Next Article in Journal
How Does Cross-City Patient Mobility Impact the Spatial Equity of Healthcare in China?
Next Article in Special Issue
Quantifying the Driving Forces of Water Conservation Using Geodetector with Optimized Parameters: A Case Study of the Yiluo River Basin
Previous Article in Journal
Spatiotemporal Evolution and Driving Mechanisms of Non-Grain Production Rate of Planting Structure in Jilin Province from the Perspective of Grain Security
Previous Article in Special Issue
The Impact of Built Environment on Mixed Land Use: Evidence from Xi’an
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Land
Volume 14
Issue 2
10.3390/land14020213
land-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Spatial Reconstruction and Determinants of Industrial Land in China’s Urban Expansion: A Theoretical Framework

by
Dan Zhao
1,2,
Kewei Liu
1,2,
Jianwei Li
1,2,* and
Jiagang Zhai
1,2
1
College of Urban and Environmental Science, Northwest University, Xi’an 710127, China
2
Shaanxi Key Laboratory of Earth Surface System and Environmental Carrying Capacity, Northwest University, Xi’an 710127, China
*
Author to whom correspondence should be addressed.
Land 2025, 14(2), 213; https://doi.org/10.3390/land14020213
Submission received: 27 December 2024 / Revised: 17 January 2025 / Accepted: 20 January 2025 / Published: 21 January 2025
(This article belongs to the Special Issue Optimization of Land Space for High-Quality and Sustainable Development)
Browse Figures
Versions Notes

Abstract

:
The evolution of industrial land layout is a significant feature of changes in urban land use types during rapid urbanization. However, the theoretical studies investigating the evolution mechanism behind industrial land are scarce. This study developed a dynamic theoretical framework examining the relationship between government and market forces. Furthermore, this study examined the determinants of industrial land evolution in Xi’an from 1994 to 2019 by using a logistic regression model. The results showed that industrial land area increased from 55.02 km2 to 126.63 km2 from 1994 to 2013 but later rapidly reduced to 106.61 km2 in 2019. The spatial distribution of industrial land exhibited significant deindustrialization and suburbanization during park-orientated agglomeration and showed a shift from large-area small-patch to multi-center large-patch agglomeration. Factors such as development zone policies, land prices, agglomeration benefits, land supply, and transportation infrastructure are key determinants of suburbanization, while land prices and agglomeration benefits are key determinants of deindustrialization. This study reveals the mechanisms driving industrial land evolution and offers guidance in improving the rationality of urban land layout and achieving industry–city integration.

1. Introduction

Industrial land is vital in both the economic and social dimensions of urban development. The expansion and efficient utilization of industrial land can foster the healthy and sustainable growth of cities by increasing job opportunities, boosting economic growth, and improving urban competitiveness [1,2,3]. Proper planning can also help mitigate issues such as the NIMBY effect1 caused by industrial pollution, urban sprawl, job–housing separation, and traffic congestion caused by industrial activities [4,5,6,7,8]. On the contrary, in the absence of the effective management of industrial land, cities face the risk of becoming a dense, high-value residential dormitory, losing their vibrancy and global appeal [9,10].
Studies on the evolution of industrial land layout can be traced back to the mid-19th century. On the one hand, the large-scale production of modern industry has greatly stimulated expansion demand in industrial land, while the rapid migration of populations to suburban areas and the convenient transportation and abundant and inexpensive land supply in urban suburbs have promoted the suburbanization of industrial land in areas such as the San Francisco Bay Area [11], Chicago [5], Cleveland [12], Greater Vancouver [13], and Montreal [14]. This suburbanization has exhibited obvious spatial agglomeration characteristics [15,16], becoming an important symbol of urbanization in North America [17]. On the other hand, after the end of World War II, urban reconstruction and expansion generated housing demand and intensified the interference of industrial production in residents’ lives. Therefore, governments promoted the suburbanization of manufacturing in the city center [18]. At the same time, traditional manufacturing firms in the central urban area went bankrupt under the impact of modern industry [16] or relocated from the city center to the suburbs based on the principle of maximizing profits [12,19]. As a result, city centers began to show significant deindustrialization characteristics [16,20], and the international transfer of manufacturing against the background of globalization further intensified the deindustrialization of developed countries [21], causing traditional manufacturing cities to generally face many social problems such as economic decline and unemployment [7,22]. Therefore, protection policies for industrial land have been implemented in developed countries to promote the return of manufacturing [3]. As a whole, developed countries have long engaged in active discussions on the evolution of industrial land layout and how to promote the rational distribution of industrial land since the Industrial Revolution [23,24,25,26,27,28,29]. However, studies in developing countries have been limited [30], especially at the municipal level [31], which remains underexplored.
Industrial land is a pivotal land use category in contemporary urban China [30,32,33,34,35,36], with large cities playing a central role in its expansion due to resource advantages and the scale effect of rapid industrialization. Between 2002 and 2019, the industrial land area of nine national central cities, namely, Beijing, Tianjin, Shanghai, Guangzhou, Chongqing, Chengdu, Wuhan, Zhengzhou, and Xi’an, increased rapidly from 1139.01 km2 to 2007.86 km2, with an increase of 1.76 times. By 2019, these cities accounted for 14.49% of the total industrial land area across 679 cities in China [37].
This expansion of industrial land area in large cities has sparked academic discussions on industrial land policies [33,38,39], utilization efficiency [40,41], transfer prices [42,43,44], and renewal theories [45,46,47]. Studies have reported that the industrial land in cities such as Beijing [31], Shanghai [32], and Hangzhou [30,48] has rapidly shifted from urban centers to suburban areas, forming a polycentric agglomeration pattern. However, the emerging regions, especially the western regions, often receive insufficient attention [49]. Meanwhile, most studies have focused on case studies and theoretical explanations regarding industrial land transformation at the parcel scale, but the quantitative analysis using mathematical and statistical methods at the city scale is insufficient [50,51]. The evolution of industrial land layout in China’s major cities has resulted in numerous urban issues, including job–housing separation and traffic congestion [8,52]. Therefore, understanding the origin and transformation of industrial land is crucial in revealing the factors driving its evolution, in order to formulate planning and management policies that promote industry–city integration [48,53].
The effective management of industrial land is essential in the high-quality development of Chinese cities [54]. Territorial spatial planning must ensure the rational management of industrial land on a wider scale, promote its intensive and efficient use, and improve its economic and social benefits. The management of industrial land layout adjustments must be improved to ensure optimal and rational distribution. Therefore, in this paper, we aim to establish a comprehensive analytical framework for understanding the evolution mechanisms behind industrial land layout, providing insights for use in improving land policies in China and promoting high-quality industry–city integration.

