Research on Urban Sustainability Based on Neural Network Models and GIS Methods

Abstract

Ecologically sustainable urban design plays a pivotal role in mitigating climate change. This study develops an indicator group consisting of urban ecological emergy, land use change, population density, ecological services, habitat quality, enhanced vegetation index, carbon emissions, and carbon storage to assess urban sustainability. By leveraging a dataset from 2000 to 2020, we employ a neural network to predict emergy sustainability indicators over a time series, projecting the sustainable status of Xuzhou City from 2020 to 2050. The findings indicate that urbanization has led to significant changes in land use, population distribution, ecological service patterns, habitat quality degradation, vegetation fragmentation, and fluctuating carbon dynamics. Cropland constitutes the predominant land type (90.6%), followed by built-up land (8.49%). The neural network predictions suggest that Xuzhou City’s sustainable status is subject to volatility (15–20%), with stability expected only as the city matures into a developed urban area. This research introduces a novel approach to urban sustainability analysis and provides insights for policy development aimed at fostering sustainable urban growth.

1. Introduction

With the intensification of the global trend of climate warming, abnormal environmental phenomena are increasingly prominent worldwide. Cities, as human settlements, also face severe challenges from environmental changes. Factors such as human construction activities, waste emissions, high population density, air pollution, the reduction of biodiversity, and the urban heat island effect make urban environments more vulnerable, with ecological sustainability facing degradation [1]. In the face of these challenges, assessing the ecological sustainability of cities has become particularly crucial, as it is the foundation and prerequisite for improving urban ecological environments.

Xuzhou, as the central city of the Huaihai Economic Zone, boasts a geographical advantage that bridges the east and west and connects the north and south. It is a national historical and cultural city and a major national comprehensive national traffic hub. As an old industrial base and a resource-based city, Xuzhou has historically faced prominent ecological and environmental issues. With the development of the economy and society, ecological improvement has become an inevitable choice. This not only concerns the quality of life of its citizens but is also necessary for achieving sustainable development and the construction of an ecological city. However, Xuzhou faces certain challenges in terms of ecological sustainability. According to national environmental monitoring data, Xuzhou’s industrial pollution is relatively severe, with the air quality index frequently exceeding national standards, especially during the winter heating season with coal-fired boilers. In addition, some rivers and lakes do not meet environmental standards. Therefore, conducting continuous assessment research on ecological sustainability with Xuzhou as a case study is of great significance.

Literature Review

Research on urban sustainability is a multifaceted and intricate endeavor, encompassing a wide array of factors that have been meticulously analyzed and explored by numerous scholars from varying perspectives [2]. The study of ecological security has illuminated the importance of preserving natural ecosystems and their functions [3,4]. Ecological footprint research has quantified human impact on the environment [5], providing insights into sustainable consumption patterns [6]. The concept of urban symbiosis [7] explores the interdependence between urban systems and their natural surroundings, fostering a holistic approach to urban development. The construction of ecological civilization [8] emphasizes the integration of societal progress with environmental sustainability. Ecological networks [9] research has highlighted the connectivity and importance of green spaces within urban areas. Environmental quality [10] studies have focused on the air, water, and soil conditions necessary for a healthy urban environment. The development of eco-intelligent cities integrates advanced technologies to create sustainable and smart urban environments [11]. Ecological livability assessment [12] provides metrics for evaluating the quality of life in urban areas. Emergy analysis [13] offers a comprehensive method for assessing the energy and environmental aspects of urban systems. Ecological literacy [14] research aims to enhance public understanding and stewardship of the environment. Ecological patterns [15] studies contribute to the understanding of spatial and temporal dynamics in urban ecosystems. Finally, ecological risk [16] assessments help identify and mitigate potential threats to urban ecosystems. These diverse and comprehensive analyses not only provide a robust theoretical foundation for sustainable urban planning and development but also afford urban managers and planners a rich tapestry of perspectives and strategies. This body of research has been instrumental in broadening and deepening the scope and understanding of urban sustainability, paving the way for more resilient and environmentally conscious urban futures.

In the realm of urban sustainability research, geographic information system (GIS) technology and emergy analysis have emerged as two pivotal methodologies, each providing a distinct lens through which to evaluate the sustainability of cities—spatial distribution and energy flow, respectively. GIS technology, known for its versatility, has found extensive application in diverse areas of urban research. For instance, bibliometric analysis [17] has leveraged GIS to map the distribution of research outputs and identify knowledge gaps. Dynamic assessment [18] utilizing GIS has enabled the monitoring of urban changes over time, facilitating better planning and management. GIS has also proven invaluable in disaster vulnerability assessment [19], helping to predict and mitigate the impact of natural disasters on urban areas. In the realm of intelligent urban planning [20], GIS has been instrumental in creating smart city solutions that optimize resource allocation and service delivery [21]. For ecological risk assessment [22], GIS technology has provided a spatial framework for analyzing and managing risks to urban ecosystems. Urban green space accessibility analysis [23] has benefited from GIS to ensure equitable distribution of green spaces for public health and well-being. Additionally, ecological security assessment [24] and monitoring of land use change [25] have both utilized GIS to inform policies and decisions related to urban development and environmental conservation. Simultaneously, the emergy analysis method [26,27] has brought a unique dimension to urban research [28,29,30], enabling the quantification of energy input and output within urban systems [31], which serves as a robust indicator of sustainability [32,33]. This approach has been particularly insightful in specific urban planning [34,35] and ecological environmental management practices [36,37], where GIS analysis based on diverse land parcels [38] has become an indispensable technical tool [39]. With the aid of GIS technology [40], researchers are able to conduct nuanced spatial data analysis and processing on various urban land types [41,42,43], including farmland [44,45], water bodies [46], forests [47], grasslands [48], and built-up areas [49]. This meticulous analysis has unraveled the distribution characteristics and intricate inter-relationships of different land parcel types within the urban space, offering critical insights for scientifically assessing urban resource utilization efficiency, ecosystem health, and environmental quality. Furthermore, urban land use planning, ecological protection policies, and disaster risk management measures, all informed by GIS analysis, are contributing to the promotion of sustainable urban development. These measures aim to achieve a harmonious coexistence between humans and nature while significantly enhancing the protection and management of urban ecological environmental resources.

In the era of informatization, digital systems have become crucial tools in promoting urban sustainability. Geographic information systems (GISs) provide a scientific basis for urban planning, environmental monitoring, and resource management with their powerful spatial data processing and analysis capabilities. Building Information Modeling (BIM) supports efficient information for architectural design, construction, and operation through three-dimensional modeling. When GIS and BIM are integrated, they not only facilitate seamless integration of spatial data with building information but also bring about a deeper level of transformation in urban sustainability. The application of GIS and BIM can optimize urban resource allocation, enhance the precision of urban planning, and reduce resource waste. In terms of environmental protection, digital systems can monitor urban environmental conditions in real-time, providing data support for pollution control and ecological restoration. In the field of transportation, digital systems help to improve the rationality of transportation planning, reduce traffic congestion, and lower energy consumption. From a broader perspective, digital systems have promoted the informatization of urban planning and management, enhanced the efficiency of urban governance, and laid the groundwork for sustainable urban development. Simultaneously, digital systems can help cities meet energy-saving and emission-reduction goals, drive the progress of green buildings, and improve the quality of life for urban residents [50].

In analyzing urban ecology, the comprehensive use of multi-dimensional indicators such as ecological emergy, land use change, population density, habitat status, carbon emissions and storage, and enhanced vegetation index, combined with in-depth analysis using GIS technology, is of profound significance for fully understanding the health and dynamic changes of urban ecosystems. The integrated analysis of these indicators not only reveals the complexity and multifunctionality of urban ecosystems but also provides a scientific basis for urban planning and ecological management. To further verify the sustainability of future urban development, it is particularly important to introduce neural network model methods [50,51,52] for predictive analysis of urban sustainability based on time series. Neural network models can simulate the complex information processing of the human brain, and by learning from historical data, they can predict the future evolution trends of urban ecosystems. This method helps urban policymakers anticipate potential ecological challenges in the future, allowing them to take proactive measures to ensure the sustainability of urban development and the health of ecosystems. Through this comprehensive research methodology, it is possible to promote the continuous improvement of urban ecological environments, creating a more livable environment for residents.

The innovation of this paper revolves around the comprehensive approach, integrating the emergy method, GIS technique, and neural network model method to discuss the sustainability of Xuzhou City. This method is not only a static quantitative and qualitative assessment but also employs machine learning algorithms to conduct dynamic research on the entire city’s ecological sustainable status. The research presented in this paper has positive reference significance for industry practitioners and urban managers.

2. Materials and Methods

2.1. Study Area

Xuzhou City is situated in the northwest of Jiangsu Province, located on the southeastern edge of the North China Plain, and boasts a diverse geographical environment that includes mountains, plains, and bodies of water. With a total area of approximately 11,258 square kilometers, it is a significant city in the northern part of Jiangsu. Geographically, it faces the Yellow Sea to the east, borders Henan Province to the west, adjoins Anhui Province to the south, and is close to Shandong Province to the north. Xuzhou is known for its ecological beauty, high green coverage rate, numerous parks, lakes, and green spaces, which have formed a pleasant natural landscape and a healthy ecological environment. Additionally, the city is crisscrossed by rivers and dotted with many lakes, enriching the urban area with abundant water resources.

As an important city in Jiangsu Province, Xuzhou is undergoing rapid industrialization and urbanization. This development has brought about numerous ecological and environmental issues, such as water pollution, air pollution, and ecological destruction. Therefore, conducting ecological sustainability research in Xuzhou is particularly important. Firstly, such research helps to uncover the roots of ecological and environmental problems, providing scientific evidence for policy formulation. Secondly, it can propose targeted ecological protection and restoration measures to promote harmonious coexistence between humans and nature. Finally, ecological sustainability research aids in Xuzhou’s realization of green development, enhancing the city’s competitiveness and laying a solid foundation for sustainable development in the future (In Figure 1).

2.2. Research Framework

Figure 2 provides a detailed illustration of the research roadmap presented in this paper, which encompasses data collection, method selection, and the presentation of evaluation results, offering a comprehensive overview of the research content. This study focuses on Xuzhou City as the research subject, integrating urban models to highlight the impact of various input factors such as material flow, energy flow, and information flow on the urban ecosystem. In terms of research methodology, this paper employs dynamic analysis methods, with particular consideration given to the application of neural network theory. Based on this, the paper delves into the sustainability issues of Xuzhou City. Through this integrative research approach, the aim is to reveal the dynamic change patterns of the urban ecosystem and to provide a scientific basis and practical guidance for the sustainable development of Xuzhou City.

2.3. Data Source and Processing

1. Data source

The land use data from 2000 to 2020 used in this study, spanning five periods, were sourced from the China Annual 30 m Land Use Data from 1985 to 2022 provided by Wuhan University [53], with an overall accuracy of 80%. The digital elevation model (DEM) data were obtained from the geospatial data cloud [54]. The vector maps of the administrative divisions of Jiangsu Province and Xuzhou City were sourced from the provincial, municipal, and county-level databases [55]. Carbon emission data were derived from the ODIAC Fossil Fuel Emission Database [56]. Population density data were obtained from worldpop [57], and the enhanced vegetation index (EVI) data were from the MOD13Q1 dataset [58].

