion Dynamics and Drivers in Peri-Urban Areas of China: A Case of Xiaoshan District, Hangzhou Metropolis (1985–2020)

Abstract

China has seen a transition from impulsive industrial development-driven urban expansion to urban planning and government control since the start of the 21st century. Urban planning attempts to ensure the sustainable development of cities and the rational use of land resources. The dual-track urbanization process does, however, often result in a significant gap between planning and actual development. This paper contends that a comprehensive dual-track urbanization research perspective is better suited for investigating urban sprawl in peri-urban areas in post-reform market economy-active regions. Taking Xiaoshan District, the peri-urban area of Hangzhou, China, as a case study, this research utilizes land use raster data from 1985, 2000, 2010, and 2020 to identify the dynamic changes in urban construction land and examines the driving elements through the logistic regression model. Based on the results, the mechanisms of market-incented and government-controlled impacts on urban expansion are explained. This study aims to (1) analyze the spatial and temporal patterns of Xiaoshan’s urban expansion during 1985–2020; (2) compare the compositional characteristics of three types of urban growth (edge growth, infilling, and leapfrog development) in each phase of Xiaoshan; (3) identify the natural environment, accessibility factors, and socioeconomic aspects that promote Xiaoshan’s urban expansion in 2000–2020; (4) discuss Xiaoshan’s dual-track urbanization from four perspectives: regional development, private business, government control, and major events. This study advanced our knowledge of the driving mechanisms of urban land expansion in small towns, from peri-urban areas to metropolitan sub-centers, and, thus, has important implications for Chinese urban planning and the management of metropolitan suburbs, especially those located in the Yangtze River Delta.

1. Introduction

As a result of the reform and opening up, China has undergone an unprecedented urbanization development. The urban spatial has expanded dramatically, the real estate market has developed rapidly, the urban landscape has achieved enormous building advancements, and the related urban function types and industrial and social structures are continuously evolving. Nationwide, the urbanization rate in China rose from 18% in 1978 to 61% in 2018 [1]. As of November 2020, there were seven super-cities with a permanent resident population of more than 10 million in China’s urban areas and 14 megacities with a permanent resident population between 5 million and 10 million [2]. In this context, urbanization has long been one of the most important issues in China. Prior to 2000, after reform and opening-up, the rapid expansion of Chinese cities was mostly attributable to the rural industrialization under the upsurge of township enterprises [3], which was conceptualized as “bottom-up urbanization” [4]. In 1983, however, the “municipal control of counties” policy altered the dynamic equilibrium between urban centers and surrounding rural regions. Due to land and real estate markets, large cities began to expand [5]. Therefore, it is widely agreed that China’s urbanization is simultaneously propelled by two tracks [6,7,8,9]. The first track is government-led urbanization, focusing on urban infrastructure development through land expropriation and capitalization. The second track is informal urbanization initiated by local villagers who have lost their land to market development and urban expansion. Villagers in rural areas where agricultural land was confiscated by the state or in regions with strong regional industry spontaneously constructed multi-story housing and rented it out in rooms to migrant workers. After 2000, the dual-track urbanization approach contributed significantly to China’s remarkable economic growth, urban infrastructure development, and urban expansion. Consequently, a comprehensive dual-track urbanization research perspective is more appropriate for explaining urban expansion in post-reform areas with thriving market economies.

The impact and drivers of metropolitan spatial patterns in China from a dual-track urbanization perspective have been extensively studied [10,11,12,13,14]. Under the guidance of the national 2035 urban agglomeration long-term construction plan, studies on the urban growth patterns and spatial reconstruction mechanisms of China’s most rapidly developing regions, such as the Yangtze River Delta and the Pearl River Delta, have been strengthened in recent years [7,9,15]. Depending on China’s economic policies or administrative shifts, these studies focus on the following three periods: (1) 1985–2000, a period of “bottom-up urbanization” led by township enterprises and the government’s “land-centered urban development”; (2) 2000–2010, the planning and construction of large cities under the country’s 10th Five-Year Plan; (3) 2010–2020, when the Chinese government proposed to engage in new urbanization, shifting the focus of development to small towns. Small towns serve as a transition zone between urban and rural areas, generally located at the edge of built-up urban areas and extending to connect low-density rural [16,17]. The rapid expansion of urban land in small towns is typical of the Yangtze River Delta region (the sixth largest urban agglomeration in the world) [18]. Although the number of small towns is far more than major cities, they are the region where the collision of dual urbanization is most intense [19]. However, due to the difficulties of data collection, studies on these towns are often inadequate. As metropolitan cities spread, towns in peri-urban areas are confronted with increasingly severe issues. Urban sprawl has been defined as a low-density, car-dependent style of urban development [20,21,22,23]. Market regulation in peri-urban areas is weaker than in built-up urban cores, and the barriers to entry for factories are lower, making them susceptible to undesirable competition with each other driven by economic interests. This form of urban development is considered an inefficient urban growth model with many social and economic drawbacks, especially environmental safety concerns and rural semi-urbanization [20,21,24]. Therefore, a systematic study of typical peri-urban areas which explores the urban expansion process and driving factors from small towns to metropolitan suburbs through “rural urbanization”, to urban suburbs through systematic planning and development, is of great theoretical value and practical significance for the sustainable development of China’s suburbs in the context of economic transformation.