2. Theoretical Framework

The research on the mechanism behind industrial land evolution originated from the exploration of manufacturing site selection through industrial location theory, and the core concept is an emphasis on the decisive role of market factors in guiding the layout of industrial land [55]. However, due to the lack of urban planning, the rapid expansion of industrial land not only brings about an increase in economic benefits, but also accelerates population agglomeration and urban land expansion, resulting in serious negative effects, such as the Great Smog of 1952. In order to reduce interference from industrial land in the lives of urban residents, the government began to control the layout of industrial land. For instance, the 1916 Zoning Resolution divided New York City into Residence Districts, Business Districts, and Unrestricted Districts and clearly defined and restricted locations for manufacturing. Subsequently, the role of governments in industrial land layout has increasingly grown [56,57].
In general, regarding the determinants of industrial land evolution, the interaction between government action and market forces has received considerable attention. First, market-driven factors such as land price, land supply, agglomeration benefits, transportation conditions, and neighborhood characteristics are often highlighted [5,10,12,14,15,16,17,58,59,60]. These factors are guided by neoclassical locational theory, which posits that industrial enterprises prioritize minimum production costs or maximum profits, thus influencing changes in the spatial layout of industrial land. Second, studies have generally emphasized the important role of government intervention in changes in the spatial layout of industrial land [6,9,23,24,54,61], and the development zone policy is considered to play a critical role in shaping the spatial evolution of industrial land in China [30,32,62]. Despite research on the factors influencing industrial land evolution in both developed and developing cities, due to the lack of an analytical framework, a gap in understanding exists regarding the dynamic mechanisms underlying the evolution of industrial land layout during urban expansion [21,50,54]. How the spatial layout of industrial land changes over time and space, especially under the influence of government forces and market factors, has not been clearly defined.
Local governments in China have played a dual role as city managers and regulators, particularly in the context of institutional reforms since the reform and opening period. As city managers, local governments demonstrate a significant “entrepreneurial spirit”, utilizing market logic to strategically influence the location of manufacturing industries. According to rent theory, industrial land rent is significantly lower than residential and commercial land rents. Faced with intensely fierce inter-city competition and performance evaluation, local officials have aimed to maximize their own benefits by reducing industrial land prices to attract industry and raising the land prices of residential and commercial areas since the 1994 centralized fiscal reform [30,38,42,43]. Consequently, local governments create development zones in suburban areas, offering cheap industrial land with a good infrastructure (such as wide and convenient roads, sufficient power, and a gas supply) and preferential fiscal and tax policies (such as tax relief, fiscal subsidies, and financial incentives) to attract manufacturers [8,54,62]. Meanwhile, local governments arrange residential land and commercial land in the city center to maximize land transfer revenue.
Second, local governments, as city regulators, must also address the conflicts arising between manufacturing activities in development zones and nearby residential and commercial areas, particularly following urban expansion. Additionally, they need to address issues such as land expansion, idle land, and inefficient land use resulting from industrial decay or bankruptcy [35], while extensive industrial land utilization appears to be the primary driver behind the aforementioned conflicts [41,51]. To overcome these challenges, local governments promote the transformation of inefficient industrial land and establish new development zones in suburban areas, encouraging the relocation of polluting or new manufacturing industries to suburban areas to solve urban problems [8].
Market factors such as land prices, land supply, agglomeration economies, and accessibility significantly influence the evolution of industrial land. On examining manufacturing from the cost–benefit perspective, we found that for development zones located in the suburbs, low land prices, sufficient land supply, good transportation infrastructure, and agglomeration economy can reduce production costs. On the contrary, the rapid expansion of cities has led to rising land prices, insufficient land supply, negative externality (such as NIMBY conflicts), and the lack of agglomeration economy, increasing the production costs of manufacturing in development zones due to urban expansion [48,62]. As a result, industrial enterprises aim to relocate to the suburbs to reduce production costs, and local governments aim to capitalize on this appreciation by establishing new development zones in suburbs, encouraging industries to relocate to new development zones.
In general, the dynamic interaction between local governments and markets during urban expansion profoundly affects industrial land evolution. Accordingly, we developed a dynamic analytical framework (Figure 1). Based on the theoretical framework, using Xi’an as an example, we established industrial land conversion models by incorporating and quantifying the factors of government (development zone policies) and markets (such as land price, land supply, traffic accessibility, agglomeration benefits, and NIMBY effects) and discussed the intensity and relationship of government and markets with regard to the evolution of industrial land layout during different periods of urban expansion.

3. Study Area, Data, and Methodology

3.1. Study Area

Xi’an is the ninth-largest national central city in China. It is located in the central part of the Guan-zhong Plain in the geographical center of China. Xi’an serves as an ideal case study for industrial land evolution because it is a typical mega-city with a long history of industrial development. It is among the eight old industrial bases formed after the People’s Republic of China was founded (1949). As an industrial base, Xi’an has evolved into a crucial hub for national aerospace, modern manufacturing, and scientific research and education. By 2019, Xi’an had eight national and seven provincial development zones, generating an industrial output value of CNY 620.87 billion and hosting a population of 10.20 million [63], making it a key industrial area in western China.
The study focused on a total area of 2073.69 km2 that includes nine districts, namely, Weiyang, Xincheng, Beilin, Lianhu, Baqiao, Yanta, Chang’an, Huyi, and Gaoling, and various development zones. These zones include economic and technological development zones, high-tech industrial development zones, and provincial and municipal industrial parks (hereinafter collectively referred to as development zones). This region displays the most intense changes in industrial land layout. To compare and demonstrate the spatial evolution trend in urban industrial land, based on existing studies [52], roads, including the first, second, and third ring roads, were used as boundaries to divide the study area into four zones: Zone I, Zone II, Zone III, and Zone IV (Figure 2). Among them, zones I, II, and III comprise the central city, and Zone IV is the suburban area.

3.2. Data Sources

Land use data were primarily obtained from Google Earth satellite images, Quickbird remote sensing images, Xi’an urban cadastral database, and Xi’an City Master Plan [64]. Benchmark land prices were gathered from the China Land Price Information Service platform (http://www.landvalue.com.cn/, accessed on 2 January 2022.) and Xi’an City Master Plan. A research team ensured the consistency and accuracy of satellite image data through visual interpretation, referencing master plans for 1995, 2008, and 2017 produced by Xi’an Urban Planning and Design Institute; the Xi’an life atlas; and the master plans of districts, counties, and development zones. Finally, the visually interpreted industrial land results were further calibrated through field surveys and interviews.

3.3. Research Method

The logistic regression model is an effective method for analyzing the driving frces behind urban expansion and land use change, particularly industrial land evolution [30,32,42]. In this study, the dependent variable (Y) is binary, representing whether industrial land has changed or not. Specifically, Y = 1 indicates that industrial land has been newly added or transformed, while Y = 0 means no change in industrial land use. The functional expression of the logistic regression model is as follows:
l o g   i t Y = β 0 + ∑ i = 1 n β i x i + e
where Y is the dependent variable, indicating the probability of industrial land evolution when the independent variable takes Xi; Xi is the explanatory variable; and βi is the regression coefficient value. A linear expression can be obtained by changing log to (Y), as follows:
P Y = 1 = exp ⁡ β ^ 0 + ∑ i = 1 n β ^ i x i / 1 + e x p β ^ 0 + ∑ i = 1 n β ^ i x i
where β ^ i is the parameter estimate that indicates the degree of contribution of the explanatory variables to industrial land evolution. A positive coefficient indicates that the probability of industrial land evolution increases with an increase in the explanatory variables, and vice versa.

3.4. Variables and Data

3.4.1. Dependent Variables and Data

The evolution of industrial land layout includes newly expanded industrial land and the transformation of existing industrial land. It is, therefore, modeled using two sub-models: a Gain Model and a Loss Model. In the Gain Model, Industrial Land Growth (IIG) = 1 indicates that non-industrial land is converted to industrial land during T1–T2 (T1 < T2) time, whereas IIG = 0 denotes that non-industrial land is not converted to industrial land during T1–T2 time. In the Loss Model, Industrial Land Loss (IIL) = 1 indicates that industrial land is converted to non-industrial land during T1–T2 time, whereas IIL = 0 represents that industrial land is not converted to non-industrial land during T1–T2 time. Finally, the dependent variable data for both models were determined by performing a two-by-two overlay analysis of industrial land in Xi’an for different time periods (Figure 3).