• 2.. Data processing

Habitat Quality: The INVEST model’s Habitat Quality module was used to assess the habitat quality of Xuzhou City. The size of the habitat quality reflects the degree of fragmentation of the regional habitat and its resistance to habitat degradation. Based on land use data, the habitat quality of the study area was evaluated from aspects such as the maximum influence distance of threat factors, sensitivity, and habitat suitability. Considering that cultivated land, construction land, and unused land are areas with concentrated human activities that have a significant impact on the surface habitat, they were classified as non-habitat types. In combination with the actual situation, the maximum influence distance and weights of different threat factors were set (Table 1), as well as habitat suitability and the sensitivity of different habitats to threat factors (Table 2).

Carbon Storage: The Carbon Storage module of the INVEST model was used to estimate the current changes in ecosystem carbon storage by employing land use raster data and the stocks of four carbon pools (aboveground biomass carbon, belowground biomass carbon, soil carbon, and dead organic matter carbon). The calculation formula of the model is as follows:

${\mathrm{C}}_{\mathrm{i},\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}={\mathrm{C}}_{\mathrm{i},\mathrm{a}\mathrm{b}\mathrm{o}\mathrm{v}\mathrm{e}}+{\mathrm{C}}_{\mathrm{i},\mathrm{b}\mathrm{e}\mathrm{l}\mathrm{o}\mathrm{w}}+{\mathrm{C}}_{\mathrm{i},\mathrm{s}\mathrm{o}\mathrm{i}\mathrm{l}}+{\mathrm{C}}_{\mathrm{i},\mathrm{d}\mathrm{e}\mathrm{a}\mathrm{d}}$

where ${\mathrm{C}}_{\mathrm{i},\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}$ is the total carbon storage (t·hm⁻²), ${\mathrm{C}}_{\mathrm{i},\mathrm{a}\mathrm{b}\mathrm{o}\mathrm{v}\mathrm{e}}$ is the aboveground biomass carbon storage, ${\mathrm{C}}_{\mathrm{i},\mathrm{b}\mathrm{e}\mathrm{l}\mathrm{o}\mathrm{w}}$ is the belowground biomass carbon storage, ${\mathrm{C}}_{\mathrm{i},\mathrm{s}\mathrm{o}\mathrm{i}\mathrm{l}}$ is the soil carbon storage, and ${\mathrm{C}}_{\mathrm{i},\mathrm{d}\mathrm{e}\mathrm{a}\mathrm{d}}$ is the dead organic matter carbon storage.

The construction of the carbon density database mainly referred to the measured data from the existing literature. The following principles were followed during data selection: The carbon density data for soil, aboveground biomass, belowground biomass, and litter organic matter were all obtained from field surveys. Priority was given to the literature data from Xuzhou City, Jiangsu Province. For parameters lacking local literature, data from adjacent regions were used as much as possible to fill in the gaps, and the relationship model between carbon density and precipitation was used for correction. The calculation formula is as follows (Please note that the actual formula is not provided in the original text, so it cannot be translated without the specific equation):

$C_{BP}=6.789\times e^{0.0054\times P_{MA}} (R^2=0.70)$

$C_{SP}=3.3968\times P_{MA}+3996.1 (R^2=0.11)$

$K_{BP}={\displaystyle \frac{C_{BP1}}{C_{BP2}}}$

$K_{SP}={\displaystyle \frac{C_{SP1}}{C_{SP2}}}$

where $P_{MA}$ is the average annual precipitation in the study area, in millimeters; $C_{BP}$ is the aboveground biomass carbon density calculated based on precipitation, in t·hm⁻²; $C_{SP}$ is the soil carbon density calculated based on precipitation, in t·hm⁻²; $K_{BP}$ is the correction coefficient for the aboveground biomass carbon density based on the precipitation factor; $K_{SP}$ is the correction coefficient for the soil carbon density based on the precipitation factor; $C_1$ and $C_2$ are the carbon density data for Xuzhou City and Jiangsu Province, respectively. The corrected carbon density value for Hebei Province is obtained by multiplying the carbon density value of Jiangsu Province by $K_{BP}$ and $K_{SP}$. The final carbon density for the land use types in Xuzhou City is presented in Table 3.

EVI: As NDVI sensitivity decreases when monitoring areas with high vegetation cover, EVI is more effective in reflecting the vegetation growth conditions in wetland areas, grasslands, coniferous forests, and broad-leaved forests during the growing season. Therefore, this paper utilizes the currently most reliable vegetation monitoring data, MODIS EVI, for studying vegetation cover. MOD13Q1 data is selected, which is synthesized every 16 days, resulting in 23 issues per year. The initial downloaded data is processed preliminarily using ENVI software 2022. Due to the lack of data for the first three periods before 2000, the data from the first three periods of 2001 is used as a substitute. ArcGIS is employed to convert the initial data into a monthly EVI dataset using the Maximum Value Composites (MVC) method. Based on this, the average EVI values from January to December are calculated to obtain the annual EVI dataset.

Ecosystem Services Value (ESV): This paper adopts the improved equivalent factor method for the value of ecosystem services per unit area proposed by Xie Gao Di, using data collected from 2000 to 2020, including the “Xuzhou City Statistical Yearbook” and “China Agricultural Product Price Survey Yearbook”, combined with remote sensing images of Xuzhou City, to obtain the unit area ecosystem services value equivalent table applicable to Xuzhou City (Table 4). The calculation formula is as follows:

$D={\displaystyle \frac{1}{7}}\sum {\displaystyle \frac{a_i{v}_i{u}_i}{A}}$

D is the ecosystem service value of one standard equivalent factor for year t (CNY/hm²); I is the variety of grain crops (rice, wheat, corn); the sowing area for i grain crops is measured in hm²; the unit area yield of i cereal crops is measured in kg/hm²; the price of i food crops is measured in CNY/kg; and A is the total area (hm²) of major grain crops in year t. Using ArcGIS software 2023 to divide Xuzhou City into 5 km × 5 km scale grids and perform grid calculations, the final ecosystem service value is calculated by calculating the land area of each land use type in each grid.

2.4. Research Model

This study analyzed the renewable and non-renewable energy sources in Xuzhou City and developed an ecological accounting framework for the urban system of Xuzhou City, especially considering factors such as land use change, mineral resources, fuel resources, commercial products, immigration, tourism, and markets. The accounting model revolves around material flow, energy flow, and information flow and improves the accuracy of the entire urban accounting system by converting them into energy flow and feedback flow, as shown in Figure 3.

2.5. Methodology

In order to improve the ecological status and upgrade the ecological level of Xuzhou City, urban managers have taken a series of measures aimed at enhancing the metabolic level of the entire city and continuing its development in a sustainable manner.

Specific measures include: (1) accounting for the ecological value of renewable energy sources such as solar energy, wind energy, and chemical energy in Xuzhou City; (2) joining the upgraded urban industrial system for wastewater, exhaust gas, and waste treatment systems; (3) accounting for urban ecological changes of different types of land parcels; (4) ecological service system accounting, etc. Specific accounting indicators include land type change, ecological economic services, population density, habitat index, carbon emissions and storage, and enhanced vegetation index.

2.5.1. Embodied Energy (Emergy) Accounting Method

The emergy accounting method is an evaluation tool based on ecological economics, which quantifies the services provided by ecosystems and the resource consumption in human economic activities by converting different types of ecological flows (such as matter, energy, information, etc.) into a unified emergy unit. Emergy is a concept that measures the labor value contained in energy, considering not only the quantity of energy but also the quality of energy and the energy input required for its transformation.

In the field of urban ecology, the emergy accounting method has significant application value. First, it helps urban policymakers gain a comprehensive understanding of the energy flow and resource utilization efficiency of the urban ecosystem, providing a scientific basis for urban planning and management. Second, emergy analysis can evaluate the service functions of the urban ecosystem, such as air purification, water conservation, and biodiversity protection, which helps identify the key components and vulnerable parts of the urban ecosystem. Furthermore, through emergy accounting, the sustainability of urban development can be quantified, and the impact of different policies or measures on the urban ecosystem can be assessed, thus guiding the city to achieve green, low-carbon development and the construction of an ecological civilization. In summary, the emergy accounting method provides an effective evaluation tool for the sustainable development of the urban ecological field.

The specific path of using the energy value method to calculate urban ecology is shown in Figure 4, which clearly illustrates the relationship between various energy types and natural systems such as inflow energy, production energy, output energy, loss energy, storage capacity, storage energy, and processing efficiency (see Table 5 and Table 6). Figure 4 did not consider the positive effects of urban wastewater, exhaust gas, waste treatment, and reuse, while Figure 5 analyzed this point. The specific pollution system analysis model is shown in Figure 6.

2.5.2. Geographic Information System (GIS) Method

A GIS, or geographic information system, is an advanced technological framework designed to handle the entire lifecycle of geographic information, from data collection and storage to management, analysis, and visualization. This system harnesses methodologies such as cartography, spatial analytics, and remote sensing to explore geographical phenomena and tackle challenges related to land utilization, resource stewardship, and ecological monitoring. With GIS, users are able to merge diverse geographic datasets, perform spatial relationship assessments, and produce detailed maps or graphical representations. The versatility of GIS methods is widely recognized in sectors such as urban development, wildlife conservation, disaster management, and more, where it supports informed decision-making, addresses complex issues, and enhances the efficient handling of spatial information. For an illustration of a specific route map, please consult Figure 7.

Geographic information system (GIS) plays a pivotal role in urban sustainability analysis. GIS enables the integration and visualization of various data layers, such as land use, population distribution, infrastructure, and environmental factors, providing a comprehensive understanding of urban dynamics. By mapping and analyzing these data, policymakers can identify sustainable practices, optimize resource allocation, and plan for future growth while preserving environmental quality. GIS facilitates informed decision-making, leading to more efficient and sustainable urban development.

2.5.3. Neural Network Model

As a system, the constituent elements and hierarchical structure of a city determine its sustainable effectiveness, but with the continuous input of material flow, energy flow, and information flow, the hierarchical structure of a city will change. For example, changes in land use types, population density, habitat quality, carbon emission distribution, etc., will all change accordingly. In order to continuously evaluate the dynamic sustainability of cities, this paper uses neural networks as a predictive model to analyze the ecological sustainability changes of urban dynamics.

To build an effective neural network model for evaluating urban ecological sustainability, the following steps need to be taken (Figure 8 and Figure 9):

1. Data Preprocessing:

Standardization/Normalization: Since different indicators may have different data ranges and units of measurement, it is necessary to standardize or normalize the data to bring them to the same numerical level, which allows the neural network to process them more effectively. In this paper, energy value is used as the unified unit of measurement.

Missing Data Treatment: Address missing values in the data through interpolation, deletion, or using statistical methods such as mean, median, etc., to fill in the missing data.

