Spatiotemporal evolution and driving mechanisms of the transformation of production-living-ecological land in peri-urban areas under the construction of the Guangdong-Hong Kong-Macao greater bay area: a case study of Huiyang district, China

Against the backdrop of rapid development in the Guangdong-Hong Kong-Macao Greater Bay Area (GBA), the spatiotemporal evolution of Production-Living-Ecological Land (PLEL) in its peri-urban regions has emerged as a critical research focus for achieving regional coordinated development. This study employs Huiyang District as a representative case study area, utilizing multisource geospatial data and analytical methodologies including land use dynamic degree, land transfer matrix, binary logistic regression, and ROC analysis to systematically investigate PLEL’s spatiotemporal evolution patterns and underlying driving mechanisms from 2007 to 2023. The empirical results demonstrate that: (1) During the observation period, aggregate production and ecological land areas exhibited a declining trend, while living land manifested persistent expansion, reflecting accelerated urbanization process and intensified population concentration. Specifically, between 2007 and 2015, production and ecological land areas decreased significantly while living land expanded markedly, reflecting accelerated urbanization. From 2015 to 2023, the decline in production land substantially moderated, whereas ecological land diminution accelerated. Living land maintained a consistent expansion throughout both periods. (2) Spatial analysis revealed a stable concentric pattern emerged, characterized by living land at the urban core, encircled by production land, with ecological land occupying the peripheral zones. Notable land conversion dynamics were observed between ecological and production land, revealing the tension between conservation and development. Living land conversions to other categories remained negligible, with its expansion predominantly sourced from production land, suggesting urban expansion has largely occurred through agricultural land appropriation. (3) Driving factor analysis identified population size as the predominant influence. Following GBA development, the impact of population size on living land expansion intensified. Model validation through ROC curve analysis confirmed robust predictive performance. This research systematically elucidates PLEL’s spatiotemporal evolution patterns and driving mechanisms within a GBA peripheral urban unit, thereby contributing scientific insights for territorial spatial optimization in rapidly urbanizing regions.

1 Introduction

The 21st century has witnessed an accelerated pace of global urbanization, characterized by intensified human activity, rapid urban expansion, and significant transformations in land use and land cover. These changes have profoundly reshaped ecological systems and development patterns worldwide (Xu et al., 2023). Projections indicate that by 2050, approximately 68% of the global population will be urban dwellers (Hassanein et al., 2019), making the optimization of Production-Living-Ecological Land (PLEL) structure a central issue for global sustainable development (Song et al., 2024). As one of the most rapidly urbanizing countries globally, China has experienced extensive urban expansion that has contributed to significant challenges, including the loss of arable land and ecological degradation (Wan et al., 2025). In response, the Chinese government formulated the National Territorial Spatial Planning strategy, designed to coordinate PLEL and achieve sustainable development (Bai et al., 2018).

As one of China’s most economically dynamic regions, the GBA has undergone rapid development, accompanied by profound transformations in land use (Yang et al., 2021).This strategic evolution has systematically reconfigured land-use patterns across the GBA, characterized by significantly intensified functional interactions between core cities and their peripheral regions (Cheng et al., 2025). Furthermore, changes in land-use demand induced by industrial relocation and population mobility have resulted in novel spatial differentiations in the trade-offs among the PLEL (Liu et al., 2023). While this dynamic has contributed to substantial gains in economic efficiency, it has concurrently precipitated a suite of challenges, including encroachment upon ecologically sensitive zones and fragmentation of farmland (Long et al., 2021). These adverse effects are particularly pronounced in peripheral regions, such as the eastern flank of the GBA, where the relentless pace of urbanization has starkly outpaced adaptive planning responses. Consequently, there is a pressing need for scientific inquiry to elucidate the evolutionary trajectories of PLEL and to inform evidence-based strategies for coordinated and sustainable regional development.

Scholars worldwide have extensively investigated land-use changes through diverse dimensions. International research, predominantly structured around land-use change modeling frameworks, has systematically synthesized a multitude of representative models to dissect the complex interactions among spatial (resolution and extent), temporal (time steps and duration), and human decision-making (agents and domains) factors. These analyses elucidate the dynamics and patterns of land-use transformation across varying scales and complexities, thereby providing a robust methodological foundation for deciphering the underlying mechanisms of land change. Furthermore, the integration of remote sensing technologies has significantly refined land cover monitoring capabilities, enabling the high-precision identification of land use and land cover conversion (Agarwal et al., 2002; Xie and Niculescu, 2021).

Domestic research has predominantly focused on regions such as the GBA, utilizing the PLEL framework to analyze spatiotemporal land-use dynamics. These studies delineated a clear core-periphery divergence, identifying high-frequency transition zones clustered along administrative boundaries and within economically vibrant corridors. Notably, nearly 80% of the transformed land in some peripheral regions has been reconverted to ecological uses, and spatial hotspots of functional land-use conflicts have been systematically mapped (Lv et al., 2022; Cheng et al., 2025). Parallel research has characterized unique land-use and land-cover change patterns in coastal urban settings (Zhang et al., 2019). To decipher the underlying drivers, quantitative methods such as logistic regression have been employed to quantify the impacts of economic, urbanization, and policy factors (Zhou et al., 2024). Methodologically, the integration of deep learning (e.g., feature enhancement via spectral indices) has significantly refined classification accuracy. Furthermore, studies have advanced multi-scenario simulations and critically examined the evolution of land-use classification systems to enhance prognostic capabilities (Xiang, 2023; Wang Y. et al., 2025; Wang et al., 2025 J.).

However, despite considerable progress, extant research has predominantly examined urban land-use transitions in the GBA and their multi-spatial drivers from a macro perspective (Meng et al., 2024), alongside analyses of ecosystem service trade-offs and synergies (Li et al., 2024). Region-specific studies have likewise concentrated on core cities such as Guangzhou and Shenzhen (Fan et al., 2007; Cheng et al., 2021), leaving peripheral urban areas critically understudied (Doran et al., 2023). A significant gap exists in the systematic investigation of the spatiotemporal evolution and driving mechanisms of the PLEL within these peripheral regions. Methodologically, there is a heavy reliance on statistical descriptions or single-source remote sensing classification, with limited integration of multisource data, including high-resolution imagery (<30 m), point-of-interest (POI) big data, and socio-economic indicators, for quantitative coupling analysis. Furthermore, temporal analyses are often restricted to continuous trends, failing to account for the discontinuous, differential impacts of pivotal policy nodes. These limitations collectively undermine the capacity of existing research to provide precise decision support for territorial spatial optimization in these peripheral regions, potentially aggravating regional disparities and falling short of the strategic demands for coordinated development within the GBA. Consequently, quantifying the impacts of GBA development on PLEL transformations in these peripheral urban units and elucidating the underlying mechanisms have emerged as pressing scientific imperatives. Against this backdrop, Huiyang District—a strategic node in the Greater Bay Area’s eastern corridor—serves as a critical case for examining the evolution and drivers of PLEL. As a representative peripheral city undergoing rapid urbanization and industrial restructuring, its land system exhibits dynamic reorganization of production space, marked living space expansion, and noticeable ecological space compression. The area’s well-documented multisource geospatial data further enables a methodologically robust analysis. Understanding PLEL dynamics in Huiyang not only elucidates how GBA policies reshape peripheral landscapes but also provides empirical support for reconciling development and conservation priorities, offering actionable insights for regional sustainability.