3.4.2. Explanatory Variables and Data

In the logistic regression model for analyzing industrial land evolution, various factors that represent government actions and market factors were selected as independent variables (Table 1). First, the dummy variable of development zone policy (DZ) captures the role of local government interventions in shaping the spatial distribution and transformation of industrial land. Second, the explanatory variables of land price, accessibility, and three neighborhood indices were selected to represent the impact of market factors on industrial land evolution. Of them, the explanatory variable of accessibility included the distance from the railway freight station (RAI), the distance from the entrance/exit of the expressway (EXP), and the distance from the main road (MAI). The explanatory variable of land price is represented by the industrial benchmark land price (BLP) in the Gain Model. In the Loss Model, two variables are used to measure the potential profitability of converting industrial land into non-industrial use: the difference in benchmark land price between industrial land used at time T1 and residential land used at time T2 (PVIR), and the difference in benchmark land price between industrial land used at time T1 and commercial land used at time T2 (PVIB). The three neighborhood indices included the percentage of the area occupied by existing industrial land (DEN), which represents the agglomeration benefits, the percentage of potential land around existing industrial areas that could be converted for industrial use (POT), which represents the land supply, and the percentage of residential land surrounding existing industrial areas (RES), which represents the NIMBY effect of industrial land.
All analyses were conducted using a square of 100 m × 100 m as the unit of observation. Therefore, dependent variables and explanatory variables were converted to a raster-based format, which might result in spatial dependence in the Gain Model and in the Loss Model [30]. To reduce the spatial autocorrelation between variables, the dataset of variables was determined by using the spatial sampling method combining regular sampling and random sampling. Among them, 300 samples were selected in the Gain Model and in the Loss Model from 1994 to 2007, totaling 600 samples; 500 samples were selected in the Gain Model and in the Loss Model from 2007 to 2013 and from 2013 to 2019, totaling 1000 samples.

4. Results

4.1. Evolution Characteristics of Industrial Land

The evolution of industrial land demonstrated the substantial and sustained suburbanization and deindustrialization of urban centers as the city underwent rapid urban expansion and industrial transformation from 1994 to 2019 (Table 2, Figure 4), which confirms our analytical framework. That is, the suburbanization and deindustrialization of urban centers are two important phenomena in urban expansion, and they jointly shape the spatial and temporal patterns in industrial land in Xi’an. In the suburban areas, the industrial land area in Zone IV continued to increase by 69.75 km2 from 1994 to 2019. This growth highlights the trend of park-orientated suburban industrial agglomeration, with much of the newly added industrial land concentrated in the suburban development zones. By 2019, the National Development Zones in the study area accounted for 61.17% of the total industrial land, especially in Xi’an Hi-tech Industrial Development Zone and Shaanxi Gaoling Jinghe Industrial Park, revealing the important role of development zone policies in guiding the evolution of industrial land. However, in the central city (Zones I–III), industrial land decreased significantly; in particular, the proportion of industrial land decreased by 56.53% from 1994 to 2019. Concretely, the industrial land area in Zones I–II continued to decrease, even directly reaching 0 in Zone I in 2019.
The expansion of industrial land in suburban areas was predominantly due to the conversion of agricultural land, which accounted for 91.69% of the total industrial land from 1994 to 2019 (Figure 5). Renewal in the central city primarily involved the conversion of land for residential, commercial, and business facility purposes, accounting for 51.65% of the total industrial land. Only a small amount of industrial land in the central city was used for logistics and warehouses, administrative and public services, and green space and square lands. Most industrial land in Zones I–II was transformed into residential land, whereas that in Zone III was primarily transformed into residential land and commercial and business facilities (Figure 6). As a whole, the spatio-temporal pattern of industrial land evolution in Xi’an is consistent with the bid–rent theory. The bid–rent capabilities of commercial and residential land are stronger than those of industrial land, resulting in a prominent “center–periphery” structure in their spatial distribution. This distribution pattern not only reflects the payment abilities and location preferences of different land users but also embodies the laws and trends of urban economic development.

4.2. Major Determinants of Industrial Land Conversion

To avoid the multi-collinearity problem, the correlation coefficient > 0.8 was used as a threshold to exclude the explanatory variable percentage of potential land around existing industrial areas that could be converted for industrial use (POT) (1994–2007) in the Gain Model, and the explanatory variables difference in benchmark land price between industrial land used at time T1 and commercial land used at time T2(PVIB) (1994–2007), and the percentage of residential land surrounding existing industrial areas (RES) and the difference in benchmark land price between industrial land used at time T1 and residential land used at time T2 (PVIR) (2007–2013) in the Loss Model (Supplementary Tables S1–S8).
The logistic regression model indicated that the development zone policy, land prices, agglomeration benefits, land supply, and transportation infrastructure are the key determinants of industrial land layout evolution in Xi’an from 1994 to 2019 (Table 3, Table 4 and Table 5). First, the explanatory variable development zone policy (DZ) was significantly associated with the dependent variable in the Gain Model since the national high-tech zone, Xi’an Hi-tech Industries Development Zone, was established in 1991. This indicated that the development zone policy (DZ) is a major factor influencing the spatial distribution of newly added industrial land. Second, the explanatory variables of industrial benchmark land price (BLP), difference in benchmark land price between industrial land used at time T1 and residential land used at time T2 (PVIR), and difference in benchmark land price between industrial land used at time T1 and commercial land used at time T2 (PVIB), which denote land price, were significantly associated with the dependent variable in the Gain and Loss Models. This indicated that land prices drive the evolution of industrial land layout. Third, the explanatory variables of distance to the main road (MAI) and distance to the entrance/exit of the expressway (EXP) were significantly associated with the explanatory variables in the Gain Model in different periods. This showed that industrial land layout evolution is closely related to changes in transportation infrastructure. Finally, the explanatory variable of existing industrial land (DEN) was significantly associated with the dependent variable in the Gain and Loss Models, indicating that the agglomeration benefits were closely related to industrial land layout evolution. The explanatory variable of the percentage of potential land around existing industrial areas that could be converted for industrial use (POT) was significantly associated with the dependent variable in the Gain Models during 2007–2013 and 2013–2019, indicating that land supply was an important reason behind industrial suburbanization.

5. Discussion

The results of the logistic regression models support our analytical framework, showing that the government and the market jointly shape the evolution of industrial land in Xi’an during urban expansion, and the government plays a more important role than the market [36].

5.1. Government Forces

First, the development zone policy (DZ) is a key driver of the suburbanization of industrial land. By 2019, the study area included six national-level development zones and four provincial-level development zones, primarily concentrated in Zone IV (suburban areas). These development zones offer ample industrial land at favorable prices and provide fiscal incentives for industrial enterprises, facilitating the expansion of industrial enterprises while reducing their operating costs. This combination of land availability and financial support encouraged the suburban clustering of industrial enterprises. The regression coefficient β of the explanatory variable development zone policy (DZ) was positive during 1994–2007 and 2013–2019, which is consistent with results for Shanghai [32] and Hangzhou [30,48]. However, the coefficient was negative during 2007–2013 because of the rapid expansion of development zones in the 1990s, which led to urban sprawl and the inefficient use of land resources. To address these issues, national policies intervened in 2004, resulting in the cancelation of 42 development zones in Xi’an, which accounted for 70% of the total number of original zones. Consequently, many small-scale industrial enterprises, which did not meet the stringent access requirements for development zones, began rapidly clustering in rural suburban areas outside the designated zones. This led to a faster expansion of industrial land outside the development zones compared to inside, which explains the negative β during this period. This reveals the spatio-temporal heterogeneity in the guiding influence of development zone policy on industrial land [28].
In addition, the explanatory variable of development zone policy (DZ) was not significantly associated with the dependent variable in the Loss Model, which is contrary to the results for Hangzhou [30] and Shanghai [32]. This is because, in Xi’an, industrial land use in the development zone was constantly changing during rapid expansion on the urban scale. For instance, 70 of the 144 plots had changed in use in Phase I of the Xi’an hi-tech Industrial Development Zone from 1991 to 2006 [65]. They predominantly changed from industrial land to residential and commercial land. In Phase II, 65 of the 120 industrial plots also changed in use, accounting for 54% of the total industrial land. According to the current Xi’an Urban Master Plan (2008–2020) (revised in 2017) [64], a large amount of existing industrial land in the central city (Zones I–III) and development zone, especially in the Xi’an hi-tech Industrial Development Zone and Xi’an Economic and Technological Development Zone, is also encountering the challenge of changes in use (Figure 7).
The results of the logistic regression model support our viewpoint that the development zone policy (DZ) is the core driving force in suburbanization but not in deindustrialization in urban centers. The model also reveals that the current development zone policies in China face severe challenges in promoting healthy and sustainable development; specifically, development zone policies cannot effectively protect the rational layout of industrial land. Existing development zone policies emphasize the economical and intensive use of industrial land, with a focus on indicators such as the comprehensive plot ratio and energy consumption per unit output value of industrial land [41]. However, little attention has been paid to the impact of the transformation of industrial land in development zones on the healthy and sustainable development of cities [28]. In fact, although developed countries have implemented many industrial land protection policies, there is still no effective mechanism to protect industrial land [66], and the shift by governments towards prioritizing the highest and best return principles in industrial land layout has led to the continuous deindustrialization of industrial land in urban centers. Therefore, the government still faces challenges in balancing the rational layout and efficient utilization of industrial land [24,25,26,39].

