Analysis of the Driving Mechanism of Land Comprehensive Carrying Capacity from the Perspective of Urban Renewal

Abstract

After a period of rapid development, the process of urbanization in China has gradually shifted from “scale expansion” to “enhanced quality”. The scarcity of urban land resources has created constraints on resources and economic development. This paper examines the carrying capacity of urban land resources from the perspective of urban renewal. A conceptual model of the driving mechanism of land comprehensive carrying capacity is constructed, incorporating six dimensions and 22 indicators, including urban renewal and urban ecology. Through questionnaire surveys and structural equation modeling, feedback data are analyzed, and measurement models, structural models, and mediation effects are examined to analyze the causal paths of factors in different dimensions on the comprehensive carrying capacity of urban land. The research results indicate that all six dimensions in the conceptual model have a direct positive impact on the land carrying capacity. In terms of direct effects, the influencing factors are ranked in descending order of magnitude as follows: urban development, urban disaster prevention and mitigation capacity, infrastructure development, urban renewal, social economy, and urban ecology. In terms of overall effects, factors are ranked in descending order of magnitude as follows: urban development, social economy, urban ecology, urban renewal, urban disaster prevention and mitigation capacity, and infrastructure development.

1. Introduction

Urban areas, as key growth poles driving economic development and important engines for achieving modernization, are not only active hubs for economic exchanges and the flow of factors, but also centers for material production and consumption [1]. The rapid advancement of urbanization in China has led to urban resources and the environment exceeding their carrying capacity, resulting in deteriorating living conditions, environmental degradation, energy and resource shortages, and a declining quality of economic growth [2]. Among these issues, land, as one of the rapidly depleting resources in the process of urban agglomeration, is closely related to urban construction activities. This is mainly due to the rapid and extensive growth of urban land use, with non-construction land in various urban fringe areas rapidly being converted into construction land [3,4].

Comprehensive land carrying capacity is the result of the comprehensive interactions between land use, regional socio-economic activities, industrial development, and policy guidance, involving numerous influencing factors and a wide range of aspects [5]. From the typical characteristics of urban spatial expansion, both domestically and internationally, land expansion exhibits a concentric distribution pattern, transitioning from a single center to a multi-center expansion [6,7]. In terms of the expansion process, it follows an upward spiral cycle of repeated growth. Economically driven growth serves as the fundamental driving force behind the expansion mechanism [8]. The assessment of the comprehensive land carrying capacity faces challenges related to regional complexity and the uncertainty of indicator systems. According to the existing research on the evaluation system of the land comprehensive carrying capacity, it has been found that the evaluation model of the land carrying capacity in different regions will have index uncertainties, due to regional reasons. The research on its influencing mechanism has gradually shifted from early-stage phenomenological descriptions and experiential judgments to quantitative determinations, such as econometric analysis and pressure–response models [9]. In theory, all factors associated with land-related activities have an impact on the comprehensive land carrying capacity, but not all factors result in significant changes [10].

Through a review of the existing research on the comprehensive carrying capacity of land, it was found that the current research on the comprehensive carrying capacity of land mainly considers the coordination relationship with urban economic development [11,12,13]. And some researchers are focused on analyzing the dynamic relationship between the vulnerability of the land carrying capacity and ecological resources [14,15,16,17]. However, with the slowing down of large-scale urban infrastructure development in China, some ultra-large urban agglomeration areas have entered a new stage of urban renewal. As a scarce resource, there is still a research gap on the influencing factors, evaluation methods, and driving mechanism of the comprehensive carrying capacity of land in the context of urban renewal.

Based on the existing research on the land carrying capacity, this paper takes an innovative perspective from urban renewal, and proposes the influencing factors of land carrying capacity in highly urbanized areas of China. It explores the driving mechanisms behind these factors. This novel contribution addresses the insufficient understanding of the relationship between the land carrying capacity and urban renewal in previous studies, while also enriching the research findings on the driving mechanisms of the land carrying capacity. On one hand, it supplements the research on the coordination between land resources and urban development. On the other hand, it promotes the improvement of the research system related to urban ecological resources.

In 2014, the National New Urbanization Plan (2014–2020) clearly pointed out that the accelerating cultivation of Chengdu-Chongqing, Central Plains, the middle reaches of the Yangtze River, Kazakhstan, and other urban agglomerations would take the lead in development, become an important growth pole to promote the balanced development of territorial space, and lead regional economic development [18]. In 2016, the Development Plan for Harbin-Chanchung urban agglomeration determined its status as the first key regional urban cluster, as a key location for the revitalization and development of the old industrial base in Northeast China, and as an important engine for the innovative development, coordinated development, green development, open development, shared development and comprehensive revitalization of the old industrial base in Northeast China. It aims to build a city cluster with core competitiveness and important influence in Northeast Asia by 2030, and become the core region of the China–Mongolia–Russia Economic Corridor [19].

The Harbin-Changchun urban agglomeration is an important region driving the development of the old industrial base in Northeast China, playing a crucial role in promoting regional economic growth and resource allocation. Firstly, from an economic perspective, the Harbin-Changchun urban agglomeration faces an urgent need for economic structural adjustment, transformation, and upgrading. Secondly, in terms of population, the northeast region has long been facing the issue of population decline, which has implications for regional development and functional restructuring, manifested in land use, infrastructure construction, and urban expansion. Thirdly, in terms of sustainable development of resources and the environment, the region is endowed with abundant natural resources and ecological environments. In-depth research on the land comprehensive carrying capacity of the Harbin-Changchun urban agglomeration contributes to assessing and managing land resource utilization and protecting the ecological environment to promote sustainable development. Finally, the study of the comprehensive land carrying capacity of the Harbin-Changchun urban agglomeration can provide key information to local governments, helping them develop sustainable development strategies for the region.

China has a vast territory, and there are huge differences in the levels of economic development and industrial structure in different regions. The urban agglomerations in each region have their own unique evolution model and driving mechanism [20]. With the industrial transformation of the old industrial base in Northeast China, and the urban renewal demand of the Harbin-Changchun urban agglomerations, the selection of indicators and survey data from existing studies are mainly based on provincial or city data, and there are few studies on urban agglomerations and insufficient attention is given to the old industrial base in Northeast China [21], which cannot directly reflect the land comprehensive carrying capacity level of the Harbin and Chang urban agglomerations.

Given that the land comprehensive carrying capacity in the context of urban renewal is influenced by multiple dimensions, and there are inherent interactions among these dimensions and their influencing factors, this study, based on the influencing factors, combines the explanation of exogenous variables and endogenous variables, using the structural equation model. It explores the complex paths between variables and the causal chains of influencing factors, tests the hypotheses regarding the mutual relationships among variables in different dimensions, and systematically elucidates the relationships between different dimensions and the impact mechanism on the comprehensive land carrying capacity. This study can address questions such as the effects of urban renewal on the land carrying capacity and its influencing mechanisms. It is of great significance for promoting the improvement in the comprehensive land carrying capacity through urban renewal and facilitating cities to move towards high-quality development. The logical relationship between the research content modules of this research is presented in Figure 1.

Therefore, this paper will first design and analyze the questionnaire on the basis of research methods and data acquisition, analyze the questionnaire data, build a model and then test it, so as to answer questions such as what impact urban renewal will have on the land comprehensive carrying capacity and the impact mechanism.