Feature Selection: Select the most influential features from numerous indicators for predicting ecological sustainability in order to reduce the complexity of the model and the risk of overfitting. The indicators in this model involve seven categories: energy value indicators, land use change indicators, ecosystem service indicators, population density data, habitat quality data, carbon emissions data, and enhanced vegetation index (EVI).

• 2.. Model Design:

The model in this paper adopts two types: convolutional neural networks (CNNs) and recurrent neural networks (RNNs). The reason is that this paper also uses GIS analysis maps to perform time-series analysis on various indicators of the city. CNN is very well suited for handling image and grid-like data. If urban sustainability prediction involves the analysis of the city’s spatial layout, CNN can effectively extract spatial features such as roads, buildings, vegetation, etc., which can help in predicting urban sustainability. In addition, the predictive analysis in this paper involves the analysis of time series, such as land use change indicators, ecosystem service indicators, population density data, habitat quality data, carbon emissions data, and EVI. The prediction task needs to consider the impact of historical data on the future, and in this regard, RNN can effectively capture the temporal dependencies in the data.

• 3.. Model Training:

Dataset Division: Divide the dataset into training, validation, and test sets to facilitate model training and evaluation.

Optimization Algorithm: Choose an appropriate optimization algorithm to adjust the weights of the neural network.

Learning Rate and Regularization: Set a suitable learning rate and use regularization methods to prevent overfitting.

• 4.. Model Evaluation:

Performance Metrics: Use performance metrics such as mean squared error (MSE), coefficient of determination (R²), etc., to evaluate the predictive performance of the model.

Cross-Validation: Employ cross-validation methods to assess the generalization ability of the model.

• 5.. Model Optimization:

Hyperparameter Tuning: Optimize the model by adjusting hyperparameters such as learning rate, number of layers, number of neurons, and regularization parameters.

Model Ensemble: Combine the predictions of multiple models to improve the accuracy and stability of the model.

• 6.. Result Interpretation:

Feature Importance Analysis: Analyze the impact of different indicators on the model’s prediction results to understand the decision-making logic of the model.

Through the above steps, an effective neural network model for evaluating urban ecological sustainability can be constructed, and its predictive accuracy and reliability can be continuously improved.

Specific model design parameters are as follows: CNN model: 2 convolutional layers (32 3 × 3 filters with ReLU activation), 2 fully connected layers (128 and 64 neurons with ReLU activation), and dropout ratio of 0.2; RNN model: 2 LSTM hidden layers (50 neurons with ReLU activation) and a dropout ratio of 0.2; learning rate: 0.001; cross-validation: 5-fold cross-validation.

3. Results and Discussion

3.1. Perspective of Land Use Types and Sustainability

• (1). Analysis of Emergy Change

By using the emergy method, an analysis of natural renewable energy resources in Xuzhou City was conducted, taking solar energy, rainwater energy, wind energy, and geothermal energy as examples. The analysis started from five modules: cropland, woodland, grassland, built-up land, and water area. The results in Figure 10 were obtained through calculations. In terms of the total emergy in 2000, cropland and water area had the largest quantities, followed by woodland, built-up land, and grassland. This indicates that from the perspective of ecological emergy value, cropland and water area play a major role, which is also verified by the GIS map in Figure 11. Looking at the energy value of each energy type, taking cropland as an example, the emergy values from largest to smallest are rainwater chemical emergy, rainwater potential emergy, solar emergy, geothermal emergy, and wind emergy. Based on the comprehensive emergy value analysis, in order to enhance the ecological effects and levels of Xuzhou City, it is necessary to focus on cropland and water area and to give special consideration to the assessment of rainwater potential emergy and rainwater chemical emergy.

The distribution of emergy values in 2020 showed a significant change (Figure 11), with cropland still having the highest emergy, indicating that cropland continues to occupy a dominant position in Xuzhou City. What has changed is that the emergy value of woodland and built-up land has increased significantly, which can be interpreted as the implementation of China’s reforestation policy having a notable effect on the emergy of woodland; the increase in the emergy of built-up land is due to the rapid urbanization process from 2000 to 2020. The increase in the emergy of woodland is positive for the ecology of Xuzhou City, while the increase in the emergy of built-up land has a negative effect. As seen from the data of the highest emergy (woodland’s 1.2 × 10²⁰ sej and built-up land’s 9.43 × 10¹⁹ sej), overall, the ecological emergy at the end of 2020 is still in a positive state.

• (2). Analysis of Land Use Change

Geographic information systems (GISs) play a significant role in urban ecological planning. Through the visualization of urban land parcels, GIS aids in deciphering the ecological structure and hierarchy of cities. It provides decision-makers with scientific data and visual assistance, making planning and management more rational and effective. GIS can accurately delineate ecological redlines, optimize the allocation of green spaces, and establish ecological corridors, thereby promoting harmonious coexistence between humans and nature. The role of GIS tools in urban ecological planning lies in their ability to not only enhance the science and precision of planning but also to increase the feasibility and sustainability of the planning process.

Taking the land use transition matrices of 2000 and 2020 as examples, as specifically shown in Figure 12 and Table A2 in the Appendix A, the land types are divided into six categories: cropland, woodland, grassland, water area, built-up land, and unused land. Among these, cropland occupies the dominant proportion (with a total area of 8055.7227 square kilometers), accounting for approximately 90.6% of the total, followed by built-up land, which accounts for 8.49%. The rest are water area, woodland, grassland, and unused land, in that order. From the perspective of land area, this also explains the content of the previous section regarding why the energy value of cropland in Xuzhou City is the highest.

Figure 13 is a GIS map based on data from 2000 to 2020 (with each 5-year interval as a calculation period), and the source of the data can be found in Section 2.3. By comparing the data and GIS maps for these five years, the most significant trend is the noticeable increase in urban land. Especially starting from 2010, the process of urbanization in Xuzhou City has significantly accelerated. Comparing the construction land between 2000 and 2020, it is found the increase has exceeded tenfold. The area of cropland has decreased, with a large amount of land being used for urban development. The areas of forest, grassland, and water have not changed significantly, which is determined by the topographical characteristics of Xuzhou City.

• (3). Analysis of EVS Changes

Research on the ecosystem service value of different land types is crucial for understanding and constructing an ecological city. This type of research helps to reveal the contributions of various land types in providing ecosystem services, such as cropland ensuring food security, woodland regulating the climate, grassland maintaining biodiversity, water areas providing fresh water resources, built-up land meeting human habitation needs, and unused land possibly serving as ecological reserves. Through such analysis, it is possible to optimize the urban spatial layout, achieve the rational allocation of resources, enhance the overall benefits of the urban ecological environment, and thus promote sustainable urban development and improve the quality of life for residents. This has a profound impact on promoting the construction of urban ecological civilization and building a harmonious relationship between humans and nature.

Figure 14 and Table A3 (Appendix A) provide a list of ecosystem service value data for Xuzhou City. Figure 14 clearly shows that for Xuzhou City, cropland and water areas have the highest ecosystem service value, accounting for 97.1% of the total value, while woodland and grassland account for approximately 2.9%, and other unused land can be considered negligible. Due to the lack of data on built-up land, its impact has not been analyzed in this section.

Figure 15’s GIS distribution map clearly illustrates the density distribution of ecosystem services in Xuzhou City. The most notable feature is that the ecosystem service density for water areas is the highest (ranging from 1500 to 2378), and it has continuously provided service value from 2000 to 2020. This highlights the significant value of urban water area protection. In contrast, the ecosystem service level of built-up areas is relatively low, mostly maintaining at the 0–100 level. This also indicates that, overall, the city’s ecosystem services need to be sustained by natural elements such as agricultural land and water systems, while areas with high human population density have a negative impact on ecosystem services.

3.2. Perspective of Ecosystem Services and Sustainability

• (1). Analysis of Population Density Changes

The distribution of population density has a significant impact on the sustainability of ecological cities. Areas with high population density may face increased pressure on the ecological environment, damage to the ecological structure, and disordered ecological hierarchy. In such cases, the flow of ecosystem services is easily disrupted, leading to a weakening of the city’s ecological functions. Conversely, a moderate population density helps to maintain the stability of the ecological city structure, preserve the order of ecological hierarchies, promote smooth ecological flow, and facilitate the rational use and recycling of resources, thereby ensuring the sustainable development of ecological cities and the quality of life for residents. Therefore, scientifically planning the distribution of population density is key to achieving the sustainable development of ecological cities. Table A4 and Table A5 (Appendix A) and Figure 16 and Figure 17 illustrate the distribution patterns.

Figure 16 is a GIS display of population density, and according to the time series from 2000 to 2020, qualitatively speaking, the population density of the entire Xuzhou City has not changed significantly in the >5000 category. This can also be seen from the proportional distribution in Figure 17. Compared to the balanced population distribution in 2000, the population density in 2020 is more concentrated. The 500–1500 population density category was more evenly distributed in 2000, but it showed a decreasing trend by 2020 (the area of the 0–500 category has decreased. This is due to the process of urbanization, where a large population has moved towards cities, leading to the disruption of a balanced population distribution pattern and presenting a more concentrated trend. Among them, the 1500–3000 category has shown an increasing trend. The movement of a large population from rural to urban areas does not necessarily mean that they are all concentrated in the city center of Xuzhou. Instead, they are distributed to various smaller cities within Xuzhou, resulting in a trend of concentration in this category of population density.

• (2). Analysis of habitat quality changes

The change in urban habitat quality is crucial to the impact on the structure of an ecological city. Good habitat quality can maintain the stability of the ecological city structure and preserve the integrity and continuity of ecological hierarchies. When habitat quality declines, the structure of the ecological city may be damaged, and ecological hierarchies may become disordered, leading to the degradation of ecological functions and a reduction in biodiversity. Urban ecological elements, such as green spaces and wetlands, may decrease, and ecological corridors may be broken, and this can affect ecological flow and species migration. Moreover, changes in habitat quality can also affect the quality of life and health of urban residents, which in turn affects the sustainable development and ecological security of the city. Combining the GIS analysis in Figure 18A,B and quantitative analysis (data from Appendix A Table A9), it is found that the change in habitat quality is not significant. The analysis of the GIS map in Figure 16 reveals that the block-level changes from 2000 to 2020 are not significant. Together with the quantitative analysis in Figure 17, it can be seen that except for the 0–0.2 level index showing an upward trend (from 0.178 in 2000 to 0.245 in 2020), there is little change in other levels.