5.2. Market Forces

First, land prices are a crucial determinant in industrial land suburbanization and the deindustrialization of urban centers. In the Gain Model, the explanatory variable of industrial benchmark land price (BLP) was significantly associated with the independent variable during 1994–2007, 2007–2013, and 2013–2019, which is consistent with the results for Beijing [42], Shanghai [32], Hangzhou [30], and Cleveland [12]. This indicates that land prices guide the agglomeration of manufacturing units to suburbanization. In the Loss Model, the explanatory variables of difference in benchmark land price between industrial land used at time T1 and residential land used at time T2 (PVIR) and difference in benchmark land price between industrial land used at time T1 and commercial land used at time T2 (PVIB), which denote land value increment, were significantly associated with the independent variables in different periods. This result indicated that land prices are a major cause of urban center deindustrialization [45,48,62]. Although deindustrialization in Xi’an occurred for various reasons, the market-orientated reform of the land system generally increased the price differences among industrial, commercial, and residential land. This increased the land transfer fee for converting industrial land to commercial or residential land, thereby increasing the chances for the local government to help relieve state-owned manufacturing units in former industrial areas and improve living conditions in urban centers (Zones I–II) [52]. These positive changes are due to the city government converting industrial land to commercial or residential land in development zones.
The significance of three land price variables in both the Gain Model and the Loss Model confirms the key role of land prices in the processes of the suburbanization and decentralization of industrial land in the analytical framework. The low price of land in suburban areas has promoted the suburbanization of industrial land, while urban expansion has led to rapid growth in land values. Driven by price factors such as real estate investment, the government, developers, and other stakeholders have accelerated the rapid transformation of industrial land in the city center, resulting in the significant deindustrialization of the city center [17,67]. China’s land finance system has intensified the intensity and scale of the government’s efforts to promote deindustrialization in urban centers [30,44].
Second, transportation conditions have a significant impact on the layout of newly added industrial land. Among them, the explanatory variable of distance to the main road (MAI) was significantly associated with the independent variable during 1994–2007, 2007–2013, and 2013–2019 in the Gain Model, but the regression coefficient β was positive in three periods, indicating that the probability of converting to new industrial land decreases with each unit of increase in distance to the main road, which is different from the study in Shanghai [32]. One reason for this is that the study area is different. The study area in Shanghai mainly focuses on the central area; our study area includes the central area (Zone I–III) and the suburbs (Zone IV). Meanwhile, transportation conditions have a profound impact on land prices in the suburbs; the closer the location is to the main road, the greater the competitiveness of land for commercial use and residential use compared to industrial use [42]. In addition, in the Gain Model, the explanatory variable distance to the entrance/exit of the expressway (EXP) was significantly associated with the independent variable during 2007–2013 and 2013–2019; however, the explanatory variable distance to the railway freight station (RAI) was not significantly associated with the independent variable during 1994–2007, 2007–2013, and 2013–2019. This is closely related to the transformation of Xi’an’s industrial structure and national expressway network. On the one hand, Xi’an’s industry transformed from manufacturing light and confidential machinery and textiles to high-tech industries such as electronic information, biotechnology, aerospace, modern medicine, new materials, and new energy around the 21st century, and these high-tech industries have a relatively weak dependence on railway freight stations; for example, Xi’an Hi-tech Industrial Development Zone and Xi’an Economic and Technological Development Zone have not set up railway freight stations. On the other hand, since 2000, with the construction and improvement of the national expressway network, the newly added industrial land started clustering around the entrance/exit of the expressway. In 2019, the industrial land within the 3 km buffer zone of the entrance/exit of the expressway accounted for 42.95%. The change in the ratio of rail to road freight turnover in Xi’an from 1993 to 2019 supports the aforementioned view (Figure 8).
Overall, the significance of the dependent and independent variables in the Gain Model indicates a close correlation between transportation and the suburbanization of industrial land. As stated in the industrial location theory, the optimal location for enterprise site selection is closely related to transportation costs. In the 20th century, traditional manufacturing had a large demand for bulk cargo transportation and therefore relied more heavily on railway stations. Since the 21st century, however, the products of Xi’an’s high-tech industry have been small in size and light in weight, mainly relying on road transportation. In particular, the proportion of industrial land within a 3 km buffer around the railway freight station decreased by 50.23% from 1994 to 2019 in Xi’an, while the proportion of industrial land near to the entrance/exit of the expressway increased rapidly by 42.95%. However, the dependent and independent variables are not significant in the Loss Model, which differs from the findings in Amsterdam [15]. The reasons for this phenomenon include the following: (1) From the perspective of industrial structures, Xi’an has achieved a transformation from traditional manufacturing to high-tech industries, and the impact of railway transportation on industrial land layout is relatively weak. (2) From the perspective of expansion modes, Xi’an’s urban expansion is circular, with highways distributed in a ring shape in the suburbs, and the entrance/exit of the expressway are set up at nearby locations, which to some extent affects the model’s results. (3) More importantly, the locations of different manufacturing sectors have spatial heterogeneity [4,29]. In 2019, the central urban area of Xi’an still retained 26.75 km2 of industrial land, which is consistent with research on London [18], Greater Vancouver [13], and Cleveland [12]. This reveals that the transportation cost reduction associated with suburbanization does not mean that the suburbs are the optimal location for all manufacturing activities. Therefore, it is crucial to accommodate manufacturing in the city center [5].
Finally, the three neighborhood indices have different effects. The explanatory variable of the percentage of the area occupied by existing industrial land (DEN), which represents the agglomeration benefits of industrial land, was significantly associated with the independent variable in the Gain and Loss Models during 1994–2007, 2007–2013, and 2013–2019, which shows that the agglomeration benefits promote the process of suburbanization and deindustrialization [30,48,61]. The explanatory variable of the percentage of potential land around existing industrial areas that could be converted for industrial use (POT), which represents the land supply, was significantly associated with the independent variable in the Gain Models during 2007–2013 and 2013–2019, which indicated that sufficient land supply guides the agglomeration of manufacturing units to suburbanization [16,18,42,43]. However, there was no significance during 1994–2007, 2007–2013, and 2013–2019 in the Loss Model, which is contrary to the results for Hangzhou [30] and Cleveland [12]. The reason for the results may be closely related to land prices and local government actions. During the period of rapid urban expansion, Xi’an’s urban construction was mainly characterized by great expansion [68]; there was a large amount of land around industrial land located in the suburbs and central area that could be converted for industrial use. However, the land transfer price for commercial or residential land in the central area is significantly higher than that for industrial use. For instance, the land transfer price for commercial use is 29 times higher than that for industrial use around Tangyan Road in Xi’an Hi-tech Industrial Development Zone in 2018 [52]. As a result, local governments transfer land in the central area for commercial or residential use rather than industrial use to maximize land transfer revenue. The explanatory variable of the percentage of residential land surrounding existing industrial areas (RES), which represents the NIMBY effect of industrial land, was weakly but significantly associated with the dependent variable in the Gain and Loss Models, contradicting the findings of studies conducted in other cities such as London [69] but aligning with the findings of those conducted in Hangzhou [30]. The reason for this phenomenon is that urban planning has long pursued the “Zoning Law” based on the principle of non-interference between different functions and spatial isolation in the traditional industrial economy period [5]. However, the “new economy” with office buildings as workshops, clean and light production modes, and pollution-free cities has not interfered with residential land in Xi’an in the 21st century [70], and industries are now returning to cities and integrating into living spaces [71].