Through the review of the existing cases in the suburbs of large cities in developing countries, this paper discusses the key driving factors and their impacts on the expansion of urban suburbs. Among these, models are one of the tools commonly used to examine the process of urban expansion and its drivers [25,26,27]. Logistic regression models (LRM) can combine socioeconomic variables with geospatial variables to analyze spatially specific drivers and explain the impact of drivers [27,28,29,30]. For example, Abubakr and Biswajeet argue that population growth and increased housing demand are the primary causes of urban sprawl [31]. Muhammad Salem et al. believe that proximity to the urban center and main roads are the most influential factors for urban expansion in the Delhi’s suburbs [32]. Population and GDP, according to Dong et al., are the primary drivers for the initial growth of settlements in China’s counties, whereas economy and accessibility are the primary drivers for the incorporation of small towns into the peri-urban areas of major cities [33]. Migrants from rural and affordable rent in semi-urbanized rural settlements, according to Braimoh and Onishi, are the most significant causes for urban development in the Lagos urban suburb, Nigeria [34]. Even though the topics of the aforementioned research all pertain to the suburbs of major cities in developing countries, the results vary per instance. Therefore, this study applies LRM to identify and analyze the drivers of urban expansion in Xiaoshan District, the peri-urban area of Hangzhou metropolis. It explores and discusses them from four perspectives: regional development, private economy, government regulation, and major events. This paper investigates the development process in the Xiaoshan District of Hangzhou from the perspective of dual-track urbanization. Previous studies on Hangzhou’s urban expansion mainly focused on the impact of state-supported urbanization and the spatial form in the central metropolitan area. In contrast, little attention has been devoted to the spontaneous process of rural urbanization in Hangzhou’s peri-urban areas. This paper attempts to address the following questions: What are the driving factors that influence the spatial form and structure of urbanization in Xiaoshan District, Hangzhou? How do the two urbanization tracks, spontaneous and state-supported, contribute to the urbanization process in the Xiaoshan District of Hangzhou?

2. Materials

2.1. Study Area

Xiaoshan District, Hangzhou (Figure 1) is located in the north of Zhejiang Province, on the south bank of Qiantang River, in the Yangtze River Delta (the sixth largest city group in the world [35]). It is also the combined part of the Hangzhou Bay Economic Corridor, one of the most economically developed and modernized regions in China. By the end of 2020, there were 2,056,000 permanent residents in Xiaoshan District, with a total land area of 1420 km². In 2020, the regional GDP reached RMB 2011.62B, the per capita regional GDP is RMB 152.3M close to developed countries, and the urbanization rate has reached 80.9%. In 1988, the State Council designated Xiaoshan county as a coastal economic open zone. The goal of the market economy was further established, and the opening was widened. The GDP topped RMB 10B for the first time in 1994. Since 1991, Xiaoshan has been one of China’s ten most economically developed county-level cities (districts). Xiaoshan was transformed from a county seat into a city. Under the “integration into the major metropolis of Hangzhou” policy, Xiaoshan has bolstered its economic strength with an emphasis on industry. At the same time, Xiaoshan has switched its development strategy from “building a modern city of medium scale” to “becoming a sub-center of Hangzhou.” With the 2016 G20 Summit as a symbol, Xiaoshan has become the primary commercial area of Hangzhou, a regional international conference, and a vital exhibition center in China.

To quantify the urban expansion process of the Xiaoshan area as a suburb of Hangzhou during the past 35 years, this paper chooses the street/town (Figure 1) as the spatial unit of investigation. Chinese cities have three levels of administrative units: (1) city; (2) district; (3) subdistrict or township. The subdistrict is located in the urban area, whereas the majority hub of the township is rural. The township (subdistricts) is the fundamental administrative unit in China, the spatial unit for many policy formulations and implementation, and the smallest geographic unit for publishing census data to the public [36]. When provincial and municipal development policies are formulated, a few townships (subdistricts) are often prioritized as demonstration sites. Therefore, it is appropriate to use townships (subdistricts) as the basic spatial unit to describe the expansion dynamics of cities in China.

2.2. Data Processing

2.2.1. Satellite Images

For land use classification, this study used Landsat satellite remote sensing cloud-free data for deriving land use in Xiaoshan District. This paper selected four cloud-free images from 1985, 2000, 2010, and 2020 to explore the land use changes. Detailed information on the chosen Landsat images is presented in

Table 1. Remote sensing images.

System	Satellite	Path/Row	Date of Acquisition	Spatial Resolution
Landsat TM	Landsat-5	119/39	22 July 1985	30 m
Landsat ETM+	Landsat-7	119/39	4 May 2000	30 m
Landsat TM	Landsat-5	119/39	24 May 2010	30 m
Landsat OLI_TIRS	Landsat-8	119/39	19 May 2020	30 m

.