This study systematically elucidates the spatio-temporal evolution of PLEL in Huaiyang area by applying land use dynamic degree, transfer matrix, and binary logistic regression analysis methods. The contributions of this paper are threefold: (1) developing an integrated framework incorporating multisource geospatial data to accurately quantify PLEL dynamics; (2) revealing how major policy interventions differentially reshape PLEL patterns in peripheral areas, capturing often-overlooked nonlinear temporal effects; and (3) establishing quantitative relationships between GBA development strategies and local land-use transformations, thereby supporting evidence-based spatial planning.

2 Materials and methods

2.1 Study area

Huiyang District (Figure 1) is under the administration of Huizhou City in Guangdong Province and is situated in the southeastern part of the province, within the eastern Pearl River Delta Economic Zone. It borders Hong Kong to the south, Huidong County to the east, Shenzhen and Dongguan to the west, and Huicheng District to the north. Serving as a pivotal hub within the “Shenzhen-Dongguan-Huizhou One-Hour Economic Circle”, the district is located approximately 190 km from Guangzhou, 58 km from Shenzhen, and 47 nautical miles from Hong Kong by sea. The district comprises 12 administrative divisions, including Zhenlong Town, Xinxu Town, Yonghu Town, Qiuchang Sub-district, Huiyang Economic Development Zone, Danshui Subdistrict, West Daya Bay Subdistrict, Liangjing Town, Pingtan Town, Shatian Town, Aotou Subdistrict (Daya Bay), and Xichong Subdistrict. The total land area is 917.13 km², which increases to 1,205.44 km² when considering the full extent of the Aotou and Xichong subdistricts. Topographically, the district exhibits an interplay between plains and hills, with extensive low mountains and shallow valleys. The terrain generally slopes from higher elevations in the west to lower elevations in the east, reaching a maximum altitude of 1,003.5 m. The principal rivers are the Xizhi and Danshui Rivers, which flow radially into Daya Bay. The climate is characterized as a humid subtropical monsoon, marked by abundant rainfall and mild temperatures. In 2023, the permanent population was 968,100, with an urbanization rate of 74.91% and a population density of 1,052 persons/km². The GDP reached 80.487 billion CNY, with the secondary and tertiary sectors accounting for 64.9% and 31.9%, respectively. The per capita disposable income was 53,097 CNY, and the urban-rural income ratio was 1.76:1, below the Guangdong provincial average of 2.36:1.

Figure 1

Figure 1. Location and DEM of the Huiyang district.

2.2 Data sources and processing

The datasets used in this study were obtained from multiple open-access and widely recognized sources. These include land use data from the CLCD, population distribution data from LandScan, digital elevation model (DEM) data from NASA, meteorological data (precipitation and temperature) from the National Tibetan Plateau Data Center (NTPDC), and road networks and hydrographic data from OpenStreetMap (OSM). The collective temporal coverage of these datasets spans the years 2007, 2015, and 2023, ensuring a robust foundation for spatiotemporal analysis (Table 1).

Table 1

Table 1. Indicators used in this study and their data sources.

The datasets underwent a comprehensive preprocessing protocol. Initially, all spatial datasets were transformed into the WGS_1984_Albers coordinate system to ensure accurate spatial alignment, which is crucial for subsequent overlay and analytical operations. Subsequently, the datasets were resampled to a uniform resolution of 30 m, ensuring consistent spatial granularity across all datasets. The dimensions of rows and columns in all raster datasets were standardized to maintain uniform spatial dimensions, facilitating seamless integration and comparative analysis. Finally, Min-Max Normalization was applied in ArcGIS to normalize the datasets, thereby eliminating the dimensional disparities among variables.

2.3 Methods

2.3.1 Classification of PLEL

This study established the PLEL classification system for Huiyang District based on the dominant function principle (Liao et al., 2019; Su et al., 2024), which categorizes land types by their primary societal or ecological role. Following current land use classification systems and established PLEL frameworks (Wang W. et al., 2025), we classified: farmland as productive land due to its primary role in providing agricultural products; forests, shrubs, grasslands, water bodies, and bare land as ecological land for their dominant ecosystem services (e.g., climate regulation, biodiversity conservation); and various impervious surfaces as living land given their primary support of human settlements, transportation, and economic activities (Table 2). This function-oriented system ensures alignment with academic consensus and offers a rational foundation for spatial planning and analysis (Su et al., 2024).

Table 2

Table 2. Classification of the PLEL system.

2.3.2 Land use dynamic degree

The land-use dynamic degree (LUDD) is a pivotal quantitative metric for characterizing land-use change dynamics within a specific region. This index calculates the change rate of a specific land-use category’s area over a defined temporal period, thereby capturing both the pace and magnitude of transformation during the study period. LUDD effectively elucidates dynamic patterns and evolutionary trends, enabling the comparison of change intensity across diverse temporal scales. The calculation formula of the LUDD is as follows (Fan et al., 2024):

$LUDD=\frac{U_b-U_a}{U_a}\times \frac{1}{T}\times 100\%(1)$

In Equation 1, LUDD represents the land use dynamic degree; U_a denotes the area of the land use type at the beginning of the study period; U_b signifies the area of the same land use type at the end of the study period; and T indicates the length of the study period in years.

2.3.3 Land use transfer matrix

The land use transfer matrix is a crucial method in the study of land use/cover change (LUCC), which aims to quantify the mutual transformation between different land use types over a specific period and reveal the direction, speed, and structural characteristics of land use change. The expression is (Xu et al., 2025):

$s_{ij}=\left[\begin{array}{cccc}{s}_{11}& s_{12}& \cdots & s_{1n}\\ s_{21}& s_{22}& \cdots & s_{2n}\\ \cdots & \cdots & \cdots & \cdots \\ s_{n1}& s_{n2}& \cdots & s_{nn}\end{array}\right](2)$

In Equation 2, S is the area of land use type; n is the land use type; S_ij represents the area of land from type i to type j at the end of the study.