2. Materials and Methods

2.1. Overview of the Study Area

The Harbin-Chanchung urban agglomeration is located in the hinterland of Northeast China, between 29°05′~31°50′ north latitude and 112°30′~116°10′ east longitude. It is adjacent to the southern city cluster of Central Liaoning and the Beijing–Tianjin–Hebei City cluster in the south, and it is adjacent to the Far East region of Russia and the Korean Peninsula in the north and east, and to Mongolia and the eastern Siberia region of Russia in the west. The Harbin-Chanchung urban agglomeration consists of two sub-provincial cities, namely the capital city of Harbin and Changchun. The nine prefecture-level cities are Daqing, Qiqihar, Mudanjiang, Suihua, Jilin, Siping, Liaoyuan, Songyuan and the Yanbian Korean Autonomous Prefecture. It has 26 county-level cities and 621 towns, with a total area of about 32 km², shown as Figure 2:

From the early state-level industrial base to the reform of state-owned enterprises, the urban industrial structure and spatial structure have undergone great changes, and the urban construction and development have also changed. Resource-intensive industries dominate the urban agglomeration of Harbin and Changchun. Rational allocation of land resources and scientific distribution of spatial patterns are important opportunities to promote the coordinated development of cities in the urban agglomeration of Harbin and Changchun. The linkage development model of core city—the Harbin-Chanchung urban agglomeration, Northeast region—guides the formation of a new industrial structure and economic growth pole, and further promotes the overall balance at the land level.

2.2. Establishment of the Land Comprehensive Carrying Capacity Driving Mechanism Analysis Model

2.2.1. Concept and Development of Land Carrying Capacity

The concept of land carrying capacity originates from the concept of carrying capacity in physics, which refers to the maximum load that an object can carry without any damage. Because of its adaptability, the concept of carrying capacity has been applied in the fields of economics, ecology and sociology.

From the perspective of the application of the concept of the comprehensive carrying capacity in the field of ecology and sociology, when it is applied to the field of ecology, it aims to reveal the number of plants and animals that land resources can support. And when applied to the field of sociology, it emphasizes the number of people that land resources can carry. Therefore, it can be considered that the concept of land carrying capacity originated from physics, and was inspired by the fields of economics, ecology, and economics.

There are some differences and connections between the carrying capacity of land and the comprehensive carrying capacity of land. From the perspective of urban construction and planning, it can be considered the maximum capacity of a land resource system to carry the population or human activities, against the background of urban renewal. The concept of the comprehensive carrying capacity of land is more inclined to reveal the development scale and construction intensity of various human activities that urban land resources can carry [22].

2.2.2. Basics of Model Construction

The Food and Agriculture Organization (FAO) of the United Nations conducted studies on the land carrying capacity for human populations in 177 developing countries, excluding China, using the agro ecological zone (AEZ) methodology. This approach combines various factors, such as land use practices and socio-economic levels, to classify different countries into distinct ecological units for the quantitative assessment of the relationships between population, society, economy, and resources.

In relevant studies, such as organizational and economic research, it has been proposed that the efficiency of resource allocation and utilization stems from three factors: drivers, qualities, and the environment. Among these factors, drivers are the primary influencers and encompass various types, such as material forces and economic forces as objective drivers, as well as non-material economic forces like human agency. This factor is closely related to the comprehensive qualities of individuals, such as a sense of social responsibility, innovation level, management level, and cultural level, all of which have an impact on improving production efficiency. Therefore, it is necessary to analyze the impact of multiple factors on transformation drivers based on core motivational factors.

Therefore, based on the value belief, the research paradigm of land carrying capacity continues to adopt a comprehensive approach that considers multiple factors. It adheres to the principle of the integrated effects of multiple factors, and focuses on the diverse needs of human production, life, and ecology, as well as various social and economic activities. It emphasizes the interconnections and compensations between different behaviors and activities. Guided by the aforementioned concepts and methods, this paper constructs a conceptual framework for the model, using the “bearing material–medium–carrier” three-element framework, as shown in Figure 3.

From the perspective of urban renewal, as the scope and complexity of urban construction activities expands, it is necessary to consider the multidimensional impact of land development and construction activities in urban renewal on the comprehensive carrying capacity of the land before establishing the analysis model for the driving mechanisms of the land’s comprehensive carrying capacity. This can be summarized into three aspects:

• Socio-economic growth orientation: Economic growth is a development goal for governments at all levels, and accelerating industrial capital aggregation and urban construction are fundamental driving mechanisms for the current spatial expansion of cities in our country. Policies aimed at attracting external investments to expand the city’s scale have resulted in the generation of many measures to increase land supply. There is a certain similarity between the trends in land development and GDP growth in our country. Therefore, economic growth serves as the fundamental driving force for the growth of urban land construction activities and the spatial expansion of cities;
• Infrastructure construction: The proportion of public infrastructure investment in China has been increasing year by year, which is closely related to the trend of increasing urban land development. Infrastructure investment and construction serve as tools to drive economic growth and act as a link for coordinated urban and rural development [23]. With the increasing demand for road facilities, the government has intensified the supply of transportation infrastructure. The gradual improvement in infrastructure enables urban land to accommodate more construction activities and larger development scales;
• Resource and ecological environment: The ecological environment plays a constraining role in the transformation of urban land use patterns, and is essential for ensuring the sustainable development of cities [24]. Guiding the orderly expansion of cities and determining the distribution of various functional areas within a city have a significant impact on improving the urban environment and enhancing livability [25]. Additionally, in the process of urban renewal, it is important to consider the resilience of cities in social and natural disaster conditions.

Therefore, the core impacts of land development and construction activities can be expanded into six dimensions: urban development scale, socio-economic factors, urban renewal, urban infrastructure development, urban disaster resilience, and urban ecology, as indicated by references [26,27,28].

Table 1. The dimensions of comprehensive carrying capacity of urban land construction activities.

Urban Land Development and Construction Activities	Dimensions of Comprehensive Land Carrying Capacity Evaluation
Social and Economic Growth Orientation [29,30]	Social Economy
	Urban Development Scale
Infrastructure Construction [31,32]	Urban Infrastructure Development
	Urban Renewal
Resource and Ecological Environment [33] (including Urban Resilience)	Urban Ecology
	Urban Disaster Resilience

illustrates the dimensions of the comprehensive land carrying capacity assessments based on urban land development and construction activities for expansion.

2.2.3. Research Variables and Hypotheses

Based on the analysis of the model’s construction foundation mentioned above, further refinement and decomposition of the six evaluation dimensions within the three core contents were conducted. This was performed to propose a clear directional hypotheses and validate the feasibility of the pathways. As a result, the following research variables and hypotheses were formulated:

• The dimension of urban development scale: Urban scale is an important manifestation of urban renewal and a concrete dimension of socioeconomic growth. Land carrying capacity is closely related to urban scale. Specific indicators within this dimension include per capita residential land, road density, per capita fixed asset investments of the entire society, and value-added indicators of the main industry utilizing land resources, such as the construction industry. These indicators directly reflect the current state of the land carrying capacity;
• The dimension of social economy: This dimension directly reflects the orientation of socio-economic growth. It is manifested by indicators such as economic density, urbanization level, per capita GDP, and the Engel coefficient. This dimension represents the integration of qualitative and quantitative analyses. The indicators within this dimension reflect the correlation between the land carrying capacity and the level of urban economic development;
• The dimension of urban renewal: This dimension focuses on the improvement and optimization of spatial form and functions in urban built-up areas, transforming the scarcity function of urban land. It primarily reflects the effectiveness of urban infrastructure development. The indicators within this dimension include urban building density, floor area ratio, urban redevelopment cost, and resident satisfaction. These indicators establish the relationship between the perception of residents and the effectiveness of urban transformation, providing an assessment of the level of land carrying capacity;
• The dimension of urban ecology: This dimension captures the carrying capacity generated by the absorption capacity of urban land and environmental greening, reflecting the connection between land resources and urban ecological conditions. Considering the close relationship between land resources and urban ecology, and the fact that the urban ecological level is also an evaluation indicator for urban renewal, specific indicators within this dimension include per capita urban green space area, built-up area greening coverage rate, and centralized treatment rate of domestic wastewater. These indicators comprehensively reflect the level of urban ecological governance from the perspectives of residents and the effects of urban transformation;
• The dimension of urban infrastructure development: Urban land serves as the foundation for urban infrastructure development, and urban land resources are crucial carriers for providing various basic needs of urban residents, in terms of living and working. Therefore, indicators within this dimension include transportation accessibility, water supply capacity, power supply capacity, and infrastructure investment;
• The dimension of urban disaster resilience: This dimension focuses on the ability of cities to reduce potential risks and minimize disasters, both natural and man-made, by implementing effective measures and utilizing land resources rationally to enhance urban safety. The concentration of the population and the expansion of urban scale impose higher requirements on urban resilience. Specific indicators within this dimension include building disaster resistance rate, per capita road area, and stability of the drainage system, which are directly related to the land carrying capacity.