Image pre-processing, classification, and accuracy assessments were performed in ERDAS IMAGINE (Leica Geosystems Geospatial Imaging, LLC., Norcross, GA, USA) and ArcGIS (ESRI Inc., Redlands, CA, USA). The images were geo-aligned in accordance with the UTM system (WGS84 datum, Zone51 North). Land use survey maps from 1996 and 2005 (scales of 1:50,000 and 1:10,000) were used to aid in selecting geometric image adjustments and land use classifications. To quantify patterns of urban expansion, five main land use categories were identified (Figure 2): agricultural land, forest land, urban, water, and other land uses. Urban land is all non-vegetated areas dominated by manufactured surfaces (e.g., roads and buildings), including residential, commercial, industrial, and transportation [37]. Since industrial agglomeration has defined the growth of the northern part of Xiaoshan after the reform and opening up, most of the rural construction land has been utilized by private individuals to establish enterprises, indicating a semi-urbanized rural community. Therefore, this area’s land patches were considered with urban land attributes. Then, 300 ground points were randomly selected for field validation. Applying the kappa index to calculate the accuracy, the overall accuracy of each land cover map is above 0.85, which is consistent with the recommended value [38].

2.2.2. Demographic and Economic Data

As of 2018, several towns in Xiaoshan District were classified as self-managed industrial agglomerations (currently Qiantang New District). This study uses yearly government reports for Xiaoshan District (http://www.xiaoshan.gov.cn) (accessed on 15 December 2021) and Qiantang District (http://qt.hangzhou.gov.cn/) (accessed on 21 December 2021), as well as Hangzhou City Statistical Yearbook (http://www.hangzhou.gov.cn/col/col805867/index.html) (accessed on 3 December 2021) for the relevant population and economic statistics. The Chinese government’s population data comprise permanent population, household registration population, and foreign population (population with local temporary residence permit), among which the foreign population has a high mobility rate, which partially overlaps with the permanent population statistics. Xiaoshan has a relatively high proportion of foreign population data as a region of “bottom-up urbanization” driven by industrialization. In 2020, the resident population in Xiaoshan District was 2.016 million, of which the household population was 1.371 million; hence, the foreign population accounted for 32% of the total population. Therefore, the relationship between population density and urban expansion is calculated in this study using data on the resident population at the townships (subdistricts) level from the fifth, sixth, and seventh national population censuses (http://www.stats.gov.cn/ztjc/zdtjgz/zgrkpc/) (accessed on 8 February 2022) conducted in China in 2000, 2010, and 2020, respectively. Data on industrial structure include the annual value added of primary, secondary, and tertiary industries.

3. Methods

This study identifies the process and pattern of urban sprawl in Xiaoshan during 1985–2020, as well as its drivers between 2000 and 2020. We use (1) the average annual growth area of urban land and the average annual urban expansion rate to compare and contrast the difference in urban development rates between Xiaoshan as a peri-urban area and the main urban area of Hangzhou; (2) the urban growth type index to describe the essential characteristics of urban sprawl in Xiaoshan; (3) the drivers of geographical conditions, accessibility, population, and economic and industrial structure are also considered to measure the impact of different conditions on urban expansion in Xiaoshan.

3.1. Urban Land Expansion

To quantify the intensity of urban expansion, it is necessary to use methods to measure the relative intensity of expansion at various periods or locations [39]. The measures can be compared between cities or individual cities across regions or time periods. Therefore, this study uses two indicators to compare the urban land expansion in the main urban area of Hangzhou and Xiaoshan District from 1985 to 2020: the average annual growth area of urban land ($UI$) [40,41] and the average annual urban expansion rate ($UE$) [23,42,43].$UI$ is appropriate for expressing the expansion of the same area in different periods, while $UE$ is more suitable for comparing different regions. The indicators are defined as follows.

$$UI=\frac{A_{end}-A_{start}}{d}$$

$$UE=\frac{A_{end}-A_{start}}{A_{start}}\times \frac{1}{d}\times 100\%$$

$A_{start}$ and $A_{end}$ refer to the area of urban construction land at the beginning and end of the calculated period, respectively, and $d$ (in years in this paper) is the calculation period.

3.2. Types of Urban Growth

Forman classified urban growth from a typological perspective into three types in 1995: edge growth, infilling, and leapfrog development [44]. Leapfrog development is considered a typical form of urban sprawl. Xiaoshan, as a peri-urban area of Hangzhou, is currently the most actively urbanized area and suffers from the problem of urban sprawl. Therefore, this study adopts Xu et al.’s measure to quantitatively identify these three types of urban growth [45] that describe the differences in urban growth in various townships (subdistricts). The index is calculated as follows.

$$S=\frac{L_{com}}{P}$$

where $L_{com}$ denotes the length of the common boundary between the newly constructed land patch and the completed patch, and P represents the perimeter of this newly constructed land patch. The three urban growth types are defined by an index (S) that takes values from 0 to 1. When S = 0, the type is leapfrog development, 0 < S ≤ 0.5 is edge growth, and S > 0.5 type is infilling. In addition, this study defines the leapfrog growth rate LPR as an index to describe the intensity of leapfrog development in each street town of Xiaoshan.

$$LPR=\frac{A_{LP}}{A_{end}-A_{start}}\times 100\%$$

The higher the $LPR$, the higher the degree of leapfrog development of the townships (subdistricts). If the $LPR$ is more than 50%, then it indicates that the growth of this township (subdistrict) in the past 20 years has mainly been dominated by leapfrog growth.