6. Conclusions

The evolution characteristics and influencing factors in industrial land in Xi’an from 1994 to 2019 were investigated by using land use data extracted from Google Earth satellite images, urban master planning data, cadastral data, etc. According to the results, the industrial land area increased from 55.02 km2 to 126.63 km2 from 1994 to 2013, but later rapidly reduced to 106.61 km2 in 2019. The spatial distribution of industrial land exhibited significant deindustrialization and suburbanization during park-orientated agglomeration; the industrial land area in the suburbs increased by 69.75 km2 from 1994 to 2019, but it decreased by 18.19 km2 in the central area. The development zone policy, land prices, the agglomeration effect, land supply, and transportation infrastructure are key determinants of suburbanization, with land prices and agglomeration benefits being the key factors guiding the deindustrialization of urban centers.
The urban problems caused by the evolution of industrial land underscore the severe challenges faced by current industrial land management policies. Our research indicates that government and market forces jointly shape the evolution of industrial land, with the government playing a pivotal role in this process. Therefore, the government should formulate active management policies to promote the rational layout and efficient use of industrial land. For cities undergoing rapid industrialization, the government should actively establish development zones to promote the agglomeration of manufacturing through market factors such as land prices, land supply, the agglomeration economy, and transportation facilities. For cities undergoing rapid deindustrialization, formulating industrial land protection policies and guiding industrial upgrading, rather than adjusting industrial land use, are effective measures to alleviate urban problems.
This study still has limitations. First, existing research has confirmed the spatial heterogeneity among different manufacturing sectors. Exploring the spatial evolution characteristics and driving factors behind diverse manufacturing sectors would contribute to enriching the theory of industrial land evolution in the future. Second, regression models have been extensively proven to be effective in analyzing the mechanisms of land use change at the macro scale. However, regression models struggle to reveal the spatial heterogeneity of micro scale influencing factors in land use change due to ignoring scale effects. Therefore, it is necessary to enhance the application of multi-scale models to fully uncover the complex mechanisms and spatial heterogeneity in land use change. Third, due to different cities varying in policy environments, regional differences, and socioeconomic contexts, the applicability and generalization of this study still need further validation in order to refine and validate the multidimensional framework.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/land14020213/s1, Table S1: Correlation coefficients in the Gain Model from 1974 to 1994; Table S2: Correlation coefficients in the Gain Model from 1994 to 2007; Table S3: Correlation coefficients in the Gain Model from 2007 to 2013; Table S4: Correlation coefficients in the Gain Model from 2013 to 2019; Table S5: Correlation coefficients in the Loss Model from 1974 to 1994; Table S6: Correlation coefficients in the Loss Model from 1994 to 2007; Table S7: Correlation coefficients in the Loss Model from 2007 to 2013; Table S8: Correlation coefficients in the Loss Model from 2013 to 2019.

Author Contributions

Conceptualization, D.Z., K.L. and J.L.; methodology, D.Z.; writing—original draft preparation, D.Z., K.L. and J.L.; writing—review and editing, D.Z., K.L., J.L. and J.Z.; visualization, D.Z. and J.Z.; funding acquisition, D.Z. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by Financial Department of Shaanxi Province’s Special Project of Industry-city Integration and the National Natural Science Foundation of China (No. 42071211).

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Conflicts of Interest

The authors declare no conflicts of interest.

Note

1
NIMBY (Not in my backyard) effect refers to the situation where residents or local entities, fearing that construction projects (such as landfills, nuclear power plants, manufacturing, funeral homes, etc.) will bring numerous negative impacts on their health, environmental quality, and property values, develop a sentiment of aversion.