The specific indicators and variable hypotheses contained within the six dimensions mentioned above are presented in

Table 2. Research variables and hypotheses for comprehensive evaluation of the land carrying capacity from the perspective of urban renewal.

Dimensions	Indicator	Code	Assumption
Urban Development (UD)	Per Capita Residential Land	UD1	H1.  The expansion of urban development scale has a positive impact on the comprehensive carrying capacity of land. H2.  The expansion of urban development scale has a positive impact on socioeconomic development. H3.  The expansion of urban development scale has a positive impact on urban infrastructure construction.
	Road Density	UD2
	Per Capita Fixed Asset Investment in the Whole Society	UD3
	Value Added of the Construction Industry	UD4
Social Economy (SE)	Economic Density	SE1	H4.  Socioeconomic development has a positive impact on the comprehensive carrying capacity of land. H5.  Socioeconomic development has a positive impact on urban ecology. H6.  Socioeconomic development has a positive impact on urban infrastructure construction.
	Urbanization Level	SE2
	Per Capita GDP	SE3
	Engel’s Coefficient	SE4
Urban Renewal (UR)	Building Density	UR1	H7.  Urban renewal activities have a positive impact on the comprehensive carrying capacity of land. H8.  Urban renewal activities have a positive impact on urban infrastructure construction.
	Plot Ratio	UR2
	Renovation Cost	UR3
	Resident Satisfaction	UR4
Urban Ecology (UE)	Per Capita Urban Green Space Area	UE1	H9.  Urban ecology has a positive impact on the comprehensive carrying capacity of land. H10.  Urban ecology has a positive impact on urban infrastructure construction. H11.  Urban ecology has a positive impact on urban disaster prevention and mitigation capability.
	Green Coverage Rate of Built-up Area	UE2
	Concentration Rate of Domestic Wastewater Treatment	UE3
Urban Infrastructure Development (ID)	Transportation Accessibility	ID1	H12.  Urban infrastructure construction has a positive impact on the comprehensive carrying capacity of land.
	Water Supply Capacity	ID2
	Power Supply Capacity	ID3
	Infrastructure Investment	ID4
Urban Disaster Prevention and Mitigation Capability (DPMC)	Building Disaster Resistance Rate	DPMC1	H13.  Urban disaster prevention and mitigation capability has a positive impact on the comprehensive carrying capacity of land.
	Per Capita Road Area	DPMC2
	Stability of Water Supply and Drainage System	DPMC3

.

The above data sources are based on the 11 prefecture-level cities in the Harbin-Changchun urban agglomeration. In view of the availability of data, and the unification of relevant indicators in the study area, and considering the specific situation of each prefecture-level city in the study area, the total time range of the study area is determined as 2010–2020. The data for the indicators are mainly obtained from the Urban Statistical Yearbook (2011–2021), the Chinese Environmental Statistical Yearbook (2011–2021), the National Economic and Social Development Bulletin (2011–2021), the Urban Construction Statistical Yearbook (2011–2021), and some data from the Water Resources Bulletin of each municipality, and public information from municipal government departments and annual reports. The data used in the evaluation were calculated based on the original data, with some data outliers removed and missing data extrapolated from other years.

2.2.4. Construction of Conceptual Model

Based on the theoretical model and research hypotheses, a conceptual model depicting the causal pathways of factors influencing the comprehensive carrying capacity of land from the perspective of urban renewal was constructed, as shown in Figure 4.

2.3. Structural Equation Modeling

Structural equation modeling (SEM) is a model for handling complex multivariate data, that has been developed based on the basic principles of path analysis proposed by the geneticist Sewall Wright [34,35,36]. The components of SEM include latent variables, observed variables, and error variables [35]. Among them, latent variables cannot be directly measured and are constructed based on theory and assumptions, while observed variables (manifest variables) can be measured. Error variables are present in each observed variable, and do not require measurement [36]. The specific variables are detailed in

Table 3. Components of structural equation model.

Components	Significance
Latent Variables	Unobservable variables that need to be measured using observed variables
Observed Variables	Quantifiable variables used to measure latent variables, obtained through direct observation or objective measurement based on the actual situation
Error Variables	Unmeasured variables representing the errors in latent variables and observed variables, as well as the random variation errors in the model
Exogenous Variables	Independent variables that influence other variables without being influenced by other variables
Endogenous Variables	Dependent variables that are influenced by other variables
Measurement Model	A model that represents the relationship between latent variables and observed variables
Structural Model	A model that represents the structural relationships among latent variables
Path Coefficients	Coefficients representing the relationships between latent variables

.

SEM is widely used in sociology and management science [37]. It differs from traditional multivariate analysis methods in the following ways: ① it can simultaneously handle multiple interacting dependent variables, providing a more comprehensive description of the research problem [38]; ② it allows for the presence of measurement errors in both independent and dependent variables, and can utilize covariance among variables to improve model fit; ③ it estimates the structure and relationships of variables simultaneously, incorporating both observed and latent variables into a single model for estimation; and ④ it allows for greater flexibility in model specification, accommodating the analysis of the impact of a single variable on multiple variables, or the analysis of multiple-order variables. In SEM, hypotheses are proposed based on the research problem, and parameter estimates are obtained through the analysis of latent variables. This provides information about the relationships and driving paths between multiple variables, allowing for the evaluation of the fit between the conceptual model and empirical observation data [39].

SEM can be further divided into two types: covariance-based structural equation modeling (CB-SEM), and partial least squares structural equation modeling (PLS-SEM) [40]. PLS-SEM has higher stability, and is suitable for research with small sample sizes. It takes into account the explained variance of the dependent variables, and is suitable for studying the relationships between multiple groups of variables. Additionally, PLS-SEM effectively prevents the occurrence of multicollinearity issues among measurement variables.

2.3.1. The Principle of Mechanism Analysis in Driving Factors

The SEM used for mechanism analysis of driving factors consists of two parts: the measurement model and the structural model. The measurement model is composed of latent variables and observed variables, as shown in Figure 1. Equations (1) and (2) provide an explanation of the measurement model.

(1)

Measurement model.

$$X={\lambda }_1\xi +\delta$$

(1)

$$Y={\lambda }_2\eta +\epsilon$$

(2)

In the model, X and Y represent reflective indicators of exogenous latent variable $\xi$ and endogenous latent variable $\eta$, respectively; $\delta$ and $\epsilon$ represent the error terms in the measurement model for X and Y, respectively; and ${\lambda }_1$ and ${\lambda }_2$ represent the correlation coefficient matrices between the observed variables and X and Y, respectively.

(2)

Structural model

The structural model consists of exogenous variables and endogenous variables, as shown in Figure 2, and is used to test the hypothesized relationships among latent variables. Equation (3) provides an explanation of this model.

$$\eta =\beta \eta +\gamma \xi +\zeta$$

(3)

where $\eta$ represents the path coefficients between exogenous variables and endogenous variables, $\beta$ represents the path coefficients among endogenous variables, $\gamma$ represents the correlation coefficient matrix between exogenous variables and endogenous variables, and $\zeta$ represents the residual terms of the structural equation.

Figure 5 illustrates a schematic diagram of a structural equation model consisting of four latent variables and twelve observed variables.