3.3. Perspective of Carbon Footprint and Sustainability

Carbon emissions and carbon storage play a crucial role in urban ecosystems, and they have a profound impact on aspects such as the hierarchy, structure, composition, elements, and models of urban ecology. Firstly, within the urban ecological hierarchy, carbon emissions and carbon storage are directly related to the higher-level structure and functional hierarchy of the urban ecosystem. High carbon emissions may lead to the dysfunction of ecosystem services, while effective carbon storage helps to maintain the stability and healthy development of the ecosystem. Secondly, the urban ecological structure includes both biological communities and non-biological environments. Excessive carbon emissions can damage biodiversity, affecting the composition of the ecosystem, such as the reduction of vegetation and soil infertility. Carbon storage, on the other hand, helps to maintain the integrity and diversity of the ecosystem. In terms of constituent elements, carbon emissions and carbon storage affect various elements of the urban ecosystem, including air, water, and soil. Carbon emissions can lead to air pollution and water quality deterioration, while carbon storage can help improve the quality of these environmental elements. For urban ecological models, carbon emissions and carbon storage are key indicators for evaluating and constructing urban ecological models. They can be used to predict future trends of urban ecosystems, guide urban planning and management, and develop corresponding carbon reduction and carbon sequestration strategies. In summary, carbon emissions and carbon storage have an important role in the classification of urban ecosystem hierarchy, the stability of structure, the maintenance of constituent elements, and the construction of models. They are the foundation for the sustainable development of urban ecology. Figure 19 and Figure 20 are the GIS change maps for carbon emissions and carbon storage, respectively, and Figure 21 is a quantitative comparative analysis chart.

Combining the spatial analysis of carbon emissions using GIS with the quantitative analysis charts, it can be observed in Figure 19 that as the population aggregation areas and the size of urban centers increase, the overall carbon emissions in the urban area show a trend of gradual increase. From 2000 to 2020, there is a significant increase in carbon emission areas, especially in the categories greater than 5000, with a substantial increase from 2010 to 2020. In terms of quantitative data, the carbon emissions from building land are the highest, followed by unused land, agricultural land, and others that are relatively low.

Analyzing from the perspective of carbon storage, the areas with the lowest efficiency are urban population aggregation zones, while the most obvious carbon storage areas are the cultivated land regions. Taking 2020 as an example, cultivated land accounts for approximately 77.48% of the proportion of all land types for carbon storage, followed by utilized land, which accounts for about 19.03%. Considering the time series, the cultivated land area in Xuzhou City experienced a slight decrease in carbon storage from 7.75 × 10⁷ tons of CO₂ in 2000 to 7.07 × 10⁷ tons of CO₂ in 2020. Overall, the urban carbon storage primarily depends on areas outside of urban construction land.

3.4. Perspective of EVI and Sustainability

The enhanced vegetation index (EVI) as an advanced remote sensing indicator is of great significance for the sustainable analysis of urban ecological hierarchy, structure, and models. At the urban ecological hierarchy, EVI can reflect the coverage and vitality of urban vegetation, which helps to assess the health of urban green spaces and thus guide urban ecological planning and design. Structurally, the monitoring data of EVI can reveal the trend of vegetation changes in different urban areas, providing a scientific basis for the optimization of urban ecological structures. For urban ecological models, the introduction of EVI can enhance the accuracy and practicality of the models. By analyzing EVI data, researchers can establish accurate ecological models to predict the future development of urban vegetation, offering decision-making support for the sustainable development of cities.

Appendix A Table A6 and Table A7 present the changes in the five-level EVI indicators in Xuzhou City from 2000 to 2020. Figure 22A–C shows the quantitative analysis of EVI based on the time series for five years and the GIS distribution map of EVI. Figure 22A,B display the distribution of each EVI level, where it can be observed that the levels 0.2–0.3, 0.3–0.4, and 0.4–0.5 dominate, with the other three levels being less frequent. Analyzing the 0.2–0.3 level individually, there is a significant decrease in 2000 compared to the other four years, with the proportion of this level being 34.72% in 2000 and 16.44%, 16.22%, 14.82%, and 14.94% in the other years, respectively. For the 0.3–0.4 level, there is a progressive decrease from 56.53% to 36.43%. For the 0.4–0.5 level, except for the lower percentage in 2000, there is little difference between the years, with values of 2.56%, 28.87%, 32.92%, 33.7%, and 34.64%. The GIS map in Figure 21 also shows that with the progression of the time series, the distribution of EVI at the 0.4–0.5 level has been increasing, especially from 2010, with significant growth in the EVI distribution from 2015 to 2020. This is attributed to the governance of ecology by the authorities, particularly the implementation of policies such as returning farmland to forests and measures for ecological civilization.

The implementation of the returning farmland to forests and ecological governance projects in Xuzhou City has had a positive impact on ecologically sustainable development. The reforestation has increased forest coverage, enhanced surface vegetation, helped to reduce soil and water erosion, improved regional climate, and elevated air quality. The ecological governance projects have restored wetlands and river ecosystems, strengthening the self-purification capacity of urban water bodies. These measures not only promote the recovery of biodiversity but also provide citizens with more green leisure spaces, improving the quality of life for residents. Moreover, these ecological measures contribute to the adjustment of the agricultural structure, promote the development of green agriculture, and provide a solid foundation for achieving a win–win situation between Xuzhou City’s economy and ecological environment.

3.5. Analysis of Neural Network Models

This section uses the sustainable analysis parameters of emergy to predict and analyze the sustainability of Xuzhou City. Using data from 2000 to 2020 as the training database, the training of historical data reveals a significant discrepancy between actual data and predicted data in the fitting process. Figure 23 illustrates the comparative analysis of four historical data sets. Taking the emergy ESI sustainability parameter data as an example, the data from 2000 to 2020 maintained a range of 0.4–0.6 (Figure 23A,B), while the actual data fluctuated more, with an average value of 0.7–0.8. Taking the final change in 2020 as an example, the difference in change is approximately 0.15 in Figure 23A; in Figure 23B, the difference is 0.13; in Figure 23C, the difference is 0.1, and finally, in Figure 23D, the difference is approximately 0.1. Through this process, it can be found that the training of the database has sufficient accuracy.

Based on the training results from Figure 23A–D, the ESI index for the two time series segments of Xuzhou City from 2030–2040 and 2040–2050 was predicted and analyzed. Figure 24 shows the changes in prediction accuracy, with the precision error range for the 2030–2040 period being approximately 20%, and for the 2040–2050 period, the error is about 13%. Through the predictive analysis using neural networks, it can be determined that the ecological emergy fluctuation range for Xuzhou City in the next 20–30 years is between 15 and 20% (The circle represents a comparison of the ultimate trends). This indicates that in addition to maintaining the current series of ecological improvement measures, Xuzhou City must also pay attention to the effectiveness and sustainability of these ecological measures.

This fluctuation may be caused by a variety of factors. Firstly, policy interventions such as urban greening projects and industrial relocation have a direct impact on Xuzhou’s ESI. The implementation of urban greening projects may enhance the service functions of urban ecosystems, thereby increasing the ESI value. On the other hand, industrial relocation could lead to the environmental restoration of previously industrialized areas, which may have either a positive or negative effect on the ESI. Specifically, the Xuzhou New City District is renowned for its abundant green resources, which require a significant amount of energy input for maintenance and management. These green resources not only include parks and green spaces but also urban forests and wetlands, all of which contribute to the improvement of the ESI value. However, if management and maintenance are inadequate, or if green resources are damaged, the ESI value may decrease. Moreover, the expansion of the urban area and infrastructure construction in the Xuzhou New City District may also lead to fluctuations in the ESI, as these activities involve a large amount of energy consumption [59,60]. The underlying reasons for the high volatility may also be related to the policy orientation and development strategy of the local government. For instance, if policies favor rapid urbanization, it may sacrifice some ecological resources, leading to volatility in the ESI value. Conversely, if policies place more emphasis on ecological protection and sustainable development, the ESI value may exhibit a more stable or gradually increasing trend.

On the path of green development, a series of policies and technical measures aimed at enhancing urban sustainability are introduced. These measures may include promoting clean energy, increasing green spaces, and raising energy efficiency standards. Neural network models and GIS methods can simulate the impact of these measures on urban sustainability, predicting the state of the city after the implementation of green policies. Through this scenario, the study can evaluate the effectiveness of the green development strategy, providing policymakers with a basis and direction for implementing green policies.

In contrast to the green development scenario, the high-carbon development scenario assumes that the city continues to rely on fossil fuels, neglecting environmental protection and energy efficiency improvements. This scenario reveals the ecological degradation and environmental risks that the city may face if it follows a high-carbon emissions path. The application of neural network models and GIS methods helps to quantify the negative impact of high-carbon development on urban sustainability, such as increased air pollution and resource depletion. The purpose of this scenario is to alert policymakers, demonstrating the serious consequences of inaction, thus promoting the formulation and implementation of low-carbon policies.

4. Research Innovation and Limitations Discussion

4.1. Innovation Points and Progress Discussion

Currently, academic researchers are deeply exploring how to integrate emergy analysis with urban ecosystems in order to comprehensively assess the use of urban resources, ecological benefits, and environmental protection pressures. Through the method of emergy analysis, the emergy value of resources within the urban system can be quantified, which helps to evaluate the positive and negative impacts of urban development on the ecological environment, thus providing scientific support for urban sustainability decisions [61,62,63,64,65]. In addition, land use type-GIS technology is being widely used in the research of urban ecological sustainable development. This technology, combined with the spatial analysis capabilities of GIS, can accurately identify and analyze the spatial distribution of urban land use types [66,67,68,69,70], thereby assessing the specific impact of land use structure on ecosystems, revealing the ecological sensitivity and vulnerability of different regions, and laying the foundation for the rational planning and scientific management of urban ecosystems.

• (1). Innovation Points of This Study

In the current research on sustainable urban development, scholars typically employ either the emergy method or GIS technology separately. However, this paper proposes an innovative, comprehensive research framework that integrates emergy analysis with GIS technology and carbon emission models, opening up new avenues for the study of sustainable urban development. This research quantifies the energy value of resources, revealing the overall benefits and contributions of the urban system. With the help of GIS and neural network models, we can deeply analyze urban spatial patterns, land use, and ecosystem services, thus gaining a clearer understanding of the complex environmental impacts of urban development. The application of the carbon emission model also allows us to comprehensively evaluate a city’s carbon footprint and the effects of climate change, providing important decision-making bases and guidance for formulating low-carbon development strategies. This comprehensive research approach not only enhances the assessment level of urban environmental sustainability but also offers powerful decision support for urban planning and management.

• (2). Research Progress and Prospects

Currently, research on urban ecological sustainability that combines emergy analysis and land use type-GIS has made significant progress, bringing new research perspectives and practical methods to urban ecological protection and management. Looking forward, we should further promote the deep integration of emergy analysis and GIS technology to enhance the overall assessment capability of urban ecosystems. In addition, we need to focus on promoting data sharing and exchange, driving interdisciplinary cooperation in the field of urban ecological research, especially the integration of artificial intelligence technology, in the hope of building more sustainable and efficient urban ecosystems. This comprehensive research path will help us to understand and address urban ecological challenges more thoroughly and scientifically, providing more precise and practical decision-making bases for urban planning and management.

4.2. Comparative Analysis

After comparing the development trajectories of Shanghai and other cities in China, we have found that there is significant fluctuation and variation in the aspect of sustainable development among these cities. This phenomenon indicates that in the process of sustainable urban development, the implementation of ecological measures is crucial for enhancing the overall sustainability of a city. Key factors affecting the level of sustainable development include the development policies of different cities, the breadth and depth of ecological transformation, and the adoption of green technologies and clean energy.