References

  1. Leigh, N.G.; Hoelzel, N.Z. Smart Growth’s Blind Side. J. Am. Plan. Assoc. 2012, 78, 87–103. [Google Scholar] [CrossRef]
  2. Karen, C. The Highest and Best Use? Urban Industrial Land and Job Creation. Econ. Dev. Q. 2014, 28, 300–313. [Google Scholar]
  3. Green, J.W. Zoning for Jobs-Industrial Land Preservation in the Nation’s Largest Cities. OSF Prepr. 2018, 1–18. Available online: https://econpapers.repec.org/paper/osfosfxxx/ud345.htm (accessed on 19 January 2025).
  4. Scott, A.J. Locational Patterns and Dynamics of Industrial Activity in the Modern Metropolis. Urban Stud. 1982, 19, 111–142. [Google Scholar] [CrossRef]
  5. Lester, T.W.; Kaza, N.; Kirk, S. Making Room for Manufacturing: Understanding Industrial Land Conversion in Cities. J. Am. Plan. Assoc. 2014, 79, 295–313. [Google Scholar] [CrossRef]
  6. Balsas, C.J.L. Reconsidering Industrial Policy in Eastern New York, U.S.A. Urban Res. Pract. 2020, 15, 505–528. [Google Scholar] [CrossRef]
  7. Hudalah, D.; Firman, T. Beyond Property: Industrial Estates and Post-suburban Transformation in Jakarta Metropolitan Region. Cities 2011, 29, 40–48. [Google Scholar] [CrossRef]
  8. Yuan, X.; Wang, X.; Wang, Y.; Li, J.; Zhang, Y.; Gao, Z.; Zhang, G. Commuting Pattern Recognition of Industrial Parks Using Mobile Phone Signaling Data: A Case Study of Nanjing, China. Land 2024, 13, 1605. [Google Scholar] [CrossRef]
  9. Frantál, B.; Greer-Wootten, B.; Klusáek, P.; Krejčí, T.; Kunc, J.; Martinát, S. Exploring spatial patterns of urban brownfields regeneration: The case of Brno, Czech Republic. Cities 2015, 44, 9–18. [Google Scholar] [CrossRef]
  10. Ferm, J. Beyond the Post-Industrial City: Valuing and Planning for Industry in London. Urban Stud. 2017, 54, 3380–3398. [Google Scholar] [CrossRef]
  11. Walker, R. Industry Builds the City: The Suburbanization of Manufacturing in the San Francisco Bay Area, 1850–1940. J. Hist. Geogr. 2001, 27, 36–57. [Google Scholar] [CrossRef]
  12. Simon, C.J. Analysis of Manufacturer Location and Relocation in the Cleveland Metropolitan Area: 1966–1971. Am. Econ. 1980, 24, 35–42. [Google Scholar] [CrossRef]
  13. Steed, G.P.F. Intrametropolitan Manufacturing: Spatial Distribution and Locational Dynamics in Greater Vancouver. Can. Geogr. 1973, 17, 235–258. [Google Scholar] [CrossRef]
  14. Lewis, R.D. A City Transformed: Manufacturing Districts and Suburban Growth in Montreal, 1850–1929. J. Hist. Geogr. 2001, 27, 20–35. [Google Scholar] [CrossRef]
  15. Molle, W.T.M. Industrial mobility-A Review of Empirical Studies and An Analysis of the Migration of Industry from the City of Amsterdam. Reg. Stud. 1977, 11, 323–335. [Google Scholar] [CrossRef]
  16. Lloyd, P.E.; Mason, C.M. Manufacturing Industry in the Inner City: A Case Study of Greater Manchester. Trans. Inst. Br. Geogr. 1978, 3, 66–90. [Google Scholar] [CrossRef]
  17. Walker, R.; Lewis, R.D. Beyond the Crabgrass Frontier Industry and the Spread of North American Cities, 1850–1950. J. Hist. Geogr. 2001, 27, 3–19. [Google Scholar] [CrossRef]
  18. Ferm, J.; Jones, E. London’s Industrial Land: Cause for Concern? University College London: London, UK, 2015; Available online: https://discovery.ucl.ac.uk/id/eprint/1461419 (accessed on 12 October 2024).
  19. Keeble, D.E. Industrial Decentralization and the Metropolis: The North-West London Case. Trans. Inst. Br. Geogr. 1968, 44, 1–54. [Google Scholar] [CrossRef]
  20. Muller, E.K.; Groves, P.A. The Emergence of Industrial Districts in Mid-Nineteenth Century Baltimore. Geogr. Rev. 1979, 69, 159–178. [Google Scholar] [CrossRef]
  21. Neuss, L.V. Globalization and Deindustrialization in Advanced Countries. Struct. Chang. Econ. Dyn. 2018, 45, 49–63. [Google Scholar] [CrossRef]
  22. Vu, K.; Haraguchi, N.; Amann, J. Deindustrialization in Developed Countries Amid Accelerated Globalization: Patterns, Influencers, and Policy insights. Struct. Chang. Econ. Dyn. 2021, 59, 454–469. [Google Scholar] [CrossRef]
  23. Wolf-Powers, L. Up-Zoning New York City’s Mixed-Use Neighborhoods. J. Plan. Educ. Res. 2005, 24, 379–393. [Google Scholar] [CrossRef]
  24. Grodach, C.; Gibson, C.R. Advancing Manufacturing?: Blinkered Visions in U.S. and Australian Urban Policy. Urban Policy Res. 2018, 37, 279–293. [Google Scholar] [CrossRef]
  25. Asabere, P.K.; Huffman, F.E. Zoning and Industrial Land Values: The Case of Philadelphia. Real Estate Econ. 1991, 19, 154–160. [Google Scholar] [CrossRef]
  26. Davis, J.; Renski, H. Do Industrial Preservation Policies Protect and Promote Urban Industrial Activity? Examining the Impact of New York City’s Industrial Business Zone Program. J. Am. Plan. Assoc. 2020, 86, 431–442. [Google Scholar] [CrossRef]
  27. Ferm, J.; Panayotopoulos-Tsiros, D.; Griffiths, S. Planning urban manufacturing, industrial building typologies, and built environments: Lessons from inner London. Urban Plan. 2021, 6, 350–367. [Google Scholar] [CrossRef]
  28. Zhang, R.; Yan, W. The 50-year Spatial Transition of Suburban Industrial Parks and Impacts on Sustainable Urbanization in the Tokyo Metropolitan Area. Sustain. Cities Soc. 2024, 113, 105679. [Google Scholar] [CrossRef]
  29. Wu, Y.; Wei, Y.D.; Li, H. Firm Suburbanization in the Context of Urban Sprawl Neighborhood Effect and Sectoral Difference. Prof. Geogr. 2020, 72, 598–617. [Google Scholar] [CrossRef]
  30. Zhang, L.; Yue, W.; Liu, Y.; Fan, P.; Wei, Y.D. Suburban industrial land development in transitional China: Spatial restructuring and determinants. Cities 2018, 78, 96–107. [Google Scholar] [CrossRef]
  31. Gao, B.; Liu, W.; Dunford, M. State Land Policy, Land Markets and Geographies of Manufacturing: The Case of Beijing, China. Land Use Policy 2014, 36, 1–12. [Google Scholar] [CrossRef]
  32. Fan, P.; Yue, W.; Zhang, J.; Huang, H.; Messina, J.; Verburg, P.H.; Qi, J.; Moore, N.; Ge, J. The Spatial Restructuring and Determinants of Industrial Landscape in a Mega City Under Rapid Urbanization. Habitat Int. 2020, 95, 102099. [Google Scholar] [CrossRef]
  33. Zhang, Z.; Liu, J.; Gu, X. Reduction of Industrial Land Beyond Urban Development Boundary in Shanghai: Differences in Policy Responses and Impact on Towns and Villages. Land Use Policy 2019, 82, 620–630. [Google Scholar] [CrossRef]
  34. Zhang, J.; Zhang, D.; Huang, L.; Wen, H.; Zhao, G.; Zhan, D. Spatial Distribution and Influential Factors of Industrial Land Productivity in China’s Rapid Urbanization. J. Clean. Prod. 2019, 234, 1287–1295. [Google Scholar] [CrossRef]
  35. Dai, B.; Gu, X.; Xie, B. Policy Framework and Mechanism of Life Cycle Management of Industrial Land (LCMIL) in China. Land Use Policy 2020, 99, 104997. [Google Scholar] [CrossRef]
  36. Kuang, W.; Liu, J.; Dong, J.; Chi, W.; Zhang, C. The Rapid and Massive Urban and Industrial Land Expansions in China between 1990 and 2010: A CLUD-based Analysis of Their Trajectories, Patterns, and Drivers. Landsc. Urban Plan. 2016, 145, 21–33. [Google Scholar] [CrossRef]
  37. Ministry of Housing and Urban-Ruaral Development of the People’s Republic of China. China City Construction Statistical Yearbook 2020. 2021. Available online: https://www.mohurd.gov.cn/gongkai/fdzdgknr/sjfb/tjxx/jstjnj/index.html (accessed on 14 January 2025).
  38. Huang, Z.; Du, X. Government Intervention and Land Misallocation: Evidence from China. Cities 2017, 60, 323–332. [Google Scholar] [CrossRef]