2.3.4 Min–max normalization method

Min-Max Normalization is a commonly used data normalization method that scales data to a specific range, such as 0 to 1. This method is employed to standardize the traditional features of block image objects by applying a linear transformation to map data values into the designated range, thereby ensuring that all features have a unified scale. This process is also referred to as linear function normalization. The expression is (Ren et al., 2024)

$x_i^{new}=\frac{x_i-x_i^{\mathit{Min}}}{x_i^{\mathit{Max}}-x_i^{\mathit{Min}}}(3)$

In Equation 3, x_i^new is the normalized traditional features of block image object, x_i is the traditional features of the original block image object, x_i^Min and x_i^Max denote the maximum and minimum values of the feature values of class i.

2.3.5 Binary logistic regression model

The binary logistic regression model associates binary dependent variables with multiple independent variables through a logit transformation. The core value is to explain the driving mechanism of “whether an event occurs” and predict its probability. In the study of land use, it can effectively identify the key factors of utilization type conversion, predict the probability of change, and provide a scientific basis for planning and decision-making. The model is as follows (Peng et al., 2021):

$\ln \left(\frac{P}{1-P}\right)=\alpha +{\beta }_1{X}_1+{\beta }_2{X}_2+\dots +{\beta }_n{X}_n(4)$

In Equation 4, prepresents the probability of land use change, α is a constant, and β1, β2, β3, …, βn are the partial regression coefficients of the logistic regression. X1, X2, X3,…, Xn are the explanatory variables.

The probability of land-use change is a nonlinear function composed of explanatory variables. The principle is as follows: in the analysis of binary dependent variables, if the probability of an event is p (0 ≤ p ≤ 1), then 1-p is the probability that the event will not occur. This probability can be calculated using the logistic function. The formula is as follows (Peng et al., 2021):

$p=\frac{\exp \left(\mathrm{\alpha }+{\mathrm{\beta }}_1{x}_1+{\mathrm{\beta }}_2{x}_2+\dots +{\mathrm{\beta }}_n{x}_n\right)}{1+\exp \left(\mathrm{\alpha }+{\mathrm{\beta }}_1{x}_1+{\mathrm{\beta }}_2{x}_2+\dots +{\mathrm{\beta }}_n{x}_n\right)}(5)$

In Equation 5, p is the probability of the occurrence of a land use type, the greater the p value, the greater the possibility of the occurrence of the land use type; x_n denotes the nth driving factor; β is the partial regression coefficient in the binary logistic regression equation; exp (β) represents the ratio of the frequency of occurrence of land use types to the frequency of non-occurrence, which is used to reflect the contribution of relevant influencing factors to the probability of occurrence of a particular land use type.

2.3.6 ROC curve and AUC area

The Relative Operating Characteristic (ROC) proposed by Pontius is typically used to assess the significance of the regression results from the logistic model (Pontius and Schneider, 2001). The area under the curve (AUC) is the area enclosed by the ROC curve and the horizontal axis. Research shows that the value range of AUC is 0.5–1. The higher the AUC, the better the fitting effect of the logistic regression model (Austin and Steyerberg, 2012). When the AUC value exceeds 0.7, the selected driving factors are considered to have good explanatory power for the land use type, indicating that the model has good prediction ability and accuracy (Zhang et al., 2025).

3 Results

3.1 Spatio-temporal variation analysis of PLEL

3.1.1 Analysis of the overall spatial change of PLEL

Overall, the land-use structure of Huiyang District changed from 2007 to 2023. Production and ecological land showed a decreasing trend, whereas living land continued to expand (Figure 2). From 2015 to 2023, the reduction rate of production land slowed, resulting in a total decrease of 5.36 km², equivalent to a 0.16% decrease. The reduction rate of ecological land was accelerated, resulting in a 28.40 km² area reduction, and the dynamic degree reached 0.55%. The living land continued to expand, with an increase of 33.75 km² and a 3.17% growth rate (Table 3).

Figure 2

Figure 2. The land area change map of Huiyang District from 2007 to 2023.

Table 3

Table 3. The change in the land use dynamic degree of PLEL in Huiyang District.

3.1.2 The spatial distribution pattern and evolution of PLEL

The spatial distribution of the PLEL in Huiyang District presents an overall pattern of living land as the core, production land surrounded, and ecological land distributed outside. From 2007 to 2023, this pattern remained relatively stable in space, but there were significant changes in land-use structure and spatial scope.

Firstly, from the perspective of spatial patterns, the living land demonstrates a concentrated patch-like distribution, primarily clustered in relatively flat terrain areas such as Danshui Subdistrict and Aotou Subdistrict of Daya Bay, showing exhibiting agglomeration characteristics. Production land is relatively dispersed in distribution, often interwoven with ecological land or arranged along its edges, mainly concentrated in peripheral towns like Pingtan Town, Liangjing Town, and Yonghu Town. These areas feature relatively flat terrain suitable for agricultural development. Ecological land exhibits extensive contiguous distribution characteristics spatially, serving as the dominant land use type in Huiyang District. It is mainly distributed in the hilly and mountainous areas of eastern and western Huiyang District, including Zhenlong Town, Xinxu Town, Shatian Town, Xichong Subdistrict, and Daya Bay, providing crucial support for the regional ecological environment. These areas are characterized by high elevation, belonging to the Lianhua Mountain Range, with significant topographic relief and extensive forest coverage (Figure 3).

Figure 3

Figure 3. The pie chart and spatial distribution map of the proportion of PLEL in Huiyang District.

From the perspective of spatial evolution characteristics, from 2007 to 2023, the spatial evolution pattern of the three types of land in Huiyang District exhibited the following features: partial contraction of ecological land, patch expansion of living land, and dispersed adjustment of production land. Among these, the spatial patches of living land have significantly expanded, showing a trend of gradual expansion from the center to the peripheral ecological and production lands, particularly notable in the southern and southeastern directions. The spatial distribution of production land has undergone dispersed adjustments, with changes in boundaries between some areas and living or ecological lands. As urban expansion continues, parts of production land have gradually been replaced by living land, especially around the central urban area and near major transportation corridors, where the reduction in production land space is more pronounced. In some areas, ecological land has been encroached upon by living and production lands, leading to a slight reduction in its contiguity, though it still maintains a dominant distribution over a large area.