According to Equations (1)–(3), the relationships between variables in the schematic diagram of the structural equation model are represented as matrix equations. Equations (4) and (5) represent the matrix equations for the measurement model, while Equation (6) represents the matrix equation for the structural model.

$$\left[\begin{array}{c}{x}_1\\ x_2\\ x_3\\ x_4\\ x_5\\ x_6\end{array}\right]=\left[\begin{array}{cc}{\lambda }_{x11}& 0\\ {\lambda }_{x21}& 0\\ {\lambda }_{x31}& 0\\ 0& {\lambda }_{x42}\\ 0& {\lambda }_{x52}\\ 0& {\lambda }_{x62}\end{array}\right]\left[\begin{array}{c}{\xi }_1\\ {\xi }_2\end{array}\right]+\left[\begin{array}{c}{\delta }_1\\ {\delta }_2\\ {\delta }_3\\ {\delta }_4\\ {\delta }_5\\ {\delta }_6\end{array}\right]$$

(4)

$$\left[\begin{array}{c}{y}_1\\ y_2\\ y_3\\ y_4\\ y_5\\ y_6\end{array}\right]=\left[\begin{array}{cc}{\lambda }_{y11}& 0\\ {\lambda }_{y21}& 0\\ {\lambda }_{y31}& 0\\ 0& {\lambda }_{y42}\\ 0& {\lambda }_{y52}\\ 0& {\lambda }_{y62}\end{array}\right]\left[\begin{array}{c}{\eta }_1\\ {\eta }_2\end{array}\right]+\left[\begin{array}{c}{\epsilon }_1\\ {\epsilon }_2\\ {\epsilon }_3\\ {\epsilon }_4\\ {\epsilon }_5\\ {\epsilon }_6\end{array}\right]$$

(5)

$$\left[\begin{array}{c}{\eta }_1\\ {\eta }_2\end{array}\right]=\left[\begin{array}{cc}0& 0\\ {\beta }_{21}& 0\end{array}\right]+\left[\begin{array}{c}{\eta }_1\\ {\eta }_2\end{array}\right]+\left[\begin{array}{cc}{\gamma }_{11}& {\gamma }_{12}\\ {\gamma }_{21}& {\gamma }_{22}\end{array}\right]\left[\begin{array}{c}{\xi }_1\\ {\xi }_2\end{array}\right]+\left[\begin{array}{c}{\zeta }_1\\ {\zeta }_2\end{array}\right]$$

(6)

2.3.2. Modeling Steps

The modeling steps of SEM can be divided into six parts: theoretical assumptions, model construction, model fitting, model evaluation, model modification, and model interpretation [41].

(1)

Theoretical assumptions: Review relevant literature and summarize theoretical assumptions. Establish directed relationships between latent variables and corresponding observed variables, and set up an initial theoretical model;

(2)

Model construction: Determine the relationships between different variable combinations and select a model that provides a simple explanation for a greater number of variables. Express the measurement model and structural model through a system of equations or a path diagram;

(3)

Model fitting: Estimate the parameters of the variables using collected data and information. The better the fit between the covariance matrix and the sample covariance matrix in SEM, the better the model fit. Common fit indices include chi-square value, goodness-of-fit index (GFI), and root mean square error of approximation (RMSEA), as shown in Table 4;

(4)

Model evaluation: Determine if the output indicator values meet the predefined fitness criteria of the model. This evaluation includes overall model evaluation and structural fit evaluation. The former assesses the fit between the sample data and the theoretical model, i.e., whether the observed variables effectively reflect the latent variables. The latter tests the causal relationships proposed by the hypotheses. The model should meet the criteria of the measurement equation errors having a mean of zero, the structural equation residuals having a mean of zero, and the errors being uncorrelated with the factors;

(5)

Model modification: If the fit indices indicate poor model fit, model modification is required to improve the fitness. Simultaneously, the adequacy of the modified model is assessed by connecting theoretical results with practical significance;

(6)

Model interpretation: interpret the meaning of the relevant data and validate the previously proposed hypothesis relationships.

Table 4. Common measurement index of SEM.

Category	Indicator	Standard
Relative Fit Indices	Goodness-of-Fit Chi-Square Test (Γ)	Acceptance range (2, 5), Fit is considered good when the value is less than 2
	Comparative Fit Index (CFI)	Acceptance range (0, 1), Closer to 1 indicates better fit
	Incremental Fit Index (IFI)
Absolute Fit Indices	Root Mean Square Error of Approximation (RMSEA)	For RMSEA, a value of ≤0.05 indicates acceptable fit
	Goodness-of-Fit Index (GFI)	Acceptance range (0, 1), Closer to 1 indicates better fit
Parsimonious Fit Index	Parsimonious Goodness-of-Fit Index (PGFI)	Acceptance range (0.5, 1), Closer to 1 indicates better fit.

Table 4. Common measurement index of SEM.

Category	Indicator	Standard
Relative Fit Indices	Goodness-of-Fit Chi-Square Test (Γ)	Acceptance range (2, 5), Fit is considered good when the value is less than 2
	Comparative Fit Index (CFI)	Acceptance range (0, 1), Closer to 1 indicates better fit
	Incremental Fit Index (IFI)
Absolute Fit Indices	Root Mean Square Error of Approximation (RMSEA)	For RMSEA, a value of ≤0.05 indicates acceptable fit
	Goodness-of-Fit Index (GFI)	Acceptance range (0, 1), Closer to 1 indicates better fit
Parsimonious Fit Index	Parsimonious Goodness-of-Fit Index (PGFI)	Acceptance range (0.5, 1), Closer to 1 indicates better fit.

Specific details are shown in Figure 6.

Meanwhile, the main methods for measuring the fit of a structural equation model are primarily based on the indices listed in Table 4.

The current study selected SmartPLS 3.0 software for model construction and analysis, which offers three advantages. First, it imposes fewer restrictions on the distribution of sample data, and does not have strict requirements for variable data, making it suitable for situations where data is skewed. Second, it provides flexibility in terms of sample size requirements, making it applicable for small- to medium-sized sample studies. Third, it comprehensively captures the relationship between latent variables and observed variables, allowing for both reflective and formative indicators. Therefore, the use of SmartPLS software emphasizes exploratory and explanatory analysis, making it conducive for hypothesis testing and achieving research objectives.

2.4. Survey Questionnaire Design and Analysis

2.4.1. Questionnaire Design

The design of the questionnaire followed the research literature-expert interview design path. Firstly, relevant studies on land use efficiency evaluation under urban renewal were referred to and, according to the basic situation and development characteristics of the Harbin-Changchun urban agglomeration, 24 measurement items, including the common problem of the comprehensive carrying capacity and the characteristic problems of the Harbin-Changchun urban agglomeration, are summarized for the questionnaire [42]. Subsequently, expert interviews were conducted by distributing the questionnaire to individuals involved in related research and experts with previous interviews and extensive experience for trial completion and discussion. This was performed to ensure the rationality and understandability of the design of observed variable items for the respondents. The collected questionnaires were compiled, and the average score difference of each respondent for each item was analyzed. A larger difference indicates a better explanatory power of the questionnaire item. Items with low explanatory power and poor suitability were removed, resulting in a final selection of 22 items.

The formal “Survey Questionnaire on Drivers of Comprehensive Carrying Capacity of Land” consists of three parts: (1) questionnaire guidelines, explaining the main purpose of this research, relevant concepts, and professional terminology to facilitate respondents’ understanding of the content of the questionnaire items; (2) respondent demographic information, including occupation category related to the study, educational level, years of experience in relevant industries, etc.; and (3) statements on the drivers of the comprehensive carrying capacity of land.

A Likert-type five-point scale was used for measurement, to assess the respondents’ agreement level with each item. The scale ranged from “strongly disagree” to “strongly agree”, with five levels of perception options, assigned values from 1 to 5 in increasing order.