  39. Lin, J.; Li, X.; Shen, J. Industrial Land Protection and Allocation Efficiency: Evidence from Guangdong, China. Land 2024, 13, 2081. [Google Scholar] [CrossRef]
  40. Chen, J.; Wang, Y. Ambiguous Property Rights as A Double-Edged Sword: A Case Study of Industrial Land Redevelopment in Shanghai, China. Cities 2022, 126, 103686. [Google Scholar] [CrossRef]
  41. Gao, J.; Qiao, W.; Ji, Q.; Yu, C.; Sun, J.; Ma, Z. Intensive-use-oriented Identification and Optimization of Industrial Land Readjustment During Transformation and Development: A Case Study of Huai’an, China. Habitat Int. 2021, 118, 102451. [Google Scholar] [CrossRef]
  42. Cheng, J. Analysis of the Factors Influencing Industrial Land Leasing in Beijing of China Based on the District-level data. Land Use Policy 2022, 122, 106389. [Google Scholar] [CrossRef]
  43. Hang, X.; Lin, Y.; Wu, Y.; Skitmore, M. Industrial Land Price Between China’s Pearl River Delta and Southeast Asian Regions: Competition or Coopetition? Land Use Policy 2017, 61, 575–586. [Google Scholar] [CrossRef]
  44. Yuan, F.; Wei, Y.D.; Xiao, W. Land Marketization, Fiscal Decentralization, and the Dynamics of Urban Land Prices in Transitional China. Land Use Policy 2019, 89, 104208. [Google Scholar] [CrossRef]
  45. Gao, J.; Chen, W.; Yuan, F. Spatial Restructuring and the Logic of Industrial Land Redevelopment in Urban China: I. Theoretical Considerations. Land Use Policy 2017, 68, 604–613. [Google Scholar] [CrossRef]
  46. Li, Y.; Chen, X.; Tang, B.; Wong, S.W. From Project to Policy: Adaptive Reuse and Urban Industrial Land Restructuring in Guangzhou City, China. Cities 2018, 82, 68–76. [Google Scholar] [CrossRef]
  47. Hu, Y.; Lu, B.; Wu, J. Value Capture in Industrial Land Renewal Under the Public Leasehold System: A Policy Comparison in China. Land Use Policy 2019, 84, 59–69. [Google Scholar] [CrossRef]
  48. Huang, L.; Shahtahmassebi, A.; Gan, M.; Deng, J.; Wang, J.; Wang, K. Characterizing Spatial Patterns and Driving Forces of Expansion and Regeneration of Industrial Regions in the Hangzhou Megacity, China. J. Clean. Prod. 2020, 253, 119959. [Google Scholar] [CrossRef]
  49. Chen, X.; Wang, H. Spatial–Temporal Evolution and Driving Factors of Industrial Land Marketization in Chengdu–Chongqing Economic Circle. Land 2024, 13, 972. [Google Scholar] [CrossRef]
  50. Tan, X.; Chen, W.; Cao, H.; Li, Q.; Zhang, S. Identifying Spatiotemporal Characteristics and Driving Factors of Industrial Land Use Morphology in China. Expert Syst. Appl. 2024, 235, 121231. [Google Scholar] [CrossRef]
  51. Yan, G.; Wang, S. Coexistence and Transformation from Urban Industrial Land to Green Space in Decentralization of Megacities: A Case Study of Daxing District, Beijing, China. Ecol. Indic. 2023, 156, 111120. [Google Scholar] [CrossRef]
  52. Zhao, D.; Liu, K.; Li, J.; Pang, G.; Meng, Y. The Characteristics and Spatial Effects of the Spatio-Temporal Pattern Evolution of Industrial Land in Xi’an. China City Plan. Rev. 2022, 31, 6–16. [Google Scholar]
  53. Cao, H.; Chen, W.; Tan, X.; Li, Q. Identification and Driving Mechanism of the Industrial Land Use Transition in China. Habitat Int. 2023, 138, 102848. [Google Scholar] [CrossRef]
  54. Xie, F.; Zhang, S.; Zhao, K.; Quan, F. Evolution Mode, Influencing Factors, and Socioeconomic Value of Urban Industrial Land Management in China. Land 2022, 11, 1580. [Google Scholar] [CrossRef]
  55. Weber, A. Theory of the Location of Industries; University of Chicago Press: Chicago, IL, USA, 1929. [Google Scholar]
  56. Will, R.A. Federal Influences on Industrial Location: How Extensive? Land Econ. 1964, 40, 49–57. [Google Scholar] [CrossRef]
  57. Blair, J.P.; Premus, R. Major Factors in Industrial Location: A Review. Econ. Dev. Q. 2016, 1, 72–85. [Google Scholar] [CrossRef]
  58. Gripaios, P. Industrial Decline in London: An Examination of its Causes. Urban Stud. 1977, 14, 181–189. [Google Scholar] [CrossRef]
  59. Stahura, J.M. Determinants of Suburban Job Change in Retailing, Wholesaling, Service and Manufacturing Industries: 1960–1972. Sociol. Focus 1982, 15, 347–357. [Google Scholar] [CrossRef]
  60. Park, J.I.; Leigh, N.G. Urban Industrial Land Loss and Foreign Direct Investment-related Manufacturing Job Sprawl: An Atlanta, Georgia MSA Case Study. J. Urban Technol. 2017, 24, 95–113. [Google Scholar] [CrossRef]
  61. Tian, L.; Ge, B.; Li, Y. Impacts of State-led and Bottom-up Urbanization on Land Use Change in the Peri-urban Areas of Shanghai: Planned Growth or Uncontrolled Sprawl? Cities 2017, 60, 476–486. [Google Scholar] [CrossRef]
  62. Wang, K.; Li, G.; Liu, H. Location Choice of Industrial Land Reduction in Metropolitan Area: Evidence from Shanghai in China. Growth Chang. 2020, 51, 1837–1859. [Google Scholar] [CrossRef]
  63. Xi’an Municipal Bureau of Statistics. Xi’an Statistical Yearbook 2020. 2021. Available online: http://tjj.xa.gov.cn/tjnj/2020/zk/indexch.htm (accessed on 14 January 2025).
  64. Xi’an Natural Resources and Planning Bureau (Municipal Forestry Bureau). Xi’an City Master Plan. 2024. Available online: http://zygh.xa.gov.cn/ywpd/cxghgsgb/ghgsgb/1.html (accessed on 14 January 2025).
  65. Cheng, A. The Proposals and Suggestions Study to Some Problems of Planning Management in Xian High-Tech Development Industries Zone; Xi’an University of Architecture and Technology: Xi’an, China, 2006. (In Chinese) [Google Scholar]
  66. Ferm, J.; Jones, E. Mixed-use ‘regeneration’ of employment land in the post-industrial city: Challenges and realities in London. Eur. Plan. Stud. 2016, 24, 1913–1936. [Google Scholar] [CrossRef]
  67. Curran, W. ‘From the Frying Pan to the Oven’: Gentrification and the Experience of Industrial Displacement in Williamsburg, Brooklyn. Urban Stud. 2007, 44, 1427–1440. [Google Scholar] [CrossRef]
  68. Geng, T.; Mao, Y.; Li, J.; Chen, H. Spatio-Temporal Characteristics and Driving Mechanism of Xi’an Urban Expansion. Econ. Geogr. 2019, 39, 62–70. Available online: https://kns.cnki.net/kcms2/article/abstract?v=1i9K2Vab0pPbrdUfWhmhVB2LhAWHRn3wRc8zZF-5Ul4btQQjd3EuD2YFJ8Ex0YmmLYkSJxF61rcG5wLm5NQYLKu1F5C-I9sQ3071p3LEKrJ22hJeIfaYITB8q8WYjijhHO2STHbB1DNVCkuoDQ8-10tx7Y9r7QkGHjRGyO8qk4tMtPfQFmhMiUliW-EJB0ud3Z395TCTUovRCiR6hK9xEA==&uniplatform=NZKPT (accessed on 19 January 2025).
  69. Keeble, D.E. Industrial Migration from North-west London, 1940–1964. Urban Stud. 1965, 2, 15–32. [Google Scholar] [CrossRef]
  70. Wang, H. Spatial Impacts of New Economies and the Implications for City Planning and Decision-making. Geogr. Res. 2007, 26, 577–589. [Google Scholar]
  71. Park, J.I. Re-urbanization Pattern of Manufacturing and Characteristics of Urban Manufacturing in South Korea. Cities 2023, 137, 104330. [Google Scholar] [CrossRef]
Land 14 00213 g001
Figure 1. Analytical framework.
Figure 1. Analytical framework.
Land 14 00213 g001
Land 14 00213 g002
Figure 2. The study area.
Figure 2. The study area.
Land 14 00213 g002
Land 14 00213 g003
Figure 3. The datasets for the dependent variable Y.
Figure 3. The datasets for the dependent variable Y.
Land 14 00213 g003
Land 14 00213 g004
Figure 4. Spatial distribution of industrial land in Xi’an from 1994 to 2019.
Figure 4. Spatial distribution of industrial land in Xi’an from 1994 to 2019.
Land 14 00213 g004
Land 14 00213 g005