Secondly, from the perspective of time evolution, from 2007 to 2023, the spatial pattern of the PLEL in the Huiyang District remained stable, but the land structure changed. According to the map of the proportion of the PLEL in 2007, 2015, and 2023 (Figure 3), the PLEL in Huiyang District is mainly comprised of ecological land, followed by production land, and the least is living land. The changes in the proportion of various types of land-use areas are as follows: the proportion of ecological and production land has continued to decrease. During the period from 2007 to 2023, the proportion of ecological land decreased by 3.93%, and the proportion of production land decreased by 2.09%. From 2007 to 2023, the proportion of living land area continued to increase, with an overall increase of 6.02% over 17 years.

3.1.3 Analysis of the structure type of PLEL

According to the land transfer change of Huiyang District (Figure 4), there were more conversions between ecological and production land from 2007 to 2023. The transfer of production and ecological land to living land is far greater than the amount transferred out, and more production land is transferred to it. Through the statistical analysis of the transfer of land use in 2007,2015 and 2023 (Table 4): in 2007–2015, the conversion of ecological land into production land was 46.54 km^2, while the conversion of production land into ecological land was 37.75 km²; in 2015–2023, the conversion of ecological land into production land was 46.85 km², while the conversion of production land into ecological land was 23.38 km². The conversion of living land to other land is less, and the living land is mainly converted from the production land. Between 2007 and 2015, 29.80 km² of production land was converted into living land, and between 2015 and 2023, it was 28.84 km².

Figure 4

Figure 4. The change of land use transfer in Huiyang District from 2007 to 2023.

Table 4

Table 4. The transfer matrix of PLEL in Huiyang District from 2007 to 2015 and from 2015 to 2023.

From the perspective of temporal and spatial distribution, from 2007 to 2015, the conversion of ecological land mainly occurred in Daya Bay West District Street, Daya Bay Xiayong street, Daya Bay Aotou street and Danshui street, and most of them were reduced, that is, ecological land was converted to other land (Figure 5a); the reduction of production land is distributed in the whole Huiyang District, with more transfer out in Qiuchang street, Huiyang economic development zone and Daya Bay West District, and more transfer in in Shatian town and Danshui Street (Figure 5c); While the living land is mainly transferred in, that is, other land is transferred to living land, and the transferred areas are concentrated in the streets of Daya Bay West and Qiuchang streets, with very few transferred out (Figure 5e).

Figure 5

Figure 5. The distribution of land use change in Huiyang District from 2007 to 2015 and from 2015 to 2023. (a,b) represent ecological land, (c,d) represent production land, (e,f) represent living land.

From 2015 to 2023, the conversion of ecological land remains dominated by reduction. The reduction in Xinxu Town and Huiyang Economic Development Zone is increasing, while the reduction in the Daya Bay area is easing (Figure 5b), and the transferred area is relatively small. The production land is still being transferred out more than it is being transferred in, and the transferred areas are primarily located in the Daya Bay West District, specifically on Street and Qiuchang Street. The transferred part is more distributed in Xinxu Town (Figure 5d). The living land area is still increasing, with the majority of the increase concentrated in the streets of Daya Bay West District (Figure 5f).

3.2 Analysis of driving factors of spatial-temporal pattern evolution of PLEL

The study highlights that the spatial pattern evolution of PLEL is the result of the long-term influence of natural factors and the short-term impact of socio-economic factors, and its influence changes dynamically over time (Hu et al., 2025). Based on the driving factor index selected by other scholars in the study on land use driving mechanism (Fan et al., 2023) and the actual situation of Huiyang District, this study selected elevation, slope, temperature, precipitation, population, distance to primary road, distance to secondary road, distance to tertiary road, distance to highway and distance to water. These ten quantifiable driving factors (Figure 6) align with the actual and quantifiable factors identified in the Huiyang District. And these driving factors were subsequently normalized. A logistic regression model was used to quantitatively analyze the driving mechanism of PLEL spatial pattern evolution and identify the dominant factors affecting the spatial differentiation of the study area. By comparing and analyzing the driving factors in 2015 and 2023, the spatial and temporal changes in land use in Huiyang District are examined before and after the construction of the GBA.

Figure 6

Figure 6. Normalized driving factors of LUCC.

3.2.1 Logistic regression analysis and results

3.2.1.1 Collinear diagnostic analysis of driving factors

Research indicates that in a significant sample, multicollinearity among independent variables in logistic regression can impact the accuracy of the regression results (Lukman et al., 2023). Therefore, it is necessary to conduct a collinearity diagnosis using indicators such as the Variance Inflation Factor (VIF) and Tolerance to improve the model’s robustness. When the tolerance (Tol) is less than 0.1, it shows that there is a high degree of collinearity between the independent variables; at the same time, the VIF value greater than 10 is used as the decision threshold (Setshedi and Modirwa, 2020), and the VIF value is positively correlated with the degree of collinearity. The predictive variables with high collinearity are excluded.

According to Table 5, the Tol of all driving factors is greater than 0.1, and the VIF is less than 10, indicating that there is no significant collinearity between driving factors. The selected driving factors were included in the regression model for analysis.

Table 5

Table 5. Results of collinearity diagnosis.

3.2.1.2 Logistic regression analysis

Binary logistic stepwise regression analysis was performed on the 10 driving factors and three types of spaces to obtain the β values of various spatial types and driving factors. The results are presented in Table 6.

Table 6

Table 6. The regression coefficient of each driven continuous factor.

From 2007 to 2015, before the construction of the GBA, the PLEL in Huiyang District was driven by multiple factors, including natural, social, and economic considerations, as well as location. Among them, the driving factors that have a greater impact on the distribution of ecological land are temperature and population. The regression coefficients for the distance to secondary and tertiary roads, as well as the distance to the river system, are positive, indicating that as these factors increase, the probability of other types of space being converted into ecological land is greater. In contrast, the elevation, temperature, precipitation, population, distance from the primary road, and distance to the highway are negative coefficients. It shows that as the number of variables increases, the probability of other spaces being converted into ecological land decreases. Production land is primarily influenced by temperature, and the positive coefficients for elevation, distance to primary roads, and highways indicate that these independent variables are positively correlated with the distribution of production land. When the value of these factors increases, the spatial agglomeration of production land becomes stronger. The negative impact of production land is mainly restricted by population size. The larger the population, the less evenly the production land is distributed. The positive effects of temperature primarily influence the living land. The higher the temperature, the more the living land is distributed, and the greater the probability of the living land space being transferred in. However, due to the negative influence of the distance to the highway, the closer the distance to the highway is, the greater the probability of the living space being transferred out.

From 2015 to 2023, after the construction of the GBA, the driving mechanism for the spatial and temporal distribution of PLEL in Huiyang District remained largely unchanged from that before the GBA’s construction. In general, temperature and population factors were still dominant, but the degree and direction of each factor were adjusted accordingly. For example, the impact of precipitation on ecological land increased in 2023 and became positive. Although the impact of temperature remained significant, it decreased, whereas the impact of population factors increased substantially. Secondly, in terms of production land, except for the influence of slope and distance from the tertiary road, which becomes insignificant, the other factors have not changed significantly. In the living land, the influence of temperature decreases significantly, and the influence of the population factor increases greatly. The influence of the distance from the secondary road changes from insignificant to negative, the distance from the tertiary road changes from positive to insignificant, and the distance from the highway changes from negative to insignificant.

In summary, the comprehensive effect of natural, social, and locational factors shapes the spatial and temporal patterns of PLEL. The construction of the GBA impacts the distribution of land use in Huiyang District through multiple pathways, including transportation, population, industry, economy, and ecology. Future land planning needs to coordinate elevation and slope, climate change, population distribution, road networks, ecological protection, and economic development to achieve regional sustainable use.

3.2.1.3 ROC regression results test

In the Logistic regression model, due to the lack of a traditional R-squared goodness-of-fit index, the ROC curve is widely used to evaluate the model’s performance. Usually, when the AUC exceeds 0.7, it can be considered that the driving factor has a significant explanatory power for specific land use types. The results of the ROC test in this study are as follows (Figure 7).

Figure 7

Figure 7. The ROC curve of the PLEL.

It can be seen from Figure 7; Table 7 that the ROC curves of the three types of land are above the diagonal, and the AUC values are higher than 0.7, indicating that the driving factors included in the model significantly affect the probability of land distribution, rather than random fluctuations.

Table 7

Table 7. The AUC of the PLEL.

4 Discussion

Driven by national strategies including the GBA initiative, rapid urbanization and regional integration have fundamentally reconfigured land use patterns in peri-urban regions. These transitional zones between urban and rural areas epitomize the fundamental tensions between development demands and ecological conservation. Empirical analysis of Huiyang District (2007–2023) demonstrates significant restructuring of PLEL, characterized by continuous living space expansion alongside shrinking productive and ecological areas, with 2015 emerging as a critical policy-induced turning point. These transformations necessitate a systematic investigation of their socioeconomic, policy, and spatial drivers. Accordingly, we situate our empirical findings within regional development theory and sustainable land governance frameworks to elucidate the complex mechanisms underlying PLEL transformations in rapidly urbanizing peripheral regions.

4.1 The expansion of living land and contraction of production and ecological land

The land use transition in Huiyang District demonstrates a typical urbanization pattern, characterized by the expansion of living space through the encroachment on production and ecological spaces (Cruz-Bello et al., 2023). The direct conversion of 58.6 km²of cultivated land into living land reveals the region’s persistent reliance on the land finance model (He et al., 2025), which conflicts with national cultivated land protection strategies (Xin and Qian, 2025). Notably, the ecological land reduction rate doubled after 2015, a trend that likely reflects the shift of urban expansion pressure towards ecological spaces following the tightening of cultivated land protection policies (Tang et al., 2025). This finding confirms the vulnerability of ecological land during the rapid development of the GBA. The results emphasize the urgent need to establish ecological space regulation mechanisms that are equally stringent as cultivated land protection, in order to resolve the increasingly prominent conservation-development contradiction in the regional integration process.

Although 2015 marks a clear turning point in the observed land use trajectories, this study uses the year primarily as a contextual reference associated with the launch of the GBA initiative. The intent is not to attribute causality to specific policies, particularly because spatially explicit data for individual GBA policy instruments remain unavailable. Thus, policy context functions as a qualitative interpretive lens rather than a quantitative model input. This clarification helps avoid over-interpreting the role of policy while acknowledging that the structural changes observed after 2015 are consistent with the broader institutional environment introduced by the GBA development process.

4.2 Policy-driven spatial restructuring: from production to living core

Under the policy-driven framework of the GBA development, the spatial pattern of Huiyang District is undergoing a fundamental restructuring from a production-dominated to a living-core configuration. Although quantitative modeling has not been conducted, the industrial spillover and population decentralization policies from Shenzhen have catalyzed a functional transformation in Huiyang, which is spatially manifested as the axial expansion of living land from the central areas toward the southeastern periphery, primarily through the conversion of former production land. This phenomenon corroborates the assertion by Fan et al. that industrial relocation is far from a mere scale expansion but systematically reconstructs the land functions and spatial structure of the receiving areas (Fan et al., 2020). The year 2015 marks a critical policy intervention point, after which the dynamic changes in production land have slowed, while the loss of ecological land has accelerated, reflecting the tension between development and ecological conservation. Consequently, the case of Huiyang profoundly reveals that regional policies, by guiding industrial distribution, and integrating multidimensional pathways such as transportation, population, and ecology, serve as a critical force in reshaping local spatial patterns and driving their fundamental transformation.

4.3 Strengthening influence of population and economic factors under the GBA development

Logistic regression model further confirms the transformation of the driving mechanism: temperature, elevation and traffic accessibility as constant core factors, highlighting the fundamental role of natural base and infrastructure (Bikis et al., 2025); However, the influence of the population factor significantly strengthened after the initiation of the Greater Bay Area development, as evidenced by a substantial increase in its regression coefficient. This finding corroborates the mechanism proposed by Tian and Mao, wherein regional integration policies intensify the correlation between population agglomeration and construction land expansion (Tian and Mao, 2022). It reflects how the GBA strategy, through industrial synergy, has markedly enhanced the population radiation effect. The high β-value of the temperature factor may stem from the fact that in Huizhou’s South subtropical region, higher temperatures are typically associated with flat river valleys, plains, and coastal areas. These zones not only represent prime traditional agricultural regions but are also preferred for urban expansion. The elevated β-value essentially reveals the highest probability pathway for land transitioning from production to living functions, highlighting the phenomenon of urban development areas expanding into high-quality agricultural land (Zhao et al., 2025). Simultaneously, the impact of transportation factors demonstrates characteristics of policy regulation. The shift of the highway distance effect on ecological land from significant to non-significant reflects policy intervention moderating the ecological impact of infrastructure. The eco-friendly traffic development emphasized in Huizhou City’s key points for promoting the construction of GBA in 2024 is an example (Huizhou Daily, 2024).

The two-stage comparison reveals that 2015 marks a significant threshold point for policy intervention: the dynamic degree of production land in the early stage of spontaneous urbanization is 0.59%, while the later policy-driven slowdown is reduced to 0.16%. Additionally, the dynamic degree of ecological land loss decreases from 0.31% to 0.55%. This contrast highlights the tension between regional integration policies aimed at protecting development and protecting ecology, and confirms the driving mechanism of regional integration policies, as outlined in the PLEL framework (Deng and Yang, 2021). Therefore, future land and space planning needs to transform traffic advantages into spatial governance advantages, transform population growth into green asset growth, and achieve a development-ecology win-win through Transit-Oriented Development, ecological compensation, and carbon sink trading (Lyu et al., 2023).

4.4 Research limitations and implications for future work

This study has several limitations. First, although 2015 is used as a policy turning point, the spatial effects of specific GBA measures were not quantitatively modeled due to limited data availability. The policy context thus provides qualitative support for interpreting land-use shifts rather than serving as an explicit explanatory variable. Second, driver analysis was limited to 2015 and 2023 due to data constraints, preventing a full tri-temporal comparison from 2007. The 30-m resolution of land-use data limits the detection of small or mixed parcels, and classification uncertainty may influence the magnitude of observed PLEL transitions. In addition, socioeconomic factors are underrepresented, with economic variables such as industrial structure, land prices, and government investment lacking. Population and transportation accessibility serve as indirect indicators of economic activity, limiting the model’s ability to comprehensively explain the mechanisms behind the expansion of living land. Furthermore, the binary logistic regression model used in this study to analyze driving factors can effectively identify the main driving factors, but its inherent structure is difficult to fully adapt to complex interaction terms. When considering the interactions between variables, the complexity of the model and the problem of multicollinearity will significantly increase, especially in cases involving a large number of variables. This simplification process to some extent limits the comprehensive understanding of the dynamics of complex human earth systems.

Future research should: (1) develop a policy quantification framework to assess the directional impact of regional initiatives; (2) incorporate longer time-series data for dynamic driver analysis; (3) employ higher-resolution imagery and advanced classification techniques to improve the recognition accuracy of small-scale or mixed use land types; and (4) integrate a wider range of socio-economic indicators, such as industrial structure, to provide a more comprehensive explanation of the mechanism behind the expansion of living land; (5) explore more advanced modeling techniques, such as machine learning algorithms or geographically weighted regression, to better capture the complex interactions between variables. These steps will enhance the understanding and predictive ability of the mechanisms underlying changes in PLEL in peri-urban regions, providing more scientific decision-making references for the sustainable development of suburban areas in the context of the GBA.

5 Conclusion

Taking Huiyang District as an example, this study explores the spatial-temporal evolution and driving mechanism of the PLEL in peripheral urban units under the background of the construction of GBA, aiming to reveal the driving mechanism of land use in the process of regional integration and provide a reference for the optimization and sustainable development of land space in related regions. The research shows that the pattern of PLEL in Huiyang District has experienced significant changes in space and driving factors from 2007 to 2023.

Spatially, Huiyang District’s PLEL pattern experienced significant shifts, with consistent expansion of living land, primarily manifesting as a Danshui-Aotou dual-core radiation. Concurrently, both production and ecological land areas decreased. The expansion of living land predominantly depended on the transformation of production land, reflecting strong urbanization and population agglomeration demands. The analysis across 2007–2015 and 2015–2023 periods revealed distinct evolutionary models: an initial phase characterized by rapid production land reduction and living land increase, followed by a period where ecological land faced accelerated pressures despite continued living land growth. This underscores that urban expansion consistently relies on agricultural land appropriation, and highlights the increasing urgency of ecological protection in the later period.

In terms of driving factors, the evolution of PLEL in Huiyang District is influenced by natural, socio-economic, and locational factors. Elevation is the core background factor, and traffic distance has a significant impact. After the construction of the GBA, the impact of population on the living land has increased significantly. The construction of the GBA has profoundly impacted the land use pattern of Huiyang District through transportation infrastructure development, population mobility, industrial adjustments, and ecological policies. The verification results of the logistic model indicated that the natural-socioeconomic-location triple factor had robust and significant explanatory power for the land use distribution in the study area.

In summary, this study enhances the theoretical understanding of land use transformation within regional integration by elucidating the policy-driven factor transition mechanism. The integrated analytical framework developed for quantifying PLEL changes provides methodological support for territorial spatial optimization in Huiyang District and serves as a reference for sustainable development in comparable regions undergoing transition. At the same time, this study acknowledges limitations such as classification uncertainty, resolution constraints, insufficient policy quantification, and incomplete temporal coverage. Future improvements will be pursued through higher-resolution data, longer time series, and integration with socioeconomic indicators to enhance the study’s mechanism analysis and predictive capabilities.

Data availability statement

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding authors.

Author contributions

BX: Writing – original draft, Methodology, Investigation, Conceptualization, Validation, Project administration, Data curation. LD: Data curation, Investigation, Visualization, Formal Analysis, Writing – original draft. GL: Writing – original draft, Formal Analysis, Investigation, Data curation, Visualization. WP: Formal Analysis, Data curation, Writing – original draft, Investigation, Visualization. XL: Writing – review and editing, Validation. ZZ: Funding acquisition, Writing – review and editing, Resources. HY: Writing – review and editing, Funding acquisition, Resources.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This study has been conducted with support from the Guangdong Provincial Department of Education Characteristic Innovation Project (grant number 2024KTSCX092), Huizhou Science and Technology Project (grant number 2022CQ010026), Huizhou Philosophy and Social Sciences Planning Project (grant number HZSK2024GJ067), and the Natural Science Foundation of Guangdong Province (2024A1515012440).

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Agarwal, C., Green, G. M., Grove, J. M., Evans, T. P., and Schweik, C. M. (2002). A review and assessment of land-use change models: dynamics of space, time, and human choice. Newtown Square, PA: U.S. Department of Agriculture, Forest Service. doi:10.2737/NE-GTR-297

CrossRef Full Text | Google Scholar

Austin, P. C., and Steyerberg, E. W. (2012). Interpreting the concordance statistic of a logistic regression model: relation to the variance and odds ratio of a continuous explanatory variable. BMC Med. Res. Methodol. 12, 82. doi:10.1186/1471-2288-12-82

PubMed Abstract | CrossRef Full Text | Google Scholar

Bai, Y., Wong, C. P., Jiang, B., Hughes, A. C., Wang, M., and Wang, Q. (2018). Developing China’s ecological redline policy using ecosystem services assessments for land use planning. Nat. Commun. 9, 3034. doi:10.1038/s41467-018-05306-1

PubMed Abstract | CrossRef Full Text | Google Scholar

Bikis, A., Engdaw, M., Pandey, D., and Pandey, B. K. (2025). The impact of urbanization on land use land cover change using geographic information system and remote sensing: a case of mizan aman city Southwest Ethiopia. Sci. Rep. 15, 12014. doi:10.1038/s41598-025-94189-6

PubMed Abstract | CrossRef Full Text | Google Scholar

Cheng, L., Feng, R., and Wang, L. (2021). Fractal characteristic analysis of urban land-cover spatial patterns with spatiotemporal remote sensing images in Shenzhen city (1988–2015). Remote Sens. 13, 4640. doi:10.3390/rs13224640

CrossRef Full Text | Google Scholar

Cheng, Z., Yang, J., and Xue, D. (2025). Conflicts at the crossroads: unpacking land-use challenges in the greater Bay area with the “production–living–ecological” perspective. Land 14, 249. doi:10.3390/land14020249

CrossRef Full Text | Google Scholar

Cruz-Bello, G. M., Galeana-Pizaña, J. M., and González-Arellano, S. (2023). Urban growth in peri-urban, rural and urban areas: mexico City. Build. Cities 4, 1–16. doi:10.5334/bc.230

CrossRef Full Text | Google Scholar

Deng, Y., and Yang, R. (2021). Influence mechanism of production-living-ecological space changes in the urbanization process of Guangdong province, China. Land 10, 1357. doi:10.3390/land10121357

CrossRef Full Text | Google Scholar

Doran, E., Golden, J., Matus, K., Lebel, L., Timmer, V., Van ‘T Zelfde, M., et al. (2023). The emerging role of mega-urban regions in the sustainability of global production-consumption systems. Npj Urban Sustain 3, 23. doi:10.1038/s42949-023-00098-w

CrossRef Full Text | Google Scholar

Fan, F., Weng, Q., and Wang, Y. (2007). Land use and land cover change in Guangzhou, China, from 1998 to 2003, based on landsat TM/ETM+ imagery. Sensors 7, 1323–1342. doi:10.3390/s7071323

CrossRef Full Text | Google Scholar

Fan, P., Yue, W., Zhang, J., Huang, H., Messina, J., Verburg, P. H., et al. (2020). The spatial restructuring and determinants of industrial landscape in a mega city under rapid urbanization. Habitat Int. 95, 102099. doi:10.1016/j.habitatint.2019.102099

CrossRef Full Text | Google Scholar

Fan, L., Cai, T., Wen, Q., Han, J., Wang, S., Wang, J., et al. (2023). Scenario simulation of land use change and carbon storage response in Henan Province, China: 1990–2050. Ecol. Indic. 154, 110660. doi:10.1016/j.ecolind.2023.110660

CrossRef Full Text | Google Scholar

Fan, W., Yang, X., Cai, S., Ou, H., Zhou, T., and Wang, D. (2024). Land-Use/Cover change and driving forces in the pan-pearl river basin during the period 1985–2020. Land 13, 822. doi:10.3390/land13060822

CrossRef Full Text | Google Scholar

Hassanein, H., Zorba, N., Han, S., Kanhere, S. S., and Shukair, M. (2019). Crowd management. IEEE Commun. Mag. 57, 18–19. doi:10.1109/MCOM.2019.8703458

CrossRef Full Text | Google Scholar

He, K., Tan, Z., and Tang, Z. (2025). Sowing uncertainty: assessing the impact of economic policy uncertainty on agricultural land conversion in China. Systems 13, 466. doi:10.3390/systems13060466

CrossRef Full Text | Google Scholar

Hu, W., Cheng, J., Zheng, M., Jin, X., Yao, J., and Guo, F. (2025). A multi-scenario simulation and driving factor analysis of production–living–ecological land in China’s main grain producing areas: a case study of the Huaihe river eco-economic belt. Agriculture 15, 349. doi:10.3390/agriculture15030349

CrossRef Full Text | Google Scholar

Huizhou Daily (2024). Huizhou’s key tasks for advancing the development of the Guangdong-Hong Kong-macao greater Bay area in 2024. Available online at: https://hmo.sz.gov.cn/zx/gqdt/content/post_1137481.html (Accessed July 27, 2025).

Google Scholar

Li, H., Xu, Q., Qiu, H., Du, J., Xu, Z., Liu, L., et al. (2024). Study on ecosystem service trade-offs and synergies in the guangdong–Hong kong–Macao greater Bay area based on ecosystem service bundles. Land 13, 2086. doi:10.3390/land13122086

CrossRef Full Text | Google Scholar

Liao, G., He, P., Gao, X., Deng, L., Zhang, H., Feng, N., et al. (2019). The production–living–ecological land classification system and its characteristics in the hilly area of Sichuan province, southwest China based on identification of the main functions. Sustainability 11, 1600. doi:10.3390/su11061600

CrossRef Full Text | Google Scholar

Liu, M., Chen, G., Li, G., Huang, Y., Luo, K., and Zhan, C. (2023). Landscape evolution and its driving forces in the rapidly urbanized Guangdong–Hong Kong–Macao greater Bay area, a case Study in Zhuhai city, South China. Sustainability 15, 13045. doi:10.3390/su151713045

CrossRef Full Text | Google Scholar

Long, H., Zhang, Y., Ma, L., and Tu, S. (2021). Land use transitions: progress, challenges and prospects. Land 10, 903. doi:10.3390/land10090903

CrossRef Full Text | Google Scholar

Lukman, A. F., Kibria, B. M. G., Nziku, C. K., Amin, M., Adewuyi, E. T., and Farghali, R. (2023). K-L estimator: dealing with multicollinearity in the logistic regression model. Mathematics 11, 340. doi:10.3390/math11020340

CrossRef Full Text | Google Scholar

Lv, C., Lian, A., Wang, Z., Jia, T., Sun, X., and Dong, R. (2022). A spatiotemporal pattern analysis of high-frequency land-use changes in the Guangdong–Hong Kong–Macao greater Bay area, from 1990 to 2018. Land 12, 102. doi:10.3390/land12010102

CrossRef Full Text | Google Scholar

Lyu, X., Li, X., Wang, K., Zhang, C., Dang, D., Dou, H., et al. (2023). Strengthening grassland carbon source and sink management to enhance its contribution to regional carbon neutrality. Ecol. Indic. 152, 110341. doi:10.1016/j.ecolind.2023.110341

CrossRef Full Text | Google Scholar

Meng, Y., Sing Wong, M., Kwan, M.-P., Pearce, J., and Feng, Z. (2024). Assessing multi-spatial driving factors of urban land use transformation in megacities: a case study of Guangdong–Hong Kong–Macao greater Bay area from 2000 to 2018. Geo-spatial Inf. Sci. 27, 1090–1106. doi:10.1080/10095020.2023.2255033

CrossRef Full Text | Google Scholar

Peng, Y., Yang, F., Zhu, L., Li, R., Wu, C., and Chen, D. (2021). Comparative analysis of the factors influencing land use change for emerging industry and traditional industry: a case study of Shenzhen City, China. Land 10, 575. doi:10.3390/land10060575

CrossRef Full Text | Google Scholar

Pontius, R. G., and Schneider, L. C. (2001). Land-cover change model validation by an ROC method for the Ipswich watershed, Massachusetts, USA. Agric. Ecosyst. and Environ. 85, 239–248. doi:10.1016/s0167-8809(01)00187-6

CrossRef Full Text | Google Scholar

Ren, Y., Xie, Z., and Zhai, S. (2024). Urban land use classification model fusing multimodal deep features. IJGI 13, 378. doi:10.3390/ijgi13110378

CrossRef Full Text | Google Scholar

Setshedi, K. L., and Modirwa, S. (2020). Socio-economic characteristics influencing small-scale farmers’ level of knowledge on climate-smart agriculture in mahikeng local municipality, North West province, South Africa. S Afr. Jnl. Agric. Ext. 48. doi:10.17159/2413-3221/2020/v48n2a544

CrossRef Full Text | Google Scholar

Song, W., Cao, S., Du, M., and He, Z. (2024). Aligning territorial spatial planning with sustainable development goals: a comprehensive analysis of production, living, and ecological spaces in China. Ecol. Indic. 160, 111816. doi:10.1016/j.ecolind.2024.111816

CrossRef Full Text | Google Scholar

Su, H., Gu, C., and Cao, G. (2024). Land use classification system of agricultural space in territorial planning. Jouranal Nat. Resour. 39, 2570. doi:10.31497/zrzyxb.20241105

CrossRef Full Text | Google Scholar

Tang, W., Li, J., Yan, H., Cha, L., Yang, Y., and Wang, L. (2025). Multi-scenario simulation of land use change based on the objectives of cultivated land, ecological protection, and economic development in Yunnan province, China. Sci. Rep. 15, 37506. doi:10.1038/s41598-025-21568-4

PubMed Abstract | CrossRef Full Text | Google Scholar

Tian, Y., and Mao, Q. (2022). The effect of regional integration on urban sprawl in urban agglomeration areas: a case study of the Yangtze river Delta, China. Habitat Int. 130, 102695. doi:10.1016/j.habitatint.2022.102695

CrossRef Full Text | Google Scholar

Wan, N.-F., Fu, L., Dainese, M., Kiær, L. P., Hu, Y.-Q., Xin, F., et al. (2025). Pesticides have negative effects on non-target organisms. Nat. Commun. 16, 1360. doi:10.1038/s41467-025-56732-x

PubMed Abstract | CrossRef Full Text | Google Scholar

Wang, J., Liu, J., Wang, T., and Jia, M. (2025a). Scenario based land use simulation for climate adaptability in coastal urban agglomerations of Guangdong Hong Kong and Macao Bay area. Sci. Rep. 15, 27489. doi:10.1038/s41598-025-11866-2

PubMed Abstract | CrossRef Full Text | Google Scholar

Wang, W., Zhao, Y., Ma, C., and Dong, S. (2025b). Identification and analysis of production–living–ecological space based on multi-source geospatial data: a case study of Xuzhou city. Sustainability 17, 886. doi:10.3390/su17030886

CrossRef Full Text | Google Scholar

Wang, Y., Zhang, W., Liu, X., Peng, H., Lin, M., Li, A., et al. (2025c). A deep learning method for land use classification based on feature augmentation. Remote Sens. 17, 1398. doi:10.3390/rs17081398

CrossRef Full Text | Google Scholar

Xiang, H. (2023). The evolution and driving mechanism of land use classification systems in China. Sci. Rep. 13, 20644. doi:10.1038/s41598-023-48093-6

PubMed Abstract | CrossRef Full Text | Google Scholar

Xie, G., and Niculescu, S. (2021). Mapping and monitoring of land cover/land use (LCLU) changes in the crozon peninsula (Brittany, France) from 2007 to 2018 by machine learning algorithms (support vector machine, Random forest, and convolutional neural network) and by post-classification comparison (PCC). Remote Sens. 13, 3899. doi:10.3390/rs13193899

CrossRef Full Text | Google Scholar

Xin, S., and Qian, H. (2025). From urban planning to territorial spatial planning: the evolution of china’s planning system and the persistent barriers to urban–rural integration. Land 14, 1520. doi:10.3390/land14081520

CrossRef Full Text | Google Scholar

Xu, C., Zhang, Q., Yu, Q., Wang, J., Wang, F., Qiu, S., et al. (2023). Effects of land use/cover change on carbon storage between 2000 and 2040 in the Yellow river basin, China. Ecol. Indic. 151, 110345. doi:10.1016/j.ecolind.2023.110345

CrossRef Full Text | Google Scholar

Xu, Z., Ke, F., Yu, J., and Zhang, H. (2025). Multi-scale analysis of land use transition and its impact on ecological environment quality: a case study of Zhejiang, China. Land 14, 1569. doi:10.3390/land14081569

CrossRef Full Text | Google Scholar

Yang, C., Liu, H., Li, Q., Cui, A., Xia, R., Shi, T., et al. (2021). Rapid urbanization induced extensive forest loss to urban land in the Guangdong-Hong Kong-Macao greater Bay area, China. Chin. Geogr. Sci. 31, 93–108. doi:10.1007/s11769-021-1177-9

CrossRef Full Text | Google Scholar

Zhang, C., Zhong, S., Wang, X., Shen, L., Liu, L., and Liu, Y. (2019). Land use change in coastal cities during the rapid urbanization period from 1990 to 2016: a case study in Ningbo city, China. Sustainability 11, 2122. doi:10.3390/su11072122

CrossRef Full Text | Google Scholar

Zhang, T., Foliente, G., Xiao, J., and Tang, L. (2025). Exploring the driving factors of land use change and spatial distribution in coastal cities: a case study of Xiamen city. Sustainability 17, 941. doi:10.3390/su17030941

CrossRef Full Text | Google Scholar

Zhao, L., Ma, R., Chen, F., Li, J., Chen, Q., and Fan, Y. (2025). Enhancing the spatial resilience of agricultural heritage through constructing a balanced production-living-ecological space protection network. Herit. Sci. 13, 368. doi:10.1038/s40494-025-01920-7

CrossRef Full Text | Google Scholar

Zhou, H., Tang, M., Huang, J., Mei, X., and Zhao, H. (2024). Driving mechanisms and multi-scenario simulation of land use change based on national land survey data: a case in Jianghan plain, China. Front. Environ. Sci. 12, 1422987. doi:10.3389/fenvs.2024.1422987

CrossRef Full Text | Google Scholar