2.4.2. Data Collection and Analysis

Due to the importance of determining the driving factors of the land carrying capacity, ensuring the accuracy of the research is the foundation for subsequent studies. To achieve this, participant selection was a key consideration during the questionnaire survey phase. The survey was conducted through three main methods and channels: Firstly, face-to-face interviews were conducted with experts in relevant fields and industries, to improve the quality of the questionnaire and ensure its rapid collection. Secondly, online tools, such as email, WeChat, and QQ, were used to distribute the survey to respondents who were not available for face-to-face interviews. Thirdly, an online survey was conducted using the questionnaire platform provided by the website ( https://www.wjx.cn accessed on 5 March 2023) to expand the sample size and eliminate spatial and geographical limitations. The data collection process lasted for two months, during which a total of 363 questionnaires were collected, and 286 of them were considered valid, resulting in a valid response rate of 78.79%.

In order to achieve a balanced approach between the scientificity and efficiency of social surveys, the census-style expert survey method can essentially be replaced by sampling surveys. Numerous practices have shown that, under the premise of a reasonable sample size, the accuracy and stability of result predictions can be ensured. Based on this, considering the time and spatial dimensions of this paper, the selected sample size is sufficient to meet the requirements of result accuracy and stability.

Due to the reliance on a self-administered questionnaire in this paper, where both the independent and dependent variables are provided by the same respondents, there is a potential for common method bias (CMB). CMB can affect the validity of survey results and the reliability of data. Following Podsakoff and P.M, who proposed causes and solutions for CMB, this paper primarily employs procedural remedies, methodological control, and statistical remedies to mitigate CMB [43].

(1)

Procedural remedies

Cross-source data were used in the questionnaire process, i.e., respondents included managers who could make decisions on corporate low-carbon transition behaviors, as well as grassroots employees who responded to corporate low-carbon transition decisions, so as to more comprehensively reflect corporate willingness to transition and perceived factors.

(2)

Methodological control

Firstly, the questionnaires were distributed in various forms, including email, telephone, face-to-face interviews, and online questionnaires, to overcome self-response bias. Secondly, anonymity was used to reduce respondents’ concerns. Finally, it was clearly stated in the interviews and questionnaires that there were no correct or incorrect answers, to maximize the authenticity and sexiness of the data.

(3)

Statistical remedies

To address and validate the potential common method bias (CMB) arising from the questionnaire survey, the researcher can employ Harman’s one-factor test. This test allows for the examination of CMB by analyzing the basic information section of the questionnaire, which includes demographic characteristics of the sample population.

In the analysis of influencing factors, the demographic characteristics of the data sample play a crucial role. Stern suggests that relevant data on demographic characteristics can predict environmental behavior [44]. When analyzing the behavior of participants in certain environmental issues, scholars have drawn some insightful conclusions through the analysis of individual demographics. For example, women are more likely than men to engage in environmental participation, and income levels can also influence individual involvement in environmental issues [45]. And factors such as age and education level can impact public participation in waste recycling management behaviors [46]. Therefore, in constructing the model of influencing factors and designing the questionnaire, it is necessary to include relevant indicators of individual demographics, and analyze the data accordingly. The specific information of the respondents is presented in

Table 5. Information of respondents with valid questionnaires.

Respondent Basic Information		Frequency	Percentage (%)
Occupation Category	University Experts	79	27.62
	Researchers from Institutes and Related Fields	75	26.22
	Land Resource Planners	37	12.93
	Land Development Technicians	68	23.78
	Urban Planners	27	9.44
Education Level	Doctorate and above	46	16.08
	Master’s degree	72	25.17
	Bachelor’s degree	143	50
	Associate degree and below	25	8.75
Years of Experience in the Construction Industry	10 years and above	62	21.68
	6–10 years	85	29.72
	3–5 years	88	30.8
	Less than 3 years	51	17.8
Age	Below 25 years old	69	24.13
	26–35 years old	98	34.27
	36–45 years old	62	21.68
	46–55 years old	37	12.94
	Over 55 years old	20	6.98
Participation in Project Scale	Large-scale projects	98	34.27
	Medium-scale projects	109	38.11
	Small-scale projects	79	27.62

.

Among the respondents with valid questionnaires, the majority consisted of university experts and researchers from institutes or related fields, accounting for 53.84% of the total sample. In terms of project participation, a significant portion (38.11%) had experience in medium-scale projects. Most respondents had been working in relevant industries for 5–10 years. They possessed a certain level of professional knowledge in land planning and development, and had accumulated relevant experience in urban renewal and construction activities through their involvement in medium-sized and larger projects. The selection of respondents for the questionnaire survey was scientifically and reasonably designed, resulting in a representative sample.

3. Results Analysis

3.1. Reliability and Validity Tests

Before conducting the analyses, the reliability and validity of the data were checked. Reliability was assessed through factor loadings, where an indicator loading above 0.7 and significant at the 5% level indicates that the observed variables explain more than 50% of the variance in their respective latent variables. Internal consistency and construct reliability were evaluated using Cronbach’s alpha (CA) coefficient and composite reliability (CR), respectively. The recommended threshold for both CA and CR is 0.7. Higher CA values indicate that all observed variables measuring a latent variable have the same range and meaning, indicating good internal consistency. When CR is equal to or greater than 0.7, it indicates good reliability of the measurement model. The measurement results indicate that the data have high reliability and credibility. As shown in

Table 6. Reliability and validity test of measurement model.

Latent Variable	Indicator	Indicator Loadings	t-Value	CA	CR	AVE	KMO
UD	UD1	0.907	63.631 **	0.907	0.935	0.782	0.836
	UD2	0.935	97.418 **
	UD3	0.928	89.036 **
	UD4	0.877	39.037 **
SE	SE1	0.802	30.670 **	0.806	0.873	0.631	0.766
	SE2	0.777	23.898 **
	SE3	0.787	22.196 **
	SE4	0.812	42.966 **
UR	UR1	0.850	43.435 **	0.846	0.896	0.684	0.808
	UR2	0.815	34.701 **
	UR3	0.821	34.456 **
	UR4	0.823	34.464 **
UE	UE1	0.877	46.237 **	0.833	0.900	0.750	0.706
	UE2	0.821	29.156 **
	UE3	0.897	68.934 **
ID	ID1	0.831	29.614 **	0.841	0.893	0.677	0.799
	ID2	0.791	25.449 **
	ID3	0.806	24.381 **
	ID4	0.862	39.419 **
DPMC	DPMC1	0.826	31.025 **	0.856	0.913	0.778	0.719
	DPMC2	0.910	60.956 **
	DPMC3	0.907	52.645 **
LCCC	LCCC1	0.911	65.903 **	0.903	0.939	0.837	0.753
	LCCC2	0.921	84.523 **
	LCCC3	0.913	74.088 **

Note: *** indicates p < 0.001; ** indicates p < 0.01; * indicates p < 0.05.

, all indicator loadings exceed the critical value of 0.7, and are significant at the 0.01 level. The Cronbach’s alpha values range from 0.806 to 0.907, and the CR values range from 0.873 to 0.939, all meeting the test criteria (>0.7), indicating good reliability of the measurement model. The average variance extracted (AVE) values are all greater than 0.5, indicating good convergent validity of the measurement model.

Validity is divided into convergent validity and discriminant validity. Convergent validity measures the degree of convergence among items within the same dimension. Discriminant validity refers to the degree of differentiation between different dimensions or constructs. Convergent validity is assessed using composite reliability (CR) and average variance extracted (AVE).

Table 7. Correlation coefficients of latent variables in the measurement model.

	UD	SE	UE	ID	UR	DPMC	LCCC
UD	0.884
SE	0.535	0.795
UR	0.483	0.588	0.827
ID	0.471	0.539	0.539	0.823
UE	0.234	0.363	0.370	0.305	0.866
DPMC	0.435	0.528	0.560	0.443	0.321	0.882
LCCC	0.545	0.570	0.567	0.529	0.385	0.555	0.915

Note: *** represents p < 0.001; ** represents p < 0.01; * represents p < 0.05.

presents the square root of AVE, which is higher than the correlation coefficients with other latent variables.

Table 8. The cross loading of observe variables in the measurement model.

	UD	SE	UE	UR	ID	DPMC	LCCC
UD1	0.879	0.484	0.415	0.190	0.419	0.400	0.433
UD2	0.872	0.433	0.437	0.198	0.418	0.411	0.513
UD3	0.903	0.498	0.427	0.238	0.423	0.364	0.454
UD4	0.882	0.478	0.431	0.203	0.408	0.364	0.523
SE1	0.418	0.802	0.536	0.324	0.452	0.463	0.448
SE2	0.354	0.777	0.511	0.283	0.415	0.430	0.430
SE3	0.366	0.787	0.408	0.267	0.401	0.407	0.450
SE4	0.543	0.812	0.420	0.280	0.442	0.383	0.481
UR1	0.420	0.481	0.850	0.321	0.445	0.508	0.512
UR2	0.389	0.503	0.815	0.304	0.455	0.389	0.469
UR3	0.395	0.499	0.821	0.313	0.440	0.480	0.477
UR4	0.395	0.464	0.823	0.285	0.444	0.475	0.411
UE1	0.204	0.268	0.310	0.877	0.256	0.267	0.330
UE2	0.202	0.335	0.310	0.821	0.259	0.226	0.294
UE3	0.204	0.338	0.340	0.897	0.275	0.332	0.370
ID1	0.392	0.395	0.403	0.179	0.831	0.349	0.416
ID2	0.387	0.465	0.427	0.247	0.791	0.340	0.451
ID3	0.372	0.441	0.477	0.331	0.806	0.395	0.437
ID4	0.400	0.467	0.462	0.238	0.862	0.372	0.434
DPMC1	0.338	0.436	0.434	0.232	0.349	0.826	0.479
DPMC2	0.403	0.479	0.527	0.301	0.412	0.910	0.478
DPMC3	0.405	0.480	0.517	0.312	0.409	0.907	0.510
LCCC1	0.478	0.472	0.522	0.390	0.504	0.514	0.911
LCCC2	0.507	0.539	0.511	0.329	0.460	0.489	0.921
LCCC3	0.509	0.552	0.522	0.338	0.488	0.518	0.913

Note: *** represents p < 0.001; ** represents p < 0.01; * represents p < 0.05.

displays the highest factor loadings of each cross-loading indicator on its corresponding latent variable. Based on these criteria, the model demonstrates good discriminant validity. Overall, the model meets the standards for reliability and validity testing, indicating its effectiveness and reliability.

3.2. Model Goodness-of-Fit Test

3.2.1. The Measurement Model

To explore the relationships and paths of the influencing the factors of comprehensive land carrying capacity, the goodness of fit of the conceptual model was examined. The fit indices, including the comparative fit index (CFI), incremental fit index (IFI), and goodness of fit index (GFI), were evaluated, and the results are presented in

Table 9. Fitting degree test.

Fitting Index	Test Value	Whether It Meets the Standard
Γ	1.313	Yes
CFI	0.981	Yes
IFI	0.924	Yes
RMSEA	0.033	Yes
GFI	0.916	Yes
PGFI	0.716	Yes

. The CFI, IFI, and GFI all exceeded 0.9, indicating a good fit. The chi-square to degrees of freedom ratio was 1.313, falling within the range of 1 to 3, and less than 2, indicating a good fit. The root mean square error of approximation (RMSEA) value was 0.033, which was lower than the reference value of 0.08. The parsimonious goodness of fit index (PGFI) was greater than 0.5. Therefore, the model exhibited a satisfactory fit to the questionnaire data, and was well-aligned with the practical issues.

3.2.2. The Structural Model

Structural model validation is the process of verifying the predictive ability and relationships between variables in a model after meeting reliability and validity standards in the measurement model. The effect size f² is an indicator of the degree to which exogenous latent variables affect endogenous latent variables. A value of f² greater than or equal to 0.02 indicates the presence of an effect of exogenous latent variables on endogenous latent variables. As shown in

Table 10. Effect size f².

Path	f2	Path	f2
UD → LCCC	0.060	UE → LCCC	0.025
UD → SE	0.401	UE → ID	0.020
SE → LCCC	0.020	UE → DPMC	0.115
SE → UE	0.152	ID → LCCC	0.028
SE → ID	0.058	UD → ID	0.039
UR → LCCC	0.021	DPMC → LCCC	0.052
UR → ID	0.073

, the f² values for each path are greater than or equal to 0.02, indicating that the latent variables along these paths have an impact on the endogenous latent variables.

3.3. Hypothesis Testing

3.3.1. Results of the Path Coefficient Test

SmartPLS is used to test path coefficients. This study proposes 13 hypotheses, and Figure 7 and

Table 11. Structural equation model direct effect path coefficient and significance test.

Path	Path Coefficient	t-Values	p-Values	Result
H1: UD → LCCC	0.214	4.001	***	Support
H2: UD → SE	0.535	10.346	***	Support
H3: UD → ID	0.188	3.485	***	Support
H4: SE → LCCC	0.140	2.590	**	Support
H5: SE → UE	0.363	5.367	***	Support
H6: SE → ID	0.252	3.463	***	Support
H7: UR → LCCC	0.141	2.542	**	Support
H8: UR → ID	0.275	4.201	***	Support
H9: UE → LCCC	0.121	2.610	**	Support
H10: UE → ID	0.067	0.943	Non-significant	No Support
H11: UE → DPMC	0.321	5.367	***	Support
H12: ID → LCCC	0.149	2.974	**	Support
H13: DPMC → LCCC	0.203	3.593	***	Support

Note: *** represents p < 0.001; ** represents p < 0.01; * represents p < 0.05.

represent the results of the structural model and the significance testing of path coefficients. At a significance level of 0.05, 12 hypotheses are supported, while 1 hypothesis is not supported. The path coefficients indicate the relationships and the degree of influence between variables in the structural model.

Urban construction scale, socio-economic factors, urban renewal, urban ecology, infrastructure construction, and urban disaster prevention and mitigation capacity all have a significant impact on the comprehensive land carrying capacity. The top three factors with the greatest influence are urban construction scale (β = 0.214, t = 4.001), urban disaster prevention and mitigation capacity (β = 0.203, t = 3.593), and infrastructure construction (β = 0.149, t = 2.974). Therefore, hypotheses H1, H4, H7, H9, H12, and H13 are supported. Urban construction scale, socio-economic factors, and urban renewal all have a significant impact on infrastructure construction, supporting hypotheses H3, H6, and H8. Among them, urban renewal (β = 0.275, t = 4.201) has the greatest impact on infrastructure. Urban ecology has a lower impact on infrastructure, with a t-value of 0.943, failing the significance test, thus hypothesis H10 is not supported. The path coefficient (β) of urban construction scale on socio-economic factors is 0.535, with a t-value of 10.346, showing a significant positive influence, supporting hypothesis H2. Socio-economic factors have a significant impact on urban ecology, with a path coefficient (β) of 0.363 and a t-value of 5.367, supporting hypothesis H5. Additionally, urban ecology has a significant impact on urban disaster prevention and mitigation capacity, with a path coefficient of 0.321, passing the significance test and validating hypothesis H11.

3.3.2. Intermediation Effects

The impact paths of each driving factor on the comprehensive land carrying capacity include both direct and indirect effects through mediating variables [47]. Therefore, it is essential to test the existence of the mediating effects. Using the bootstrapping method with a confidence interval test in SmartPLS 3.0, a sample size of 5000 and a confidence level of 0.95 are selected. The bias-corrected confidence interval test in the bootstrapping method shows that when the intersection of the upper and lower limits does not include zero, the mediating effect is significant.

Table 12. Intermediation effect coefficient test of structural model.

NO.	Path	Indirect Effect	t-Values	Confidence Interval	Results
1	UR → DPMC → LCCC	0.065	3.009	[0.027, 0.111]	Accept
2	SE → UE → LCCC	0.044	2.325	[0.009, 0.084]	Accept
3	UD → SE → LCCC	0.075	2.523	[0.019, 0.136]	Accept
4	UD → ID → LCCC	0.028	2.312	[0.007, 0.054]	Accept
5	SE → ID → LCCC	0.038	2.168	[0.009, 0.076]	Accept
6	UR → ID → LCCC	0.041	2.308	[0.011, 0.080]	Accept

shows the results of the variables in the model with mediating effects on LCCC (comprehensive land carrying capacity). There are six mediating paths: Two paths from the dimension of urban renewal (UR) have mediating effects, namely UR → DPMC → LCCC (0.065) and UR → ID → LCCC (0.041). Two paths from the dimension of socio-economic factors (SE) are SE → UE → LCCC and SE → ID → LCCC, with mediating effects of 0.044 and 0.038, respectively. From the dimension of urban construction scale (UD), the mediating paths are UD → SE → LCCC (0.075) and UD → ID → LCCC (0.028).

3.3.3. Correlation Analysis of Indicators

Correlation analysis can explain the degree of correlation between variables. In order to further explain the relationship between variables, Pearson correlation analyses were used to analyze the relationships between variables in this study. The result is shown in

Table 13. Results of correlation analyses.

	UD	SE	UR	UE	ID	DPMC
UD	1
SE	0.623 **	1
UR	0.493 **	0.478 **	1
UE	0.556 **	0.562 **	0.575 **	1
ID	0.529 **	0.527 **	0.440 **	0.513 **	1
DPMC	0.325 **	0.351 **	0.314 **	0.303 **	0.339 **	1
Mean value	3.35	3.20	3.18	3.46	3.78	3.71
Standard deviation	0.88	0.89	1.07	0.94	0.82	0.91

Note: ** represents p < 0.01.

.

Table 13 shows that the correlation coefficients of UE, SE, UR, UE, ID, and DPMC are 0.623, 0.493, 0.556, 0.529, and 0.325, respectively, and the corresponding p values are all less than 0.01, which indicates statistical significance. UE, SE, UR, UE, ID, and DPMC all have significant positive correlations.

3.4. Analysis of Model Results

Table 14. Path of action with significant effect driving factors.

	Path	Direct Effect	Indirect Effect	Total Effect
UD	→ LCCC	0.214		1.096
	→ SE → LCCC		0.131
	→ ID→ LCCC		0.028
	→ SE	0.535
	→ ID	0.188
SE	→ LCCC	0.140		0.861
	→ UE → LCCC		0.068
	→ ID → LCCC		0.038
	→ UE	0.363
	→ ID	0.252
UR	→ LCCC	0.141		0.522
	→ ID → LCCC		0.041
	→ DPMC → LCCC → ID	0.275	0.065
UE	→ LCCC	0.121		0.442
	→ DPMC	0.321
DPMC	→ LCCC	0.203		0.203
ID	→ LCCC	0.149		0.149

displays the effects of various driving factors on the comprehensive land carrying capacity through significance testing (i.e., direct, indirect, and total effects). From the overall effect results, the impacts of the six dimensions on the comprehensive land carrying capacity, from high to low, are as follows: urban construction scale (UD), with a value of 1.096; socio-economic factors (SE), with a value of 0.861; urban renewal (UR), with a value of 0.522; urban ecology (UE), with a value of 0.442; urban disaster prevention and mitigation capacity (DPMC), with a value of 0.203; and infrastructure construction (ID), with a value of 0.149.

• Urban development (UD)

There is a direct correlation between the urban construction scale and the comprehensive land carrying capacity, with a direct effect coefficient of 0.214. Urban construction scale is an important factor influencing comprehensive land carrying capacity. Urban construction activities rely on regional land resources, and a reasonable construction scale promotes optimal land development and utilization [48]. On one hand, urban construction is consistent with land use planning, where urban functions, industrial structure, and economic development trends determine land demand. Strengthening overall land use planning plays a regulatory role in urban spatial layout and development direction [49]. On the other hand, the key to construction activities is the rational use of construction land, which should be coordinated with urban industrial development. It should be based on factors such as population distribution, economic layout, and resource and environmental endowments to achieve optimal land efficiency.

The urban construction scale has an indirect effect on the comprehensive land carrying capacity through two paths: urban construction scale → social economy → comprehensive land carrying capacity, and urban construction scale → infrastructure construction → comprehensive land carrying capacity, with indirect effect values of 0.131 and 0.028, respectively. From the perspective of land resource utilization, an increase in construction scale creates conditions for adjusting the land use structure, promoting the optimization of industrial structure, and improving the quality of economic development [50]. In terms of land use form, the high-density buildings in construction land can accommodate more population and industrial activities, which is the main manifestation of the urban construction scale, leading to high agglomeration economic benefits [51]. From the perspective of land function, urban construction promotes the transition of land from production and ecological functions to primarily carrying functions [52]. Considering the previous influence of the urban construction scale on infrastructure construction, it is inferred that these two indirect paths are valid.

2.

Social economy (SE)

The development of the social economy has a direct pathway and two indirect pathways that influence the comprehensive carrying capacity of land. The total effect is 0.861, which is relatively high among the driving factors. According to the results of the model validation, the social economy has a positive impact on the comprehensive carrying capacity of land. Land resources serve as the foundation and vehicle for economic activities, and as the urban scale expands, economic activities experience exponential growth. Economic growth drives improvements in technology and management levels, guiding the coordinated development of various components of the urban system. In terms of inter-regional land allocation, when the social economy is at a higher level, elastic distribution of construction land based on the degree of urban agglomeration can meet a larger population size, reduce economic development gaps between regions, alleviate urban pressures, maximize the overall benefits of land resources, and enhance the carrying capacity. In terms of spatial and regional optimization, high-quality economic development improves the level of urban management technology, breaks institutional barriers, promotes the formation of multi-center urban agglomerations, and reduces mismatches of resources and factors between and within cities.

The social economy has indirect effects on the comprehensive land carrying capacity through two paths: social economy → urban ecology → comprehensive land carrying capacity, and social economy → infrastructure construction → comprehensive land carrying capacity, with indirect effect values of 0.068 and 0.038, respectively. The social economy along these paths provides guarantees for infrastructure construction and renewal. Considering the impact of urban ecology and infrastructure on the comprehensive land carrying capacity and the significant mediating effects, it can be determined that these paths are valid.

3.

Urban renewal (UR)

Urban renewal has both direct and indirect pathways that affect the comprehensive carrying capacity of land, with a total effect of 0.522. In the direct pathway, the direct effect value is 0.416. According to the model test results, urban renewal has a positive impact on the comprehensive carrying capacity of land, indicating that it promotes land carrying capacity. In the long term, urban renewal activities provide an effective channel for adjusting land use allocation and improving land use efficiency, effectively increasing land supply [53]. The goal of urban renewal is to improve living conditions and enhance land suitability by transforming and reconstructing, thereby creating efficient utilization of land resources and an environment conducive to economic development [54].

In the indirect pathways, firstly, urban renewal has an indirect effect on the comprehensive carrying capacity of land through infrastructure construction, with the pathway being urban renewal → infrastructure construction → comprehensive carrying capacity of land. The model test results show that urban renewal has a positive impact on infrastructure construction. According to the current policy orientation in China, many cities, including Beijing, Shanghai, Shenzhen, and Guangzhou, require renewal projects to prioritize the increase in public service facilities, with the main focus on municipal infrastructure construction and improvement. Considering the goals of urban development stages and spatial layout characteristics, urban renewal projects effectively enhance the potential of existing land and optimize functional layout [55]. Secondly, urban renewal has an indirect effect on the comprehensive carrying capacity of land through disaster prevention and mitigation capabilities, and the mediating effect is significant. The pathway is urban renewal → urban disaster prevention and mitigation capabilities → comprehensive carrying capacity of land. Urban renewal activities are complex and involve multiple stakeholders, but their positive promotion of urban disaster prevention and mitigation capabilities is evident [56]. Considering the positive impact of urban disaster prevention and mitigation capabilities on the comprehensive carrying capacity of land, it can be inferred that urban renewal can have a positive effect on the land carrying capacity through urban disaster prevention and mitigation capabilities.

4.

Urban ecology (UE)

In the study of land carrying capacity and territorial spatial optimization, urban ecological environment plays a crucial role [57]. Starting from urban ecology, there is a direct pathway that influences the comprehensive carrying capacity of land. The model results indicate a direct correlation between urban ecology and the comprehensive carrying capacity of land. The impact of the ecological environment on land carrying capacity is evident, as a stable ecological environment serves as the foundation for sustainable urban development. From the perspective of urban renewal and construction activities, rational allocation of resources and appropriate development intensity can effectively reduce the consumption of the ecological environment and alleviate the imbalance between humans and the environment [58]. In terms of disaster prevention and mitigation capabilities, the city, as an open artificial ecosystem, ensures the preservation of regional topography, hydrological systems, urban vegetation, and other factors, which can reduce the risks associated with disaster-causing factors, thereby significantly contributing to the enhancement of the land carrying capacity [59].

5.

Urban disaster prevention and mitigation capacity (DPMC)

According to the examination results of the structural model, urban disaster prevention and mitigation capacity directly affect the comprehensive carrying capacity of land. Currently, cities face various complex natural disasters and frequent human-induced disasters. Enhancing urban disaster prevention and mitigation capacity has two main benefits. Firstly, it can prevent the deepening of disaster impacts, reduce the scope of influence, and ensure the orderly operation of urban activities. Secondly, improving the ability to prevent disaster risks can effectively restrict the uncontrolled expansion of urban land, reduce human intervention and impact on natural ecological spaces, and have a positive impact on the comprehensive carrying capacity of land [60,61].

6.

Infrastructure development

The hypothesis that infrastructure construction has a positive impact on the comprehensive carrying capacity of land is validated in the structural model. Urban infrastructure serves as a crucial foundation for regional development, and has a strong reliance on land resources. Establishing a modern infrastructure system is beneficial for supporting land planning and utilization, as well as optimizing the spatial layout of the national territory [62]. Existing studies indicate that improving the standards for construction land and conducting efficiency assessments of infrastructure construction are helpful in preventing conflicts between land supply and demand, and minimizing environmental damage. These measures can effectively enhance the comprehensive carrying capacity of land [63,64].

4. Discussion and Conclusions

4.1. Discussion

• The research results indicate that all six dimensions in the conceptual model have a direct positive impact on the land carrying capacity. In terms of direct effects, the influencing factors are ranked in descending order of magnitude as follows: urban development, urban disaster prevention and mitigation capacity, infrastructure development, urban renewal, social economy, and urban ecology. In terms of overall effects, factors are ranked in descending order of magnitude as follows: urban development, social economy, urban ecology, urban renewal, urban disaster prevention and mitigation capacity, and infrastructure development;
• According to the conceptual model, the paths with significant correlations with the integrated land carrying capacity are determined as UD → LCCC, UD → SE → LCCC, UD → ID → LCCC, SE → LCCC, SE → UE → LCCC, SE → ID → LCCC, UR → LCCC, UR → ID → LCCC, UR → DPMC → LCCC, UE → LCCC, DPMC → LCCC, and ID → LCCC. The paths of influence relationships between the drivers of each dimension are UD → SE, UD → ID, SE → UE, SE → ID, UR → ID, and UE → DPMC;
• The United States, Canada, Australia, Malaysia, and other countries have conducted studies on the comprehensive land carrying capacity with their resource endowments and social development patterns. They have identified the mechanisms through which factors such as population, economy, and natural resources influence the land carrying capacity. This paper aligns with the theoretical framework of the AEZ method. It integrates existing indicator systems based on different countries, and combines them with China’s economic development characteristics and urban renewal conditions to investigate the regional land carrying capacity. The findings are consistent with the quantitative evaluation results of the AEZ method, and also reveal the driving mechanisms of land carrying capacity under China’s urban development characteristics. Therefore, this paper has broad applicability in the field, and provides a unique perspective within the context of China’s economic development. It has regional relevance, and enriches the existing research outcomes in this field;
• Through the findings of this paper, it can be observed that, from the perspective of urban renewal, multiple dimensions have a positive impact on the land carrying capacity, and there are interactions and mediating effects among the factors. These conclusions align with the research conducted by Tian G et al. [65] and Irwin E G et al. [66], which indicate that urban renewal activities have a promoting effect on land carrying capacity. Through the above analyses, the ultimate goal of this paper is to study the driving mechanism of comprehensive land carrying capacity from the perspective of urban renewal, taking a specific region in China as the research object, and to provide theoretical and technical support with practicality and operability for various aspects of urban construction, urban resource allocation, and land resource conservation.

In the existing literature, the factors affecting the comprehensive carrying capacity of land mainly include population structure, economic development, etc. [21]. This paper conducted a survey on the Harbin-Changchun urban agglomerations, and optimized and improved them from the perspectives of urban renewal, urban scale and urban resilience on the basis of existing studies. The paper carried out a study on the impact indicators and driving mechanism of the land resource carrying capacity of the Harbin-Changchun urban agglomeration from the perspective of urban renewal, providing support for the study of the quantitative evolutionary trend of the land resource carrying capacity, the identification of the main indicators affecting the land resource carrying capacity, and the improvement in the land resource carrying capacity.

4.2. Conclusions

This study proposes research hypotheses based on the perspective of urban renewal regarding the driving factors and interrelationships among factors of comprehensive land carrying capacity. Firstly, the research variables are determined, and a conceptual model of driving factors for comprehensive land carrying capacity is constructed, including the association between six dimensions of driving factors and the comprehensive land carrying capacity. Secondly, a survey questionnaire suitable for research needs is developed, and statistical analyses are conducted to examine the measurement model, structural model, and mediating effects among factors using the PLS-SEM method. Lastly, the study summarizes the paths of the factors identified through empirical testing, and analyzes the mechanisms of the driving factors that influence the comprehensive land carrying capacity.

The conceptual model proposed in this study demonstrates that urban construction scale, socio-economic factors, urban renewal, urban ecology, infrastructure construction, and urban disaster prevention and mitigation serve as important driving factors and mediating variables, all positively affecting comprehensive land carrying capacity. There are interactions among the driving factors. For example, urban construction scale has a positive influence on socio-economic factors and infrastructure construction, with the largest effect on socio-economic factors. Socio-economic factors have a positive influence on urban ecology and infrastructure construction. Urban renewal has a positive influence on infrastructure construction. Urban ecology has a positive influence on infrastructure construction and urban disaster prevention and mitigation capacity. Urban renewal, urban construction scale, and socio-economic factors exhibit varying levels of mediating effects on comprehensive land carrying capacity, ranging from strong to weak.

Considering the influence of other aspects on the comprehensive land carrying capacity in the perspective of urban renewal, there may be human factors. Due to the lack of support in the literature, and the limited availability of data, the limitations of this paper are the lack of indicators related to the cultural promotion of urban land resources and urban management. In future research, the influence of urban citizens on land carrying capacity from the perspective of urban governance should be further considered. In addition, further analysis can be carried out from a broader perspective on the relationship between factors in different dimensions, so as to explore a more specific mechanism of land carrying capacity.