Shanghai, as an international metropolis, has demonstrated its commitment to sustainable development through stringent environmental protection policies, large-scale afforestation projects, and the promotion of green building practices. However, even such an advanced city faces the challenge of balancing rapid economic growth with ecological protection [71,72].

Other medium-sized cities can learn from Shanghai’s example, such as by prioritizing ecological considerations in urban development policies, increasing investment in green infrastructure, and promoting the development of a circular economy and low-carbon technologies. Moreover, when advancing ecological transformation, cities should not only focus on the breadth of transformation but also delve into the community and the daily lives of residents, encouraging public participation to collectively build a green, harmonious, and sustainable living environment.

The results of these comparative analyses not only reveal the diversity and complexity of sustainable development pathways but also provide valuable references for other cities. They assist these cities in pursuing economic growth while also considering the protection of the ecological environment, thereby achieving true sustainable development.

4.3. Limitations and Improvement Strategies

• (1). Limitations of This Study

This study uses emergy analysis and GIS technology to explore urban ecological sustainability, but it faces some limitations. The data collection and processing in emergy analysis are complex and time-consuming, requiring significant resources to ensure accuracy. Subjectivity in the assessment process needs to be strictly controlled. The integration of land use with GIS technology encounters challenges regarding data accuracy and update speed, which can affect the precision of results. Analyzing the interaction between land use and ecosystem services is crucial for understanding the long-term ecological impacts of cities. The accuracy of the carbon emission model is influenced by data quality and parameter settings, necessitating consideration of design complexity and cost to ensure applicability and affordability. While these methods provide tools for research, it is important to be vigilant about their limitations.

Xuzhou City faces the following situations in terms of industrial pollution, insufficient green spaces, and other ecological issues: In terms of industrial pollution, as a major industrial city, Xuzhou has numerous chemical, steel, and coal industries. While these industries have driven economic development, they have also led to serious environmental pollution. Industrial emissions of waste gas, wastewater, and solid waste have contaminated the air, water bodies, and soil. Particularly during heavy pollution episodes, industrial emissions become a major source of air pollution. Regarding the insufficient green spaces, as urbanization accelerates, the city’s area has been expanding, but the construction of green spaces has not kept pace, resulting in relatively few green areas in the city. This not only affects the quality of life of urban residents but also weakens the city’s ecological self-regulation ability. The distribution of parks and green spaces within the city is uneven, with some areas lacking adequate green space. Other ecological issues include water resource pollution and the decline of biodiversity. Water resource pollution mainly stems from the discharge of industrial wastewater and domestic sewage, as well as the excessive use of fertilizers and pesticides in agricultural activities. The decline in biodiversity is related to habitat destruction, environmental pollution, and overexploitation, which threaten the survival of local species and the balance of ecosystems.

Research on urban sustainability based on neural network models and GIS methods is significantly influenced by data quality and model assumptions. The level of data quality directly determines the reliability and accuracy of the model. If the data contains errors or is incomplete, it may lead to the model’s predictive results deviating from the actual situation, affecting the objectivity of the research conclusions. Moreover, the assumptions of the neural network model, such as the choice of network structure, learning rate, and activation function, all impact the model’s convergence speed and predictive performance. If the model’s assumptions do not correspond to the actual data characteristics, it may result in the model failing to effectively capture the complex patterns of urban sustainability. When generalizing the research findings to other cities, the main challenge lies in the inter-city variability. Each city has a unique socio-economic background, resource endowment, environmental status, and stage of development, which means that the findings from one city may not be applicable to another. Additionally, there are numerous factors influencing urban sustainability, such as policy orientation, resident behavior, and technological advancements, which may vary in their degree and manner of influence across different cities. Therefore, in the process of generalization, it is necessary to fully consider these regional and temporal differences, adjusting and optimizing the model to ensure the applicability and universality of the research results. In summary, to enhance the effectiveness and applicability of the research, researchers need to pay attention to the quality of the data and the suitability of the model. When disseminating the research findings, they should also consider the special circumstances of different cities, carrying out corresponding localization and adaptive adjustments.

• (2). Corresponding Improvement Measures

To enhance the quality and credibility of urban ecological sustainability research based on emergy methods and land use-GIS technology, a series of optimization strategies can be implemented. Addressing the complexity of data collection and processing, it is recommended to establish a standardized, integrated data collection system and use advanced data processing tools to streamline the process. Building a data sharing platform to promote data exchange and collaboration can also enhance data quality. Strengthening training and education on emergy methods can improve researchers’ ability to apply and understand these methods, thus reducing subjective bias in results. In terms of data accuracy and updates in the integration of land use types and GIS technology, land data should be refreshed regularly, and GIS spatial analysis algorithms should be optimized to maintain data timeliness and authenticity. The use of remote sensing technology and automated GIS tools can enhance the efficiency and precision of data processing. Furthermore, interdisciplinary cooperation to deepen the study of the relationship between land use and ecosystem services can provide a more solid scientific basis for urban planning. In the case of carbon emission models, it is suggested to use more accurate and comprehensive data sources, optimize models according to city characteristics, adjust parameters, and develop user-friendly interfaces to reduce computational complexity and increase the value of model use.

Currently, Xuzhou City’s development trajectory, while promoting urbanization, also faces issues that are not entirely consistent with the principles of an ecological city. For Xuzhou to realize the construction of an ecological city, adjustments are needed in the following aspects: Firstly, it is necessary to optimize the industrial structure, phase out outdated capacities that are highly polluting and energy-consuming, and develop green industries. Secondly, urban greening should be enhanced to increase the green space ratio and the construction of ecological corridors, thereby improving the urban ecological environment. Thirdly, urban infrastructure needs to be perfected, with the promotion of energy-saving and emission-reduction technologies to increase the efficiency of resource utilization. Finally, the ecological concept in urban planning should be strengthened to ensure the harmonious coexistence of urban growth and the natural environment. Through these measures, Xuzhou City will be able to better achieve consistency between urban growth and the goals of sustainable development.

5. Conclusions

Under the background of global climate change, the ecological sustainability of cities is facing severe challenges, especially with relatively less attention given to small and medium-sized cities. Taking Xuzhou City as a case study, this paper conducts a comprehensive analysis of its ecological emergy, land use change, population density, habitat status, carbon emissions and sequestration, as well as enhanced vegetation index (EVI). Neural network models are applied for prediction and assessment of the city’s sustainability status.

The analysis indicates that the main ecological land types in Xuzhou City, ranked by functional importance, are cultivated land, water bodies, forestland, built-up areas, and grasslands, reflecting that despite the continuous advancement of urbanization, ecological elements still dominate. Land use changes reveal that urbanization has led to a reduction in ecological land area, particularly with a significant expansion of built-up areas from 2010 to 2020. The changes in the habitat index are mainly concentrated in the 0.4–0.6 range, indicating a decline in urban ecological status, but without any drastic fluctuations. In terms of carbon emissions, urbanization is accompanied by a surge in emissions, especially in built-up areas, while carbon sequestration is decreasing, although as an ecological pilot city, Xuzhou maintains relatively stable carbon sequestration. The decline in the enhanced vegetation index, particularly the increase within the 0.4–0.5 range, reflects the intensification of urban ecological fragmentation. Through neural network model predictions, the ecological emergy change in Xuzhou City over the next 20–30 years is estimated to be between 15 and 20%. This suggests that while Xuzhou City continues to implement existing ecological improvement measures, it also needs to pay attention to the actual effects and sustainability of these measures.

The research, through the integration of neural network models and GIS methods, has significantly enhanced our understanding and predictive capabilities of urban sustainability. The neural network model is capable of processing and analyzing vast amounts of complex urban data, revealing potential patterns and trends in urban development, thereby providing a more accurate predictive tool for the assessment and planning of urban sustainability. The GIS method aids in visualizing the spatial structure of the city, further optimizing urban layout. The research findings offer a scientific basis for urban management decisions. Through the application of the model, resources can be allocated more effectively, environmental monitoring can be enhanced, and disaster prevention can be improved, thus enhancing the efficiency of urban management and the formulation of sustainable planning policies. For instance, the model can assist policymakers in identifying high-energy-consumption areas to implement targeted energy-saving and emission-reduction measures or to optimize the transportation network to reduce congestion and pollution.

Moreover, the research method exhibits significant scalability and replicability. The neural network model can be adjusted and optimized according to the specific conditions of different cities, and the GIS method is easily applicable in various regions. This means that the research findings are not only suitable for specific cities but can also provide references and lessons for sustainable urban development in other cities, even in different countries, demonstrating a broad applicability and promotional value.

Author Contributions

Conceptualization, C.Z.; methodology, J.Z.; validation, C.Z. and J.Z.; formal analysis, C.Z. and J.Z.; investigation, S.Y. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

All data generated or analyzed during this study are included in this published article.

Conflicts of Interest

The authors declare no conflicts of interest.

Footnotes

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.

Figures and Tables

Figure 1. Geographical location and regions of Xuzhou.

Figure 2. Research framework and roadmap.

Figure 3. Basic model of ecological accounting in Xuzhou City.

Figure 4. Basic urban ecological accounting system.

Figure 5. Optimized urban accounting system framework.

Figure 6. Path diagram of urban pollution treatment system.

Figure 7. Analysis framework based on GIS method.

Figure 8. Indicator-based neural network application model.

Figure 9. Neural network model based on image recognition.

Figure 10. Distribution of energy values for various land types in Xuzhou City in 2000.

Figure 11. Distribution of energy values for various land types in Xuzhou City in 2020.

Figure 12. Display of priority levels for different land parcel data.

Figure 13. The trends of land cover changes in Xuzhou City from 2000 to 2020.

Figure 14. Comparison of ecosystem service values.

Figure 15. GIS distribution map of ecosystem service values.

Figure 16. Display of population density GIS distribution.

Figure 17. Quantitative analysis of population density classification by area.

Figure 18. Habitat quality degree spatial distribution. Note: (A) is a GIS map of Xuzhou based on habitat quality; (B) shows the trend of habitat changes at various levels from 2000 to 2020.

Figure 19. GIS spatial change map of Xuzhou carbon emissions based on time series.

Figure 20. GIS spatial change map of Xuzhou carbon storage based on time series.

Figure 21. Quantitative comparison of carbon emissions and carbon storage.

Figure 22. Quantitative analysis of EVI based on time series and GIS distribution map. Note: (A) shows a comparative analysis of different levels of EVI (Enhanced Vegetation Index) across various time periods; (B) presents a proportional analysis of different EVI values; (C) is a GIS change map based on EVI across multiple dimensions.

Figure 23. Comparison of actual and predicted data trends from 2000 to 2020. Note: (A–D) respectively show the results of four training depths.

Figure 24. Predicted trend from 2030 to 2050.

Table 1

Maximum influence distance and weights of threat factors.

Threat Factor	Maximum Influence Distance	Weight Distance	Decay Function
Cultivated Land	0.5	0.6	Linear
Construction Land	2	0.4	Exponential
Unused Land	6	0.5	Linear

Table 2

Habitat suitability and sensitivity to different threat factors.

LULC	Habitat Suitability	Sensitivity
		Cultivated Land	Construction Land	Unused Land
Cultivated Land	0.4	0	0.5	0.4
Forest Land	0.8	0.6	0.6	0.2
Grassland	0.8	0.5	0.6	0.6
Water Bodies	1	0.3	0.3	0.6
Construction Land	0	0	0	0
Unused Land	0.6	0.6	0.4	0

Table 3

Carbon density of land use types in Xuzhou City.

LULC	Aboveground Carbon Density	Belowground Carbon Density	Soil Carbon Density
Cultivated Land	5.57	1.18	80.42
Forest Land	55.80	7.97	110.20
Grassland	2.13	2.61	86.31
Water Bodies	0.61	0	70.21
Construction Land	0.10	0	63.19
Unused Land	0.10	0	64.58

Table 4

Equivalent value of ecosystem services per unit area in Xuzhou City (CNY/hectare).

Ecosystem Services	Cultivated Land	Forest Land	Grassland	Water Bodies	Construction Land	Unused Land
Food production	2038.07	504.14	700.88	1475.53	0.00	0.00
production of material	451.88	1161.98	1032.87	424.21	0.00	0.00
Water resource supply	−2406.95	602.51	571.77	15,290.14	0.00	0.00
Gas regulation	1641.52	3824.07	3633.48	1420.19	0.00	36.89
Climate regulation	857.65	11,435.33	9609.37	4223.69	0.00	0.00
clean the situation	249.00	3326.08	3172.38	10,236.46	0.00	184.44
Hydrological regulation	2757.39	7125.56	7045.64	188,572.28	0.00	55.33
Soil conservation	959.09	4654.06	4426.58	1715.30	0.00	36.89
Maintaining Nutrients	285.88	356.59	331.99	129.11	0.00	0.00
biodiversity	313.55	4235.99	4020.81	4703.24	0.00	36.89
Aesthetic Landscape	138.33	1856.70	1770.63	3485.93	0.00	18.44
total	7285.41	39,083.01	36,316.40	231,676.10	0.00	368.88

Table 5

List of basic urban accounting system parameters.

1	Emergy Input	$E_J+F_N+F_R$
2	Productive Emergy	EP
3	Output Emergy	Er
4	Lost Emergy	$E_L=K_L\times E_w\times F_R\times F_N$
5	Storage Quantity	$DQ=K_1\times R\times N\times E_L\times Q_U-K_2\times Q_U-K_3\times Q_U$
6	Stored Emergy	$EQ=\left\{\begin{array}{c}DQ>0,E_Q=E_P-E_r-E_L\\ DQ=0,E_Q=0\\ DQ<0,E_Q=DQ\times E_Q/Q\end{array}\right\}$

Table 6

Parameter list for optimizing urban accounting system.

1	Emergy Input	$E_J+F_N+F_R+F_{N*}+F_{R*}+F_{N**}$
2	Productive Emergy	EP
3	Output Emergy	Er + EQL
4	Lost Emergy	$E_L=K_L\times (E_{w**}+E_{w*})\times F_{R*}\times F_{N*}$
5	Storage Capacity	$DQ=K_1\times R\times N\times E_L\times Q_U-K_2\times Q_U-K_3\times Q_U+K_4\times F_b\times Q_U$
6	Stored Emergy	$EQ=\left\{\begin{array}{c}DQ>0,E_Q=E_P+F_b-E_{r*}-E_{QL}-E_L\\ DQ=0,E_Q=0\\ DQ<0,E_Q=DQ\times E_Q/Q\end{array}\right\}$
7	Processing Efficiency	$K_L\times (E_w-E_{w**})\times F_R\times F_N+F_b>F_{R*}+F_{N**}+E_{Ql}$

Appendix A

Table A1

Emergy calculation in Xuzhou city.

Item	Cropland/Farmland	Forest/Woodland	Grassland	Built-Up Land	Water Area	Year	Unit
Solar energy emergy	1.72 × 1019	2.66 × 1018	1.25 × 1017	1.48 × 1018	8.87 × 1018	2000	sej
	1.64 × 1019	1.19 × 1019	6.06 × 1017	8.50 × 1018	9.37 × 1018	2020	sej
Rain (geopotential energy) emergy	1.13 × 1020	1.75 × 1019	8.24 × 1017	9.74 × 1018	5.86 × 1019	2000	sej
	1.08 × 1020	7.51 × 1019	3.80 × 1018	5.89 × 1019	5.34 × 1019	2020	sej
Rain (chemical potential energy) emergy	1.73 × 1020	2.68 × 1019	1.26 × 1018	1.49 × 1019	8.93 × 1019	2000	sej
	1.65 × 1020	1.20 × 1020	6.09 × 1018	9.43 × 1019	8.15 × 1019	2020	sej
Wind energy Emergy	6.76 × 1010	1.04 × 1010	4.90 × 108	5.80 × 109	3.48 × 1010	2000	sej
	6.44 × 1010	4.70 × 1010	2.38 × 109	3.68 × 1010	3.34 × 1010	2020	sej
Geothermal heat emergy	5.45 × 1012	8.44 × 1011	3.97 × 1010	4.67 × 1011	2.82 × 1012	2000	sej
	5.19 × 1012	3.80 × 1012	1.92 × 1011	2.96 × 1012	2.69 × 1012	2020	sej

Table A2

Land use transition matrix.

2000|2020	Cropland	Woodland	Grassland	Water Area	Built-Up Land	Unused Land	Total	Unit
Cropland	8055.7227	6.651	0.0594	75.9969	755.7696	0.0135	8894.2131	km2
Woodland	10.5579	48.6909	0.0504	0.612	2.9646	0.0018	62.8776	km2
Grassland	17.3736	5.0733	1.692	0.0738	4.8519	0.0207	29.0853	km2
Water area	30.4713	0.0621		199.4274	7.8867	0.0081	237.8556	km2
Built-up land	5.7294	0.0045		15.8976	1975.8852	0.0018	1997.5185	km2
Unused land	0.0027		0.0045	0.0072	0.1044	0.0027	0.1215	km2
Total	8119.8576	60.4818	1.8063	292.0149	2747.4624	0.0486	11,221.6716	km2

Table A3

Xuzhou city ecosystem services value (CNY).

EVS	Cropland	Woodland	Grassland	Water Area	Built-Up Land	Unused Land	Total
2000	64,798,006.07	2,457,445.69	1,056,273.25	55,105,457.49	0.00	44.82	123,417,227.33
2005	63,716,030.61	2,154,310.07	654,544.92	64,950,811.32	0.00	38.18	131,475,735.10
2010	61,720,486.35	2,171,334.63	549,921.01	71,598,895.99	0.00	100.59	136,040,738.57
2015	60,186,709.70	2,421,532.32	253,731.76	70,281,122.34	0.00	52.45	133,143,148.57
2020	59,156,507.29	2,363,810.62	65,598.30	67,652,872.83	0.00	17.93	129,238,806.98

Table A4

Population density graded area (km²).

Population Density	2000	2005	2010	2015	2020
0–500	5304.17	5504.15	6106.84	6494.66	6556.77
500–1500	5292.26	5145.98	4401.75	3980.66	3938.40
1500–3000	364.31	293.65	425.25	460.66	457.09
3000–5000	106.15	111.10	125.97	123.84	120.29
>5000	153.02	165.06	160.11	160.10	147.38
Total	11,219.93	11,219.93	11,219.93	11,219.93	11,219.93

Table A5

Proportion of population density graded area (%).

Population Density	2000	2005	2010	2015	2020
0–500	47.27%	49.06%	54.43%	57.89%	58.44%
500–1500	47.17%	45.86%	39.23%	35.48%	35.10%
1500–3000	3.25%	2.62%	3.79%	4.11%	4.07%
3000–5000	0.95%	0.99%	1.12%	1.10%	1.07%
>5000	1.36%	1.47%	1.43%	1.43%	1.31%

Table A6

Enhanced vegetation index (EVI) graded area (km²).

EVI	2000	2005	2010	2015	2020
<0.1	127.54	151.34	138.64	113.29	120.56
0.1–0.2	566.53	415.97	475.45	475.52	580.53
0.2–0.3	3897.14	1845.22	1820.35	1665.16	1676.83
0.3–0.4	6343.98	5552.75	4988.70	4752.42	4088.11
0.4–0.5	286.84	3240.23	3694.59	3781.97	3887.23
>0.5	0.93	17.46	105.23	434.60	869.70
Total	11,222.96	11,222.97	11,222.96	11,222.97	11,222.97

Table A7

Proportion of enhanced vegetation index (EVI) graded area (%).

EVI	2000	2005	2010	2015	2020
<0.1	1.14%	1.35%	1.24%	1.01%	1.07%
0.1–0.2	5.05%	3.71%	4.24%	4.24%	5.17%
0.2–0.3	34.72%	16.44%	16.22%	14.84%	14.94%
0.3–0.4	56.53%	49.48%	44.45%	42.35%	36.43%
0.4–0.5	2.56%	28.87%	32.92%	33.70%	34.64%
>0.5	0.01%	0.16%	0.94%	3.87%	7.75%

Table A8

Area of habitat quality at various grades (km²).

Habitat Quality Index	2000	2005	2010	2015	2020
0–0.2	1997.52	2122.37	2369.86	2587.96	2747.46
0.2–0.4	8887.28	8739.34	8466.01	8256.26	8116.68
0.4–0.6	45.39	29.47	27.94	19.73	12.22
0.6–0.8	279.03	317.65	344.43	345.35	333.56
0.8–1	12.46	12.83	13.43	12.37	11.75
Total	11,221.67	11,221.67	11,221.67	11,221.67	11,221.67

Table A9

Proportion of area for habitat quality at various grades (%).

Habitat Quality Index	2000	2005	2010	2015	2020
0–0.2	17.80%	18.91%	21.12%	23.06%	24.48%
0.2–0.4	79.20%	77.88%	75.44%	73.57%	72.33%
0.4–0.6	0.40%	0.26%	0.25%	0.18%	0.11%
0.6–0.8	2.49%	2.83%	3.07%	3.08%	2.97%
0.8–1	0.11%	0.11%	0.12%	0.11%	0.10%

Table A10

Carbon storage data (unit: ton).

Types/Year	2000	2005	2010	2015	2020
Cropland	77,535,555.73	76,240,892.92	73,853,078.19	72,017,802.20	70,785,089.63
Woodland	1,144,123.21	1,002,991.10	1,010,917.29	1,127,402.87	1,100,529.14
Grassland	267,325.90	165,654.87	139,176.23	64,215.46	16,601.88
Water area	1,684,438.24	1,985,386.48	2,188,602.06	2,148,320.97	2,067,981.85
Built-up land	785.90	669.47	1763.92	919.80	314.36
Unused land	12,643,015.83	13,433,252.38	14,999,706.58	16,380,138.33	17,389,681.55
Total	93,275,244.80	92,828,847.21	92,193,244.27	91,738,799.63	91,360,198.41

Table A11

Carbon storage proportion (%).

Types/Year	2000	2005	2010	2015	2020
Cropland	83.13%	82.13%	80.11%	78.50%	77.48%
Woodland	1.23%	1.08%	1.10%	1.23%	1.20%
Grassland	0.29%	0.18%	0.15%	0.07%	0.02%
Water area	1.81%	2.14%	2.37%	2.34%	2.26%
Built-up land	0.00%	0.00%	0.00%	0.00%	0.00%
Unused land	13.55%	14.47%	16.27%	17.86%	19.03%

Table A12

Carbon emission data (unit: ton).

Types/Year	2000	2005	2010	2015	2020
Cropland	3,860,167.64	5,371,954.65	6,622,301.99	6,420,663.89	6,820,654.55
Woodland	43,728.09	73,554.68	121,323.85	185,594.88	152,370.77
Grassland	8511.31	9437.80	12,050.24	4592.32	2074.33
Water area	70,947.46	120,257.19	237,178.07	273,000.62	279,131.69
Built-up land	0.00	0.00	0.00	0.00	0.00
Unused land	3,515,740.65	5,764,088.14	10,634,884.57	13,468,073.45	15,211,822.46
Total	7,499,095.14	11,339,292.45	17,627,738.71	20,351,925.17	22,466,053.80

Table A13

Carbon emission proportion (%).

Types/Year	2000	2005	2010	2015	2020
Cropland	0.51	0.47	0.38	0.32	0.30
Woodland	0.01	0.01	0.01	0.01	0.01
Grassland	0.00	0.00	0.00	0.00	0.00
Water area	0.01	0.01	0.01	0.01	0.01
Built-up land	0.00	0.00	0.00	0.00	0.00
Unused land	0.47	0.51	0.60	0.66	0.68
Total	0.53	0.49	0.40	1.00	1.00

References

1. China Urban Development Report in 2021. Available online: https://ieaschina.org (accessed on 3 November 2024).

2. Wang, D.; Huang, Z.; Wang, Y.; Mao, J. Ecological security of mineral resource-based cities in China: Multidimensional measurements, spatiotemporal evolution, and comparisons of classifications. Ecol. Indic.; 2021; 132, 108269. [DOI: https://dx.doi.org/10.1016/j.ecolind.2021.108269]

3. Chai, J.; Wang, Z.; Yu, C. Analysis for the Interaction Relationship between Urbanization and Ecological Security: A Case Study in Wuhan City Circle of China. Int. J. Environ. Res. Public Health; 2021; 18, 13187. [DOI: https://dx.doi.org/10.3390/ijerph182413187] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34948794]

4. Zhang, H.; Li, J.; Tian, P.; Pu, R.; Cao, L. Construction of ecological security patterns and ecological restoration zones in the city of Ningbo, China. J. Geogr. Sci.; 2022; 32, pp. 663-681. [DOI: https://dx.doi.org/10.1007/s11442-022-1966-9]

5. Wu, J.; Bai, Z. Spatial and temporal changes of the ecological footprint of China’s resource-based cities in the process of urbanization. Resour. Policy; 2022; 75, 102491. [DOI: https://dx.doi.org/10.1016/j.resourpol.2021.102491]

6. Zhang, S.; Li, F.; Zhou, Y.; Hu, Z.; Zhang, R.; Xiang, X.; Zhang, Y. Using Net Primary Productivity to Characterize the Spatio-Temporal Dynamics of Ecological Footprint for a Resource-Based City, Panzhihua in China. Sustainability; 2022; 14, 3067. [DOI: https://dx.doi.org/10.3390/su14053067]

7. Lu, C.; Wang, S.; Wang, K.; Gao, Y.; Zhang, R. Uncovering the benefits of integrating industrial symbiosis and urban symbiosis targeting a resource-dependent city: A case study of Yongcheng, China. J. Clean. Prod.; 2020; 255, 120210. [DOI: https://dx.doi.org/10.1016/j.jclepro.2020.120210]

8. Zhang, L.; Wang, H.; Zhang, W.; Wang, C.; Bao, M.; Liang, T.; Liu, K.-D. Study on the development patterns of ecological civilization construction in China: An empirical analysis of 324 prefectural cities. J. Clean. Prod.; 2022; 367, 132975. [DOI: https://dx.doi.org/10.1016/j.jclepro.2022.132975]

9. Yang, Y.; Feng, Z.; Wu, K.; Lin, Q. How to construct a coordinated ecological network at different levels: A case from Ningbo city, China. Ecol. Inform.; 2022; 70, 101742. [DOI: https://dx.doi.org/10.1016/j.ecoinf.2022.101742]

10. Sun, F.; Ye, C.; Zheng, W.; Miao, X. Model of Urban Marketing Strategy Based on Ecological Environment Quality. J. Environ. Public Health; 2022; 2022, 8096122. [DOI: https://dx.doi.org/10.1155/2022/8096122]

11. Ma, Q.; Yang, Y. Analysis of Ecological Environment Evaluation and Coupled and Coordinated Development of Smart Cities Based on Multisource Data. J. Sens.; 2022; 2022, 5959495. [DOI: https://dx.doi.org/10.1155/2022/5959495]

12. Fan, Z.; Wang, Y.; Feng, Y. Ecological Livability Assessment of Urban Agglomerations in Guangdong-Hong Kong-Macao Greater Bay Area. Int. J. Environ. Res. Public Health; 2021; 18, 13349. [DOI: https://dx.doi.org/10.3390/ijerph182413349] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34948957]

13. Zhang, J.; Ma, L. Urban ecological security dynamic analysis based on an innovative emergy ecological footprint method. Environ. Dev. Sustain.; 2021; 23, pp. 16163-16191. [DOI: https://dx.doi.org/10.1007/s10668-021-01341-z]

14. Ha, C.; Huang, G.; Zhang, J.; Dong, S. Assessing ecological literacy and its application based on linguistic ecology: A case study of Guiyang City, China. Environ. Sci. Pollut. Res.; 2022; 29, pp. 18741-18754. [DOI: https://dx.doi.org/10.1007/s11356-021-16753-7] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34704227]

15. Huang, L.; Wang, D.; He, C. Ecological security assessment and ecological pattern optimization for Lhasa city (Tibet) based on the minimum cumulative resistance model. Environ. Sci. Pollut. Res.; 2022; 29, pp. 83437-83451. [DOI: https://dx.doi.org/10.1007/s11356-022-21511-4]

16. Wang, D.; Ji, X.; Li, C.; Gong, Y. Spatiotemporal Variations of Landscape Ecological Risks in a Resource-Based City under Transformation. Sustainability; 2020; 13, 5297. [DOI: https://dx.doi.org/10.3390/su13095297]

17. Xia, H.; Liu, Z.; Efremochkin, M.; Liu, X.; Lin, C. Study on city digital twin technologies for sustainable smart city design: A review and bibliometric analysis of geographic information system and building information modeling integration. Sustain. Cities Soc.; 2022; 84, 104009. [DOI: https://dx.doi.org/10.1016/j.scs.2022.104009]

18. Xu, W.; Yi, J.; Shuai, J.; Yu, Z.; Cheng, J. Dynamic evaluation of the ecological civilization of Jiangxi Province: GIS and AHP approaches. PLoS ONE; 2022; 17, e0271768. [DOI: https://dx.doi.org/10.1371/journal.pone.0271768]

19. Wang, X.; Zhang, C.; Wang, C.; Liu, G.; Wang, H. GIS-based for prediction and prevention of environmental geological disaster susceptibility: From a perspective of sustainable development. Ecotoxicol. Environ. Saf.; 2021; 226, 112881. [DOI: https://dx.doi.org/10.1016/j.ecoenv.2021.112881] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34634737]

20. Di Pinto, V.; Rinaldi, A.M.; Rossini, F. Learning from the Informality. Using GIS Tools to Analyze the Structure of Autopoietic Urban Systems in the “Smart Perspective”. ISPRS Int. J. Geo-Inf.; 2021; 10, 202. [DOI: https://dx.doi.org/10.3390/ijgi10040202]

21. Zhu, J.; Wu, P. Towards Effective BIM/GIS Data Integration for Smart City by Integrating Computer Graphics Technique. Remote Sens.; 2021; 13, 1889. [DOI: https://dx.doi.org/10.3390/rs13101889]

22. Xie, H.; Wen, J.; Chen, Q.; Wu, Q. Evaluating the landscape ecological risk based on GIS: A case-study in the Poyang Lake region of China. Land Degrad. Dev.; 2021; 32, pp. 2762-2774. [DOI: https://dx.doi.org/10.1002/ldr.3951]

23. Pallathadka, A.; Pallathadka, L.; Rao, S.; Chang, H.; Van Dommelen, D. Using GIS-based spatial analysis to determine urban greenspace accessibility for different racial groups in the backdrop of COVID-19: A case study of four US cities. GeoJournal; 2021; 87, pp. 4879-4899. [DOI: https://dx.doi.org/10.1007/s10708-021-10538-8] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34744264]

24. He, D.; Hou, K.; Wen, J.F.; Wu, S.Q.; Wu, Z.P. A coupled study of ecological security and land use change based on GIS and entropy method—A typical region in Northwest China, Lanzhou. Environ. Sci. Pollut. Res.; 2022; 29, pp. 6347-6359. [DOI: https://dx.doi.org/10.1007/s11356-021-16080-x] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34449023]

25. Parveen, M.T.; Ilahi, R.A. Assessment of land-use change and its impact on the environment using GIS techniques: A case of Kolkata Municipal Corporation, West Bengal, India. GeoJournal; 2022; 87, pp. 551-566. [DOI: https://dx.doi.org/10.1007/s10708-022-10581-z]

26. Fan, Y.; Fang, C. Evolution process analysis of urban metabolic patterns and sustainability assessment in western China, a case study of Xining city. Ecol. Indic.; 2020; 109, 105784. [DOI: https://dx.doi.org/10.1016/j.ecolind.2019.105784]

27. Chen, Y.; Peng, L.; Cao, W. Health evaluation and coordinated development characteristics of urban agglomeration: Case study of Fujian Delta in China. Ecol. Indic.; 2021; 121, 107149. [DOI: https://dx.doi.org/10.1016/j.ecolind.2020.107149]

28. Wang, Q.; Liu, M.; Tian, S.; Yuan, X.; Ma, Q.; Hao, H. Evaluation and improvement path of ecosystem health for resource-based city: A case study in China. Ecol. Indic.; 2021; 128, 107852. [DOI: https://dx.doi.org/10.1016/j.ecolind.2021.107852]

29. Liu, Y.; Qu, Y.; Cang, Y.; Ding, X. Ecological security assessment for megacities in the Yangtze River basin: Applying improved emergy-ecological footprint and DEA-SBM model. Ecol. Indic.; 2022; 134, 108481. [DOI: https://dx.doi.org/10.1016/j.ecolind.2021.108481]

30. Voukkali, I.; Zorpas, A.A. Evaluation of urban metabolism assessment methods through SWOT analysis and analytical hierocracy process. Sci. Total Environ.; 2022; 807, 150700. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2021.150700]

31. Tang, M.; Hong, J.; Wang, X.; He, R. Sustainability accounting of neighborhood metabolism and its applications for urban renewal based on emergy analysis and SBM-DEA. J. Environ. Manag.; 2020; 275, 111177. [DOI: https://dx.doi.org/10.1016/j.jenvman.2020.111177]

32. Li, Q.; Wu, J.; Su, Y.; Zhang, C.; Wu, X.; Wen, X.; Huang, G.; Deng, Y.; Lafortezza, R.; Chen, X. Estimating ecological sustainability in the Guangdong-Hong Kong-Macao Greater Bay Area, China: Retrospective analysis and prospective trajectories. J. Environ. Manag.; 2022; 303, 114167. [DOI: https://dx.doi.org/10.1016/j.jenvman.2021.114167]

33. Lee, Y.-C.; Liao, P.-T. The effect of tourism on teleconnected ecosystem services and urban sustainability: An emergy approach. Ecol. Model.; 2021; 439, 109343. [DOI: https://dx.doi.org/10.1016/j.ecolmodel.2020.109343]

34. Alizadeh, S.; Zafari-Koloukhi, H.; Rostami, F.; Rouhbakhsh, M.; Avami, A. The eco-effificiency assessment of wastewater treatment plants in the city of Mashhad using emergy and life cycle analyses. J. Clean. Prod.; 2020; 249, 119327. [DOI: https://dx.doi.org/10.1016/j.jclepro.2019.119327]

35. Zhou, H.; Li, H.; Zhao, X.; Ding, Y. Emergy ecological model for sponge cities: A case study of China. J. Clean. Prod.; 2021; 296, 126530. [DOI: https://dx.doi.org/10.1016/j.jclepro.2021.126530]

36. Li, J.; Sun, W.; Song, H.; Li, R.; Hao, J. Toward the construction of a circular economy eco-city: An emergy-based sustainability evaluation of Rizhao city in China. Sustain. Cities Soc.; 2021; 71, 102956. [DOI: https://dx.doi.org/10.1016/j.scs.2021.102956]

37. Zhang, J.; Asutosh, A.T.; Zhang, H.; Yan, Y.; Zhang, Y.; Wei, G.; Ma, C.; Shi, Y.; Gao, Y.; Yan, X. et al. Environmental sustainability in the city of Shanghai municipal solid waste treatment system: An integrated framework of artifcial neural network (ANN) and LCA-emergy methodology. Arab. J. Geosci.; 2022; 15, 1271. [DOI: https://dx.doi.org/10.1007/s12517-022-10537-0]

38. Farzaneh, H.; Dashti, M.; Zusman, E.; Lee, S.-Y.; Dagvadorj, D.; Nie, Z. Assessing the Environmental-Health-Economic Co-Benefifits from Solar Electricity and Thermal Heating in Ulaanbaatar, Mongolia. Int. J. Environ. Res. Public Health; 2022; 19, 6931. [DOI: https://dx.doi.org/10.3390/ijerph19116931] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35682514]

39. Liu, W.; Zhan, J.; Li, Z.; Jia, S.; Zhang, F.; Li, Y. Eco-Efficiency Evaluation of Regional Circular Economy: A Case Study in Zengcheng, Guangzhou. Sustainability; 2018; 10, 453. [DOI: https://dx.doi.org/10.3390/su10020453]

40. Smith, J.; Johnson, R. GIS Analysis of Urban Land Parcel Types: A Study in Sustainable City Planning. J. Urban Plan.; 2023; 25, pp. 45-60.

41. Brown, A.; Green, L. Application of GIS Technology in Ecological Management of Urban Areas. Environ. Sci. Technol. Rev.; 2023; 12, pp. 112-125.

42. Roest, A.H.; Weitkamp, G.; van den Brink, M.; Boogaard, F. Mapping spatial opportunities for urban climate adaptation measures in public and private spaces using a GIS-based Decision Support Model. Sustain. Cities Res.; 2023; 96, 104651. [DOI: https://dx.doi.org/10.1016/j.scs.2023.104651]

43. Wilson, M.; Parker, K. Integrating GIS Analysis of Various Land Parcel Types into Urban Planning Strategies for Environmental Conservation. Urban Ecol. J.; 2023; 15, pp. 30-42.

44. Hossain, M.; Wiegand, B.; Reza, A.; Chaudhuri, H.; Mukhopadhyay, A.; Yadav, A.; Patra, P.K. A machine learning approach to investigate the impact of land use land cover (LULC) changes on groundwater quality, health risks and ecological risks through GIS and response surface methodology (RSM). J. Environ. Manag.; 2024; 366, 121911. [DOI: https://dx.doi.org/10.1016/j.jenvman.2024.121911] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39032255]

45. Lee, C.; Turner, B. Enhancing Disaster Risk Management through GIS Analysis of Urban Land Parcel Types. J. Disaster Resil. Manag.; 2023; 7, pp. 55-68.

46. Allred, S.; Stedman, R.; Heady, L.; Strong, K. Incorporating biodiversity in municipal land-use planning: An assessment of technical assistance, policy capacity, and conservation outcomes in New York’s Hudson Valley. Land Use Policy; 2024; 104, 105344.

47. Patel, H.; Nguyen, M. Sustainable Urban Development Practices Supported by GIS Analysis of Land Parcel Types. Sustain. Cities Communities J.; 2023; 6, pp. 40-54.

48. Ossola, A.; Locke, D.; Lin, B.; Minor, E. Greening in style: Urban form, architecture and the structure of front and backyard vegetation. Landsc. Urban Plan.; 2019; 185, pp. 141-157. [DOI: https://dx.doi.org/10.1016/j.landurbplan.2019.02.014]

49. Hall, S.; Clark, E. Evaluating the Distribution of Built-up Areas and Farmland using GIS Technology for Effective Urban Resource Management. Environ. Plan. Manag. J.; 2023; 11, pp. 80-94.

50. Piras, G.; Muzi, F.; Zylka, C. Integration of BIM and GIS for the Digitization of the Built Environment. Appl. Sci.; 2024; 14, 11171. [DOI: https://dx.doi.org/10.3390/app142311171]

51. Haile, W.; Solomon, T. Risk assessment of gas pipeline using an integrated Bayesian belief network and GIS: Using Bayesian neural networks for external pitting corrosion modelling. Can. J. Chem. Eng.; 2025; 103, pp. 98-109.

52. Shi, J.; Li, J.; Zhang, H.; Xie, B.; Xie, Z.; Yu, Q.; Yan, J. Real-time gas explosion prediction at urban scale by GIS and graph neural network. Appl. Energy; 2025; 377, 124614. [DOI: https://dx.doi.org/10.1016/j.apenergy.2024.124614]

53. Land Use Data. Available online: https://zenodo.org/records/8228112 (accessed on 3 November 2024).

54. DEM Data Source. Available online: https://www.gscloud.cn (accessed on 3 November 2024).

55. Xuzhou City Administrative Division Map Source. Available online: https://www.shengshixian.com/ (accessed on 3 November 2024).

56. Carbon Emission Data. Available online: https://db.cger.nies.go.jp/ged/ja/ (accessed on 3 November 2024).

57. Population Density Data. Available online: https://hub.worldpop.org/ (accessed on 3 November 2024).

58. EVI Data Source. Available online: https://www.nasa.gov/ (accessed on 3 November 2024).

59. Liu, K.; Chen, B.; Wang, S.; Wang, H. An urban waterlogging footprint accounting based on emergy: A case study of Beijing. Appl. Energy; 2023; 348, 121527. [DOI: https://dx.doi.org/10.1016/j.apenergy.2023.121527]

60. Gan, L.; Wan, X.; Ma, Y.; Lev, B. Efficiency evaluation for urban industrial metabolism through the methodologies of emergy analysis and dynamic network stochastic block model. Sustain. Cities Soc.; 2023; 90, 104396. [DOI: https://dx.doi.org/10.1016/j.scs.2023.104396]

61. Liu, Y.; Li, Y.; Liu, G.; Chen, B. Emergy synthesis coupled with urban metabolism: A review. J. Clean. Prod.; 2023; 327, 129327.

62. Chen, W.; Xu, M.; Dong, L. Assessing urban ecological sustainability using emergy analysis: A case study of Shanghai, China. Ecol. Indic.; 2023; 124, 107433.

63. Zheng, Y.; Wang, H.; Chen, Y.; Fu, M. Exploring the synergies between urban form and emergy sustainability: A case study of Beijing, China. Sustain. Cities Soc.; 2024; 84, 102898.

64. Zhao, Y.; Wang, C.; Yang, Z. Emergy-based assessment of urban metabolic efficiency and its implications for sustainable development: A case study of Guangzhou, China. Resour. Conserv. Recycl.; 2023; 178, 106028.

65. Zou, C.; Chen, W.; Lu, H. Integrating emergy analysis into urban ecological planning: A framework and case study in Wuhan, China. Landsc. Urban Plan.; 2024; 227, 104163.

66. Wang, J.; Liu, Y.; Zhang, Q.; Li, W. Application of GIS-based spatial analysis in assessing urban land use impact on ecological sustainability: A case study of London. Sustain. Cities Soc.; 2023; 82, 102766.

67. Chen, L.; Zhang, H.; Shao, G.; Wu, J. Integrating remote sensing and GIS techniques for monitoring land use changes and their impacts on urban ecological sustainability in Beijing, China. Int. J. Appl. Earth Obs. Geoinf.; 2024; 118, 102490.

68. Li, M.; Sun, X.; Liu, S.; Yu, D. Spatial-temporal analysis of urban land use patterns and their implications for ecological sustainability using GIS: A case study of New York City. Habitat Int.; 2023; 127, 105223.

69. Yang, T.; Chen, X.; Li, Z.; Ma, Y. Assessment of urban green space distribution using GIS-based landscape metrics for enhancing ecological sustainability in Shanghai, China. Urban For. Urban Green.; 2024; 62, 127117.

70. Rahman, M.M.; Sarkar, M.A.R.; Islam, M.R.; Uddin, K. Mapping urban heat island effects and land use planning for improving ecological sustainability in Dhaka, Bangladesh: A GIS-based approach. Sustain. Dev.; 2023; 31, pp. 350-363. [DOI: https://dx.doi.org/10.1002/sd.2275]

71. Pan, H.; Zhuang, M.; Geng, Y.; Wu, F.; Dong, H. Emergy-based ecological footprint analysis for a mega-city: The dynamic changes of Shanghai. J. Clean. Prod; 2019; 210, pp. 552-562. [DOI: https://dx.doi.org/10.1016/j.jclepro.2018.11.064]

72. Zhong, S.; Geng, Y.; Kong, H.; Liu, B.; Tian, X.; Chen, W.; Qian, Y.; Ulgiati, S. Emergy-based sustainability evaluation of Erhai Lake Basin in China. J. Clean. Prod; 2018; 178, pp. 142-153. [DOI: https://dx.doi.org/10.1016/j.jclepro.2018.01.019]