3.3. Driving Factors

The drivers affecting urban expansion can be identified quantitatively by constructing a logistic regression model [46] and explaining the relationship between the factors (independent variable x) and the probability of urban expansion. In accordance with the requirements of the logistic regression model, the dependent variable y represents the change of land from non-urban to urban, and its value is set to 1; otherwise, it is 0. The corresponding measurement model is as follows.

$$ln\left(\frac{P}{1-P}\right)=a+b_1{x}_1+b_2{x}_2+\dots +b_n{x}_n$$

P denotes the probability of urban land use change; a denotes the constant term; $x_n$ represents all factors affecting the results; $b_n$ indicates the partial regression coefficient of the logistic regression model. Meanwhile, this study used the ROC approach to test the regression results with the actual urban land use as the state variable and the anticipated change probability as the detection variable. The model fit improves as the ROC curve value grows, and when the ROC curve value is more than 0.8, it indicates that the model fit is good [47]. Concerning scholars’ previous studies on major cities in the Yangtze River Delta, it can be summarized that the driving factors of land transformation into urban mainly include geographical conditions, accessibility population, and economic and industrial structure-related planning and policies [48,49,50]. Xiaoshan District is characterized by hills and plains with dense river networks. The effect of geographical conditions on urban expansion can be discussed from the perspectives of the slope, river network density, farmland density, etc. From the perspective of accessibility, the effect of location conditions may be investigated in terms of the distance of different levels of transportation roads from the old city and township centers. The impact of socioeconomic factors such as population and GDP on urban expansion varies by region and development stage [42]. Moreover, land use policies and related plans may influence urban sprawl [51].

4. Results

4.1. Comparison of the Urban Expansion of Xiaoshan District and the Main Urban Area of Hangzhou

The land use results of the image analysis for 1985, 2000, 2010, and 2020 are shown in Figure 2. Figure 3 compares the composition of the five land use categories in Xiaoshan and Hangzhou’s main urban areas from 1985 to 2020. In Xiaoshan, the proportion of urban land use increased from 6.56% to 21.03% in 35 years. The construction land in the main urban area of Hangzhou has expanded, and the rate of change has accelerated annually. The period from 1985 to 2000 was crucial for urban growth, with construction land undergoing the drastic transformation. After 2000, the expansion rate of construction land decreased, but the overall growth rate remained high. On the whole, urban land expansion in both Xiaoshan and Hangzhou’s main urban areas implies a decrease in agricultural land. In particular, between 2000 and 2010, the water area in the Xiaoshan decreased, but the agricultural land increased. This can be explained by the fact that, during this period, Xiaoshan implemented the last phase of a six-year reclamation program, converting 80 km² water area into agricultural land. The southern part of Xiaoshan is hilly with little open agricultural land, so the forest area, which is difficult to develop, is relatively stable, and the urbanization rate of the region is relatively low.

Table 2. UI (km²), UE (%) of Xiaoshan District and Hangzhou’s main urban area for three time periods from 1985 to 2020.

	1985–2000	2000–2010	2010–2020	Average
UI (km2)
Xiaoshan	2.19	8.42	8.58	5.80
Hangzhou’s main urban area	2.83	15.78	14.18	9.77
UE (%)
Xiaoshan	2.38	6.74	4.11	6.30
Hangzhou’s main urban area	1.20	5.66	3.25	4.13

displays the average annual growth area of urban land in Xiaoshan District, the main urban area of Hangzhou (UI), as well as the average annual urban expansion rate (UE) changes. Xiaoshan District experienced rapid urban expansion over 35 years, and the area of urban construction land grew from 91.98 km² to 294.82 km². In 1985, Xiaoshan was an independently developed county-level city, and its average annual growth area was the same as that of Hangzhou’s main urban area. However, it can be seen from Figure 3 that Xiaoshan’s urban expansion was relatively more scattered and disorderly during that period. During the period of 2000–2010, Xiaoshan District was transformed from an independently developed county-level city to a district under the jurisdiction of Hangzhou City, and the initial stage of UE amounted to 5.66%. After 2010, the urban construction of Hangzhou’s main urban area was completed, and a large number of national industrial parks and exhibition centers began to be planned for the peri-urban area. During the same period, industrial parks in Xiasha, Yuhang District, and Xiaoshan District in the suburbs of Hangzhou were built on a large scale [52]. After 2015, the Hangzhou Municipal Government unified the planning and construction of the venues for the 2016 Hangzhou G20 Summit and the 2022 Hangzhou Asian Games venues in the north of Xiaoshan District, which is along the Qiantang River. Therefore, the overall urban construction of Xiaoshan developed northward, and the urban sub-city across the Qiantang River from the main urban area of Hangzhou began to take shape. The trend of spatial expansion of urban construction land demonstrates that the urban land use structure mode of Hangzhou has experienced the single center structure of Hoyt sector mode and is transitioning towards the multiple nuclei model by Harris & Ullman [53]. In terms of urban space, the core of Hangzhou’s main urban area around the West Lake is the first to carry out the land urbanization process and has reached the equilibrium stage. The zone, which is located between the urban core and its fringes, went through a transformational phase during the first 25 years and is now in a stable phase since it has adopted the central area’s urbanization process. Xiaoshan District, which is located in the suburb of the city, is undergoing rapid urbanization and is currently in the transition stage.

4.2. Leapfrog Growth Urban Expansion

Figure 4 depicts the characteristics of the spatial pattern of urban growth in Xiaoshan from 1985 to 2020. In general, before 2010, Xiaoshan’s urban expansion was characterized by a monocentric pattern of outward expansion from the urban core; after 2010, the two towns of Guaoli and Linpu expanded rapidly under the influence of the policy of cultivating small cities in Zhejiang Province, forming the secondary core of Xiaoshan District. From 1985–2000, edge growth was the primary type of urban growth in Xiaoshan, accounting for 87.27% of all growth areas (

Table 3. The Statistics of urban growth types.

Period	Growth Type	Area (km2)	Percentage (%)	Rate of Industrial Land (%)
1985–2000	Infilling	3.14	7.10	16.85
	Edge growth	30.70	87.27	16.09
	Leapfrog	2.59	9.63	16.29
	All	36.43	100.00	16.17
2000–2020	Infilling	48.50	28.45	6.84
	Edge growth	84.97	49.84	32.29
	Leapfrog	37.00	21.71	80.31
	All	170.47	100.00	35.47

). In 1988, Xiaoshan was classified as a coastal economic development zone. With the rapid rise of the export-oriented economy, urban development jumped out of the small-peasant economic model and rapidly entered the track of industrialization and urbanization. In 1996, Xiaoshan divided its urban development into three parts: old city, new district, and development zone. Urban land growth in this stage mainly revolved around the expansion of the original town centers of Chengxiang, Beigan, and Xintang Subdistrict. At that stage, Xiaoshan’s township enterprises, such as the Sunan model, played a crucial role in the regional economy, significantly contributing to the rise of towns as local industrial centers [54]. However, due to the beginning of rural urbanization, the privatization of township enterprises had not yet been completed, and the urban construction land was scattered, so the cluster scale was not significant. During the 20 years from 2000 to 2020, leapfrog and edge growth types jointly contributed 71.55% of the urban growth. By this time, the privatization of Xiaoshan township enterprises was complete and growing rapidly, adding a new dimension to the rural urbanization process. Driven by the market, towns and villages set aside considerable land for industrial parks to attract industrial development to create jobs, tax revenues, and land rents [54]. In this context, after being reclaimed as farmland, extensive areas of mudflats in the northern towns of Xiaoshan were rapidly converted into industrial parks, and further urbanization was accomplished (Figure 5). In this stage, industrial land accounts for 80.31% of the leapfrog development of urban construction land.

4.3. Results of Logistic Regression Model

Since the statistical standards of the socioeconomic data of each subdistrict and township before and after 2000 are different, and the data before 2000 are difficult to obtain, this study analyzed the influencing factors of urban expansion in the Xiaoshan area from 2000 to 2020. Accessing the ArcGIS 10.8 platform, the 13 influencing factors screened were assigned to the corresponding regions to spatialize the urban expansion, rasterize the influencing factors and randomly extract 4000 valid data points evenly distributed in the study area into the logistic regression modeling. Due to the low statistical value, two elements, water network density and distance to the airport, were excluded from the model, and the total number of variables was reduced from 13 to 11 (Figure 6).

The data were imported into SPSS 26 software for logistic regression, and the regression equation was obtained (see

Table 4. Global logistic regression model for the probability of non-urban to urban land conversion.

Driving Factors	B	Standard Error	Wald	Significance	Exp(B)
$x_1$ slope	−0.01546	0.02880	0.288	0.59156	0.98466
$x_2$ density of agricultural land	−0.48702	0.29413	2.742	0.09776	0.61445
$x_3$ distance to inter-city highway	0.00001	0.00004	0.059	0.80795	1.00001
$x_4$ distance to local artery roads	−0.00064	0.00011	34.119	0.00003	0.99936
$x_5$ distance to major city centers	−0.00005	0.00001	15.413	0.00009	0.99996
$x_6$ distance to town/subdistrict centers	−0.00038	0.00006	45.242	0.00012	0.99962
$x_7$ value of population density change	0.00015	0.00016	0.844	0.35835	1.00015
$x_8$ Added value of GDP	−0.39291	0.47606	0.681	0.40918	0.67509
$x_9$ primary industrial increment	0.34223	0.48487	0.498	0.48030	1.40808
$x_{10}$ secondary industrial increment	0.39221	0.47566	0.680	0.40962	1.48025
$x_{11}$ tertiary industrial increment	0.41081	0.47534	0.747	0.38796	1.50803
Constants	0.24862	0.34610	0.516	0.47154	1.28226

for specific values).

$$\begin{array}{c}ln\left(\frac{P}{1-P}\right)=0.248621-0.015455x_1-0.487022x_2+0.00001x_3-0.000637x_4\\ -0.000045x_5-0.000377x_6+0.000147x_7-0.392907x_8\\ +0.342228x_9+0.392208x_{10}+0.410806x_{11}\end{array}$$

The urban expansion in Xiaoshan is affected by both internal and external influences. The ROC test value reaches 0.875, and the model fits well, indicating that the logistic regression model can explain the impact of urban expansion in the Xiaoshan district (Figure 7).

The slope affects the construction cost and the convenience of living. It is also evident from the data that the urbanization rate of the northern polder plains in Xiaoshan is significantly higher than that of the southern hilly areas. It is worth noting that the density of agricultural land in the southern part of Xiaoshan, although providing considerable land for urban development, has a large area of preserved agricultural land due to the arable land preservation policy, which also has a negative impact on land use conversion. Among the distance variables, the distance to the local artery roads has the most decisive negative impact on the probability of land conversion. These results suggest that urban development in the Xiaoshan area depends on road infrastructure development and that local roads are a more important determinant than highways. This may be due to the different sizes of the study area, where previous studies on models of large cities have shown a significant effect of highways on the expansion of large regional urban agglomerations. On the other hand, Xiaoshan District is smaller in scope. Similar to the research results of Shu et al. on the port towns of Taicang City in China, the urban expansion pattern of Xiaoshan District during the last 20 years has alternated between leapfrog and infill growth, which is a form of low-density urban sprawl [55]. Therefore, the relative impact of town centers and main roads is greater compared to the distance of old urban areas from the highway, which is similar to the findings of Luo, J et al. in the related research on Nanjing metropolitan suburb [48]. It also means that the development of Xiaoshan townships (subdistricts) has been centered around their respective centers and that “bottom-up urbanization” has significantly impacted the overall urban expansion of Xiaoshan. It is also worth noting that the distance to the airport is excluded because of its low significance. It can be explained by the fact that although Xiaoshan has established a new airport district based on Xiaoshan Airport and tried to promote the construction of the new city, this unified planning model has not worked. In terms of socioeconomic variables, the growth of secondary and tertiary industries has a greater impact on urban expansion. The change of the added value of the primary industry is negatively related to the transformation of urban and rural development, which is consistent with the conclusion of Ren et al. on the urban development of the Pearl River Delta [56].

5. Discussion

5.1. Regional Development Drives the Transformation of Xiaoshan from Towns to the Peri-Urban Area

There are two factors that drive regional development on urban expansion. The first is the ability to expand externally under the influence of regional economic growth. The other is the differentiation of cities and towns affected by topography, resources, and other factors in the expansion process. After the Yangtze River Delta was designated as an open region in 1985, it benefited from many preferential policies in the national development plan. It became one of the pioneers in developing an externally-oriented economy [57]. As a frontier region in China’s urbanization, Xiaoshan’s urban expansion was influenced by regional development and echoed the shift in regional economic policies. In the 1980s, relying on Shanghai’s technology and market power, cities in the Yangtze River Delta began to develop small and medium-sized private enterprises based on state-owned factories in the planned economy. In order to reduce production costs and attract surplus rural labor, these small enterprises tended to set up in small towns, and Xiaoshan’s development from 1985–2000 was based on this. Guali Town and Xintang Subdistrict flourished during that period, showing a spontaneous urbanization process of scattered development. After 2000, the Yangtze River Delta region improved its economic regionalization by deepening market reforms. Like Xiaoshan, which was transformed from a county-level city to become the peri-urban area of Hangzhou, many big cities in the Yangtze River Delta region annexed neighboring towns and counties and transformed them into new cities and industrial parks. In this context, each of the 27 townships (subdistricts) in the Xiaoshan District has developed a unique development strategy based on its unique geography and resource base. Town development in Xiaoshan has been incorporated into a stable, mutually supportive, interdependent, and dynamic system. The “Yangtze River Delta Regional Plan” released by the State Council in 2005 was based on the model of single-center development around Shanghai. By 2016, the “Yangtze River Delta City Cluster Development Plan” had become a multi-center regional development plan. These national and regional planning changes indirectly promoted Xiaoshan’s gradual integration into Hangzhou and completed the regional urban spatial reconstruction.

5.2. The Private Economy Influences the Pattern of Xiaoshan’s Urban Expansion

Small towns were never included in the socialist industrialization program until the 1980s. Following the 1980s, the government encouraged rural industrialization, which facilitated the growth of rural businesses. As more of the rural surplus population was absorbed by these enterprises, a plethora of regional economic activity sprung up around them. Rural enterprises have emerged as a key force in manufacturing in the coastal provinces, representing a dramatic industry shift from urban cores to the more expansive rural areas. The transfer of industries from big cities to small towns has promoted the rapid development of the private economy in the Xiaoshan area. Industrial villages grew in importance, rural households actively participated in industrial production, and de facto urbanization occurred in all these places. To facilitate the transportation of raw materials and products, private enterprises mainly selected the main transportation routes to distribute to industrial sites. However, since they were all small and medium-sized enterprises, factories built in the Xiaoshan area were scattered along the roads during this period. After 2000, Hangzhou’s municipal and Xiaoshan governments encouraged industries to locate in the suburbs. They built a national economic and technological development zone in Xiaoshan District, resulting in the conversion of a sizable portion of agricultural land into industrial land. Urban development in Xiaoshan was greatly aided (35.47%) by industry-oriented growth, which represented 32.29% and 80.31% of the total marginal and leapfrogging growth, respectively (Table 3). The government encouraged the construction of industrial development zones through the provision of industrial and public service land and preferential land grant policies, which generated low-density sprawl growth in these peripheral areas of Guaoli Town and Yipong Town. The distribution of industrial land was transformed from point to sheet as industrial scale and agglomeration expanded.

The dense semi-urban settlements are another manifestation of how the private economy has promoted urban sprawl. These semi-urban communities are the product of urban expansion and the visual expression of dual-track urbanization. On the national-led urbanization track, the traditional agricultural land was expropriated by the government and turned into industrial parks. Simultaneously, on the remaining land, villagers who lost land or were affected by industrialization constructed multi-story flats under their ownership that are available for migrant workers with affordable housing rent and various services. The two paths are interdependent. The former supplies the necessary funding and facilities for urbanization, while the latter serves as the pillars of urban social and economic growth by accommodating migratory labor. Because of the lack of systematic planning, the phenomena of semi-urbanized rural communities propelled by self-built flats are common in Xintang Subdistrict, Wenyan Subdistrict, and Gua Li Town, townships (subdistricts) with developed private enterprises. A large amount of agricultural land developed for industrial use is the basic space for the leapfrog development of rural areas in the Yangtze River Delta. Rural industrial land is progressively being consolidated into towns and industrial parks thanks to government planning, but there are still many rural industrial sites distributed across the villages in Xiaoshan. The diverse leapfrog development that happened in different townships (subdistricts) is related to the control methods and implementation efforts, as well as the capacity and management techniques of rural industrial land.

5.3. The Government Directly Influences the Urbanization Process of Xiaoshan through Policies and Regulations, Planning, and Infrastructure Construction

Government regulatory forces influence the direction and emphasis of regional growth through planning, significant infrastructure development, and regulations that have a direct effect on the urbanization process. Since the mid-1980s, the Chinese central government has forced local governments to implement a number of arable land protection policies and regulations [58]. However, local governments frequently amend their land use plans to reduce the area of agricultural land protection and increase the amount of land available for commercial growth. Several towns in the eastern part of Xiaoshan were initially farms that were gradually established based on the Qiantang River polder to advance agriculture. However, our review of land use planning (1997–2010 and 2006–2020) in Xiaoshan during the last two decades reveals that the quantity of primarily agricultural land has been drastically reduced. Between 2006 and 2020, a substantial amount of urban expansion happened beyond Xiaoshan’s land use planning limits. The leapfrog growth converted a substantial quantity of basic agricultural land into urban construction land (Figure 8).

In 1998, with the reform of China’s housing system, the real estate market grew dramatically. The growth of the real estate market transformed the property into a significant source of government income and supplied funds for the construction of urban infrastructure [59]. There were three peaks in the development of Xiaoshan’s transportation infrastructure. One was the construction begun in the 1990s, with the Xiaoshan segment of the first transit highway completed in 1995. The second was the construction and expansion of Hangzhou Xiaoshan International Airport at the turn of the 21st century; Hangzhou Xiaoshan International Airport had its inaugural flight at the end of 2000, completed the second phase of the airport in 2012, and the third phase of the airport expansion in 2022. The third is the construction of rail transportation that started around 2010, including the construction of Hangzhou South Railway Station and Hangzhou Metro Lines 1, 2, 5, and 7. These transportation constructions greatly contributed to the urbanization of Xiaoshan, altered its spatial structure after 2000, and accelerated Xiaoshan District into a sub-city of Hangzhou. Moreover, the rapid population and economic growth in Xiaoshan after 2000 led to the ineffectiveness of plans that suited the previous institutional environment. The changing development environment forced the government to revise its plans frequently (1993–2010, 1997–2010, and 2001–2020). For example, the construction of large-scale projects such as the Airport New City and a large exhibition center led to the development of Ningwei Town in northern Xiaoshan and the polder zone in eastern Xiaoshan, a national economic and technological development zone (Figure 2, converted a large amount of agricultural land into industrial land with low building density, and drove the urbanization of the surrounding towns. Since 2001, Hangzhou and Xiaoshan District governments have further relaxed development controls in their new master plans to accommodate projects that were inconsistent with earlier land use plans. As can be seen from Figure 8, in the Xiaoshan District land use plan revised from 2014 to 2020, industries in the eastern region that were constructed in a disorganized way in the early years would be dismantled, moved, and planned as industrial parks for centralized construction. However, the plan was not totally successful in the real world.

5.4. Major Events Accelerate Changes in Xiaoshan’s Urban Spatial Structure

In the context of globalization, major events are instruments for city governments to achieve strategic urban development objectives, and they may directly affect changes in the spatial structure of cities. With major urban events such as the 2016 Hangzhou G20 Summit and the 2022 Asian Games held in Xiaoshan District, Hangzhou has steadily transferred the focus of urban development from the main urban area to the Xiaoshan District. Major events are government-directed or led by significant investment, large audiences, and great impact. Therefore, the impact on Hangzhou’s whole urban spatial structure involves changes in urban functions, spatial expansion, urban agglomeration, industrial structure adjustment, social population distribution, traffic structure, and other aspects. Hangzhou took the G20 Summit as an opportunity to complete the construction of Qianjiang Century City. It is located north of Xiaoshan District, with a large exhibition center, extensive sports facilities, and large commercial plazas, forming a new urban functional area in Hangzhou and attracting population migration. The development of the G20 Summit promotes the development of the new functional area and the improvement of the functional layout of Xiaoshan. It directly promotes the expansion of Xiaoshan to the north, with Ningwei Subdistrict and Yinfeng Subdistrict completing the leapfrog transformation from rural to urban. Secondly, as increasingly more large-scale exhibitions are chosen to be held in the Hangzhou International Expo Center, Qianjiang Century City will continue to develop and maintain a high level of vitality. The region is the core of Xiaoshan’s northward growth, bolstering Hangzhou’s urban economic and social clustering and radiation functions. Furthermore, the Asian Games are no exception. This is the opportunity for Xiaoshan to accelerate the construction of Xiaoshan International Airport, Hangzhou South Railway Station, and other service support facilities. Moreover, as the primary gymnastics arena is located in GuaLi town, this is an excellent chance to enhance municipal infrastructure and public facilities construction. The significance of this for GuaLi town is not only limited to improving the infrastructure, but also to promoting the construction of GuaLi town as a small city pilot in Zhejiang Province. As major events have a specific time node, infrastructure and venues are often built quickly with the goal of being completed prior to the event. In the past, the growth of the main city of Hangzhou north of the Qiantang River was superior to that of the Xiaoshan area. Currently, the hosting of these two important events in Xiaoshan will link Xiaoshan’s growth with that of Hangzhou’s urban core. This combination will combine the city, traffic, landscape, and, most importantly, the city functions together. Xiaoshan will become the metropolitan sub-center of Hangzhou, propelling the balanced growth of the entire city and even the urban agglomeration around Hangzhou Bay.

6. Conclusions

Although many previous studies have been devoted to exploring urban expansion, this study focuses on the suburb of the city. It comprehensively outlines the urban expansion dynamics of Xiaoshan District, the peri-urban areas of Hangzhou metropolis in China, from a county to a sub-city, including its driving factors and spatial patterns. Moreover, it distinguishes three basic forms of urban growth. It was found that Xiaoshan’s urban growth from 2000–2020 was dominated by edge growth and accompanied by urban sprawl represented by leapfrog growth. Therefore, the study further analyzed the drivers of urban expansion in this period through a logistic regression model. The findings indicate that accessibility and socioeconomic status are the most influential variables in Xiaoshan’s urban growth. In general, the closer to town/sub-district center and local artery roads, the higher the GDP, and the more likely it is that an area will be urbanized. In addition, China’s agricultural land protection policy has had an essential impact on rapidly growing towns since its implementation, although there has been poor regulation. Our research did not differentiate the accessibility of natural conditions from socioeconomic circumstances as possible drivers. Since the assessment of social and economic circumstances, such as the GDP, whose measurement is a consequence of the process, it can be evaluated individually when statistics are available. Moreover, the full span of 20 years is selected for the analysis of data samples. While the results are desirable, relevant research suggests that the order of importance of drivers varies at different stages of town development. At the same time, further comparative studies can be carried out on individual towns within the limits of Hangzhou’s suburbs.

The expansion of Xiaoshan is a representative example and microcosm of China’s fast urbanization. Within just a few decades, its land use pattern and urban form have swiftly evolved into a contemporary metropolis with millions of inhabitants, propelled by a sophisticated private sector. Compared to the typical urban development method of “construction after planning,” the early development mode in Xiaoshan is marked by spontaneity, chaos, and multi-point distribution, resulting in the fragmentation of settlement form, the waste of land resources, and environmental contamination. Nevertheless, under the impact of multiple factors (such as the intrinsic necessities of significant economic growth, policy guidance, the adjustment of administrative divisions, the construction of major infrastructure, etc.), the region has developed into a healthy and active metropolitan area. This approach, which is also the “dual-track urbanization” discussed in this paper, is one of the characteristics and intrinsic mechanisms of rapid and even ultra-fast urbanization in China, and it demonstrates fairly distinctive traits in the global urbanization process. How this dual-track system can be implemented and its relevance from the standpoint of land usage, landscape pattern, stage division, and the underlying causes may be the subject and significance of this paper.

This paper expands our understanding of land growth in China’s metropolitan peri-urban towns, with significant implications for policy formation and other relevant studies. China has transitioned from a rapid urbanization period to a new phase focusing on developing small towns and cities. With China’s economy slowing down significantly, excessive reliance on land financing has increased the debt level of local governments. On the other hand, more migrant workers choose to stay in cities. In this context, urban policies must be able to meet new challenges and promote more sustainable urbanization models. Therefore, to ensure sustainable peri-urban development, it is essential to comprehend regional drivers’ spatial and temporal shifts and to focus on the significant impact of “spontaneous urbanization” on peri-urban sprawl to establish differentiated land management strategies. When considering the impact of regional development, private economy, and government control on urban sprawl, it is crucial to link “bottom-up” innovation and “top-down” management. In this way, the two tracks of urbanization can be more meaningfully integrated, and the development of the peri-urban area can be more stable and sustainable. Our findings will help practitioners in understanding the origins, driving factors, and processes of urbanization in China’s metropolitan suburbs, and will contribute to a broader discussion on sustainable urbanization strategies.