Figure 5. Sankey diagram of industrial land sources and transformation proportions in Xi’an from 1994 to 2019. Note: 1 Due to the fact that some industrial buildings in satellite images have been demolished and the land use cannot be defined, the statistics are presented according to unutilized land. 2 Due to the fact that the land use in satellite images cannot be defined, the statistics are presented according to other land.
Figure 5. Sankey diagram of industrial land sources and transformation proportions in Xi’an from 1994 to 2019. Note: 1 Due to the fact that some industrial buildings in satellite images have been demolished and the land use cannot be defined, the statistics are presented according to unutilized land. 2 Due to the fact that the land use in satellite images cannot be defined, the statistics are presented according to other land.
Land 14 00213 g005
Land 14 00213 g006
Figure 6. The spatial distribution of industrial land transformation in Xi’an from 1994 to 2019.
Figure 6. The spatial distribution of industrial land transformation in Xi’an from 1994 to 2019.
Land 14 00213 g006
Land 14 00213 g007
Figure 7. Adjustment of industrial land layout in Xi’an urban master plan (2008–2020) (revised in 2017).
Figure 7. Adjustment of industrial land layout in Xi’an urban master plan (2008–2020) (revised in 2017).
Land 14 00213 g007
Land 14 00213 g008
Figure 8. Ratio of railway to highway freight turnover.
Figure 8. Ratio of railway to highway freight turnover.
Land 14 00213 g008
Table 1. Description of explanatory variables.
Table 1. Description of explanatory variables.
VariablesDescriptionData SourcesData Processing
DZWhether or not it is in a national or provincial development zone (yes = 1)Xi’an urban master plan and development master planDraw the boundaries of development zones in different periods.
RAIDistance from the railway freight station (m)Remote sensing image dataCalculate the Euclidean distance from the explanatory variable to the nearest railway freight station.
EXPDistance from the high-speed entrance and exit (m)Remote sensing image dataCalculate the Euclidean distance from the explanatory variable to the nearest high-speed entrance and exit.
MAIDistance from major roads (m)Remote sensing image dataCalculate the Euclidean distance from the explanatory variable to the nearest main road.
BLPIndustrial land price (CNY/m2)China land price monitoring network, benchmark land price map of Xi’an and Xixian New AreaDraw the benchmark land price map of industrial land in different periods.
PVIRLand value increment (CNY/m2)China land price monitoring network, benchmark land price map of Xi’an and Xixian New AreaThe difference between the benchmark land price for residential use at T2 time and the benchmark land price for industrial use at T1 time.
PVIBLand value increment (CNY/m2)China land price monitoring network, benchmark land price map of Xi’an and Xixian New AreaThe difference between the benchmark land price for commercial services at T2 time and the industrial benchmark land price at T1 time.
POTPercentage of available land (%)Remote sensing image data, urban cadastral data, etc.Establish a buffer zone with a radius of 1 km and calculate the percentage of land potentially converted to industrial use in the buffer zone.
DENExisting industrial land density (%)Remote sensing image data, urban cadastral data, etc.Establish a buffer zone with a radius of 1 km and calculate the percentage of existing industrial land in the buffer zone.
RESExisting residential land density (%)Remote sensing image data, urban cadastral data, etc.Establish a buffer zone with a radius of 1 km and calculate the percentage of existing residential land in the buffer zone.
Table 2. Area of industrial land in Xi’an from 1974 to 2019.
Table 2. Area of industrial land in Xi’an from 1974 to 2019.
Year1994200720132019
Zone IArea (km2)0.340.080.040.00
Proportion (%)0.620.070.030.00
Zone IIArea (km2)11.327.832.281.34
Proportion (%)20.577.211.801.26
Zone IIIArea (km2)33.2564.1152.6925.41
Proportion (%)60.4358.9941.6123.83
Zone IVArea (km2)10.1136.6571.6279.86
Proportion (%)18.3833.7356.5674.91
TotalArea (km2)55.02108.67126.63106.61
Proportion (%)100100100100
Table 3. Estimated results for regression models in 1994–2007.
Table 3. Estimated results for regression models in 1994–2007.
Gain ModelLoss Model
VariableβSDpExp (β)VariableβSDpExp (β)
DZ1.3470.5830.0213.845DZ−0.5650.7740.4660.569
RAY0.0000.0000.3331.000RAY0.0000.0000.3091.000
MAI−0.0020.0010.0190.998MAI−0.0030.0030.2190.997
BLP−0.0800.0330.0150.923EXP0.0000.0000.9881.000
DEN0.1010.0280.0001.107PVIR0.0020.0010.0011.002
RES0.0250.0440.5671.025POT−0.0040.0190.8310.996
Constant0.4020.7520.5931.495DEN−0.0610.0230.0090.941
RES−0.0080.0290.7990.993
Constant−0.8051.8510.6630.447
Note: β means coefficient. Exp (β) is the factor change in odds for one unit increase in variables.
Table 4. Estimated results for regression models in 2007–2013.
Table 4. Estimated results for regression models in 2007–2013.
Gain ModelLoss Model
VariableβSDpExp (β)VariableβSDpExp (β)
DZ−0.9450.4080.0200.389DZ−0.3300.2600.2040.719
EXP0.0000.0000.0071.000EXP0.0000.0000.2601.000
MAI−0.0060.0010.0000.994MAI−0.0010.0010.1950.999
RAY0.0000.0000.0731.000RAY0.0000.0000.1171.000
BLP−0.0050.0020.0090.995POT−0.0120.0090.1890.988
POT0.0520.0140.0001.053DEN−0.0770.0110.0000.926
DEN0.1150.0260.0001.122PVIB0.0000.0000.0461.000
RES−0.0010.0420.9810.999Constant1.3720.5770.0173.943
Constant0.0251.5940.9871.025
Note: β means coefficient. Exp (β) is the factor change in odds for one unit increase in variables.
Table 5. Estimated results for regression models in 2013–2019.
Table 5. Estimated results for regression models in 2013–2019.
Gain ModelLoss Model
VariableβSDpExp (β)VariableβSDpExp (β)
DZ1.7330.4110.0005.657DZ−1.4481.0240.1570.235
EXP0.0000.0000.0031.000EXP0.0000.0000.8911.000
MAI−0.0020.0010.0120.998MAI0.0030.0020.3101.003
RAY0.0000.0000.1661.000RAY0.0000.0000.0031.000
BLP−0.0150.0030.0000.985POT−0.0280.0200.1560.972
POT0.0250.0080.0021.025DEN−0.0810.0290.0060.922
DEN0.1070.0230.0001.112RES1.7150.4120.0005.557
RES0.0410.0320.1931.042PVIR0.0020.0000.0021.002
Constant2.3391.4270.10110.370PVIB−0.0040.0010.0000.996
Constant6.6693.6430.067787.345
Note: β means coefficient. Exp (β) is the factor change in odds for one unit increase in variables.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Zhao, D.; Liu, K.; Li, J.; Zhai, J. Spatial Reconstruction and Determinants of Industrial Land in China’s Urban Expansion: A Theoretical Framework. Land 2025, 14, 213. https://doi.org/10.3390/land14020213

AMA Style

Zhao D, Liu K, Li J, Zhai J. Spatial Reconstruction and Determinants of Industrial Land in China’s Urban Expansion: A Theoretical Framework. Land. 2025; 14(2):213. https://doi.org/10.3390/land14020213

Chicago/Turabian Style

Zhao, Dan, Kewei Liu, Jianwei Li, and Jiagang Zhai. 2025. "Spatial Reconstruction and Determinants of Industrial Land in China’s Urban Expansion: A Theoretical Framework" Land 14, no. 2: 213. https://doi.org/10.3390/land14020213

APA Style

Zhao, D., Liu, K., Li, J., & Zhai, J. (2025). Spatial Reconstruction and Determinants of Industrial Land in China’s Urban Expansion: A Theoretical Framework. Land, 14(2), 213. https://doi.org/10.3390/land14020213

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop