Introduction

In recent decades, the rapid pace of industrialization, urbanization, and population growth has intensified human-environment interactions, creating complex challenges for the sustainable management of natural resources1. As nations and regions continue to develop, the competition for land, energy, and water resources has escalated, often at the expense of ecological balance2,3. This has resulted in the degradation of ecosystems, loss of biodiversity, and increased pollution, which in turn impacts human health, agricultural productivity, and overall well-being4,5. Consequently, achieving sustainable development necessitates a harmonious balance between economic activities, human settlements, and ecological preservation—a central challenge for regional planning and governance worldwide.

Understanding the interplay between ‘production’ space (areas primarily supporting economic and industrial activities)6,7’living’ space (areas dedicated to human habitation, social functions, and well-being)8,9and ‘ecological’ space (areas providing essential ecosystem services and environmental regulation)10,11,12 is crucial for navigating this challenge13,14. For instance, the expansion of production space may encroach upon and degrade ecological areas, while urban growth (living space) can consume valuable agricultural land or green spaces, diminishing ecosystem services vital for human well-being15,16,17. Analyzing the ‘coupling’ (intensity of interaction) and ‘coordination’ (degree of harmony and synergy) between these Production-Living-Ecological (P-L-E) spaces offers a valuable framework for assessing the sustainability of regional development pathways and informing integrated land management strategies.

Despite the recognized importance of P-L-E interactions, much of the existing research has focused on analyzing these spaces in isolation or examining only pairwise relationships. For example, the expansion of industrial production can lead to environmental degradation18,19while urban growth can result in the loss of green spaces and biodiversity20,21. Furthermore, many studies lack sufficient spatiotemporal depth to capture the dynamic evolution of these interactions over time, particularly within regions experiencing significant socioeconomic transformation. A research gap thus exists in comprehensively assessing the dynamic, ternary coupling and coordination across P-L-E spaces, identifying how their spatial patterns and functional relationships evolve under pressures like rapid urbanization.

This study addresses this gap by developing and applying an integrated framework to analyze the spatiotemporal evolution of P-L-E coupling coordination in Ganzhou City, Jiangxi Province, China, from 2000 to 2020. Ganzhou City serves as a representative case study area due to its significant economic development, rapid urbanization, and rich ecological resources, presenting a microcosm of the development-conservation tensions faced by many regions22,23. Specifically, we employ a grid-based functional assessment methodology to quantify the P-L-E functions across the landscape. We then utilize a ternary coupling coordination degree model to evaluate the level of synergy between these three functions, moving beyond simpler bivariate analyses. Furthermore, spatial autocorrelation techniques (Global and Local Moran’s I) and trend analysis (Mann-Kendall test) are applied to reveal the spatial clustering patterns and temporal evolution of P-L-E coordination.

By integrating these approaches, this study aims to: (1) Quantify the spatiotemporal changes in production, living, and ecological functions in Ganzhou City from 2000 to 2020; (2) Evaluate the evolution of the coupling coordination degree among the P-L-E spaces; (3) Identify the spatial distribution characteristics and clustering patterns of P-L-E coordination; and (4) Analyze the temporal trends in coordination levels across the study area. The findings are intended to provide valuable insights for policymakers seeking to optimize land use allocation, promote balanced regional development, and enhance the overall sustainability of Ganzhou City and other similar rapidly developing regions. This research contributes a refined methodological approach for integrated P-L-E analysis and offers empirical evidence on the complex dynamics shaping human-environment systems during urbanization.

Materials and methods

Study area

The study area focuses on Ganzhou City, located in the southern part of Jiangxi Province, China. Ganzhou is an important regional center with a significant economic, ecological, and cultural influence in the region (Fig. 1). The city spans a diverse landscape, characterized by mountainous terrain, fertile plains, and a mix of urban and rural areas. It serves as a representative example of a rapidly developing region, where urbanization and industrialization are increasingly encroaching on ecological spaces. The city is divided into 18 districts and counties, each exhibiting distinct characteristics in terms of land use patterns, development stages, and ecological conditions. The study focuses on both urban and peripheral areas to capture the variation in land use and functional interactions across the region. This area is of particular interest due to the ongoing challenges related to balancing urban development, industrial growth, and ecological preservation. The region’s climate is classified as subtropical humid, with abundant rainfall and warm temperatures, which supports a rich diversity of vegetation. As such, Ganzhou offers a unique opportunity to investigate the spatial and functional dynamics of production, living, and ecological spaces, particularly as they evolve in response to urban expansion and environmental conservation efforts.

Fig. 1
figure 1

Study area. The land use types depicted in the figure are derived from the CNLUCC database for the year 2020. (Fig. 1 was generated by the manuscript author using ArcGIS 10.2 software. Software version number is 10.2, the link to https://desktop.arcgis.com/.)

Full size image

Research framework

This study divides Ganzhou City into 10 km grid cells, which are used as the basic evaluation units. Using land use data, the multi-factor weighted summation method is applied to calculate the functional scores for production, living, and ecological land use within each grid cell. Following this, bivariate spatial autocorrelation analysis is employed to assess the spatial relationships among the three functional areas—production, living, and ecological spaces. To further evaluate the interaction and coordination between these spaces, the ternary coupling coordination model is applied to compute the coupling coordination degree of the three functional types. Lastly, the Mann-Kendall trend test is utilized to analyze the temporal evolution of the coupling coordination degree, providing insights into the changes and trends in spatial coordination over time. The specific research framework and methodology are illustrated in Fig. 2.

Fig. 2
figure 2

Technical process circuit diagram.

Full size image

Data source and pre-processing

This study divides Ganzhou City into a grid with a cell size of 10 km, using each grid as an evaluation unit. The research focuses on five time points: 2000, 2005, 2010, 2015, and 2020, utilizing land use data from these years. The land use data are sourced from the Multi-period Land Use Remote Sensing Monitoring Database of China (CNLUCC), which is maintained by the Resource and Environmental Science Data Center of the Chinese Academy of Sciences (http://www.resdc.cn). This dataset employs a three-level classification system, where the first level categorizes land cover and its utilization attributes into six primary classes: cropland, forestland, grassland, water bodies, construction land, and unused land. The second level further subdivides these categories into 25 subclasses based on the natural attributes of the land resources. The spatial resolution of the dataset is 30 m, with an overall accuracy exceeding 85%.

Evaluation index system

To evaluate the spatial structure of P-L-E spaces in Ganzhou City, a systematic approach was adopted. A 10 km grid was established as the basic evaluation unit, resulting in a total of 3,892 evaluation units across the study area. Within each unit, the area of different land use types was calculated, and a multi-factor weighted sum method was applied to assess the functional scores of production land, living land, and ecological land. The classification of land use types and their corresponding functional scores is presented in Table 1. The functional intensity scores assigned to each land use type, presented in Table 1, were determined based on the established methodologies for P-L-E space functional assessment and classification within the relevant literature, drawing specifically on the framework proposed by Liu et al.24. The evaluation system categorizes land use into three primary functional types: production land, living land, and ecological land. Each category is further divided into three subcategories—strong, moderate, and weak—based on their functional intensity25. This approach, which links land use types to their primary and secondary P-L-E functions.

Table 1 Functional scoring system for production-living-ecological spaces.
Full size table

The functional score for each evaluation unit was computed using the weighted sum method, as expressed in the following formula (Eq. 1):

$$\:{W}_{j}=\sum_{i=1}^{n}{S}_{i}\times\:{V}_{i}$$
(1)

where \(\:{W}_{j}\) represents the total functional score of the \(\:j\)th evaluation unit for production, living, or ecological functions. The index \(\:i\) denotes the land use type, and n is the total number of land use types within the evaluation unit. \(\:{S}_{i}\) refers to the area of the \(\:i\)th land use type within the evaluation unit (unit: km2, and \(\:{V}_{i}\) represents the functional score assigned to each land use type.

To ensure comparability between different evaluation units, the functional scores were further normalized using min-max normalization. The normalization formula is as follows (Eq. 2):

$$\:{W}_{j}^{norm}=\frac{{W}_{j}-{W}_{min}}{{W}_{max}-{W}_{min}}$$
(2)

where \(\:{W}_{j}^{norm}\:\)is the normalized functional score of the \(\:j\)th evaluation unit. \(\:{W}_{max}\) and \(\:{W}_{min}\) represent the maximum and minimum functional scores across all evaluation units, respectively. This normalization process ensures that all functional scores are scaled to the range [0,1], facilitating comparison and spatial analysis.

Spatial autocorrelation analysis

To examine the spatial correlation between production, living, and ecological functions, a bivariate spatial autocorrelation analysis was conducted26. Specifically, the Moran’s I index and Local Indicators of Spatial Association (LISA) maps were used to evaluate the pairwise spatial relationships among production-living (P-L), production-ecological (P-E), and living-ecological (L-E) functions.

The bivariate Moran’s I index measures the spatial correlation between two variables across adjacent spatial units27. It is defined as follows (Eq. 3):

$$\:I=\frac{{\sum\:}_{i=1}^{n}{\sum\:}_{j=1}^{n}{w}_{ij}\left({X}_{\dot{i}}-\stackrel{-}{X}\right)\left({Y}_{\dot{j}}-\stackrel{-}{Y}\right)}{{S}_{X}{S}_{Y}\sum\:_{i=1}^{n}\sum\:_{j=1}^{n}{w}_{ij}}$$
(3)

where \(\:I\) is the bivariate Moran’s I index, indicating the degree of spatial correlation between two variables. \(\:{X}_{\dot{i}}\) and \(\:{Y}_{j}\) represent the functional scores of the two land functions (e.g., production and living) in spatial units \(\:i\) and \(\:j\), respectively. \(\:\stackrel{-}{X}\) and \(\:\stackrel{-}{Y}\) denote the mean values of the two functional scores across all spatial units. \(\:{w}_{ij}\) is the spatial weight between units \(\:i\) and \(\:j\), reflecting their spatial adjacency relationships. \(\:{S}_{X}\) and \(\:{S}_{Y}\) are the standard deviations of \(\:X\) and \(\:Y\), respectively.

A positive Moran’s I value indicates that high values of one function are spatially clustered with high values of another, while a negative Moran’s I suggests spatial dispersion or inverse correlation.

To further explore the spatial patterns at the local level, the LISA analysis was employed to identify significant spatial clusters and outliers28. The local bivariate Moran’s I is expressed as follows (Eq. 4):

$$\:{I}_{i}=\frac{({X}_{i}-\stackrel{-}{X})}{{S}_{X}}\sum_{j=1}^{n}{w}_{ij}\frac{({Y}_{j}-\stackrel{-}{Y})}{{S}_{Y}}$$
(4)

where ​\(\:{I}_{i}\) represents the local spatial correlation for unit \(\:i\).

This metric enables the identification of distinct spatial association patterns, including High-High (HH) clusters, where areas with high values of one function are surrounded by high values of another; Low-Low (LL) clusters, where areas with low values of one function are surrounded by low values of another; High-Low (HL) clusters, where high values of one function are surrounded by low values of another; and Low-High (LH) clusters, where low values of one function are surrounded by high values of another.

Coupling coordination degree model

To assess the spatial coupling and coordination of production, living, and ecological functions, the ternary coupling coordination degree model was applied29,30. This model evaluates the degree of interaction and synergy among the three functional spaces, revealing whether they develop harmoniously or exhibit imbalances.

The coupling degree (C) quantifies the interdependence and interaction intensity among production, living, and ecological functions. It is defined as follows (Eq. 5):

$$\:C=3\times\:{\left(\frac{P\cdot\:L\cdot\:E}{{\left(P+L+E\right)}^{3}}\right)}^{\frac{1}{3}}$$
(5)

where P, L and E represent the normalized functional scores of production, living, and ecological functions within each evaluation unit. The coupling degree ranges from 0 to 1, with higher values indicating stronger interactions among the three functions.

While the coupling degree reflects the intensity of interactions, it does not indicate whether these interactions are positive or negative. Thus, the coupling coordination degree (D) is introduced to evaluate the extent to which the three functions develop in a balanced and coordinated manner. The formula is expressed as (Eq. 6):

$$\:D=\sqrt{C\times\:T}$$
(6)

where T is the comprehensive development index of the three functions, calculated as (Eq. 7):

$$\:T=\alpha\:P+\beta\:L+\gamma\:E$$
(7)

where \(\:\alpha\:\), \(\:\beta\:\), and \(\:\gamma\:\) are weighting coefficients that reflect the relative importance of production, living, and ecological functions. These weights are typically assigned based on expert judgment or empirical analysis, ensuring that they sum to 1 (\(\:\alpha\:+\beta\:+\gamma\:=1,\:\alpha\:=\beta\:=\gamma\:=1/3\)).

In this study, we assigned equal weights to each function (\(\:\alpha\:=\beta\:=\gamma\:=1/3\)). This decision stems from the objective of providing a balanced and impartial assessment of the coordination state among the three fundamental spatial functions, treating each system—production, living, and ecology—as equally crucial pillars for sustainable regional development in this foundational spatiotemporal analysis. While we acknowledge that the relative importance of these functions might differ across various stages of regional development or under specific policy orientations, adopting equal weights establishes a standardized and objective baseline. This approach minimizes subjective bias in the absence of empirically derived local weights or clearly defined, historically varying policy priorities for the specific context of Ganzhou City over the 2000–2020 period. It allows for a clear evaluation of the inherent coordination dynamics and temporal trends across the study area before potentially exploring scenario-specific variations in future work.

The D also ranges from 0 to 1, with higher values indicating a more balanced and sustainable spatial interaction among production, living, and ecological functions. Based on D, the spatial coupling coordination level can be classified into different stages, such as Table 2.

Table 2 Classification of coupling coordination types based on D.
Full size table

Trend analysis using the Mann-Kendall test

To analyze the temporal evolution of the D and assess whether there is a statistically significant trend over time, the Mann-Kendall (MK) trend test was applied31,32. This non-parametric test is widely used for detecting monotonic trends in time series data, as it does not require the data to follow a specific distribution and is resistant to outliers.

The Mann-Kendall test statistic (S) is calculated as follows (Eqs. 8 and 9):

$$\:S=\sum_{i=1}^{n-1}\sum_{j=i+1}^{n}sgn({D}_{j}-{D}_{i})$$
(8)

where

$$\:sgn\left({D}_{j}-{D}_{i}\right)=\left\{\begin{array}{c}+1,\:if{D}_{j}-{D}_{i}>0\\\:0,\:if{D}_{j}-{D}_{i}=0\\\:-1,\:if{D}_{j}-{D}_{i}<0\end{array}\right.$$
(9)

here, \(\:{D}_{i}\) and \(\:{D}_{j}\) represent the D at time points\(\:\:i\:\)and \(\:j\:\)respectively, and \(\:n\) is the total number of time periods analyzed. A positive \(\:S\) value indicates an increasing trend, while a negative \(\:S\) value suggests a decreasing trend.

For large sample sizes \(\:S\) approximately follows a normal distribution with mean \(\:E\left(S\right)=0\), and the variance is given by (Eq. 10):

$$\:Var\left(S\right)=\frac{n(n-1)(2n+5)}{18}$$
(10)

The standardized test statistic\(\:\:Z\) is then computed as (Eq. 11):

$$\:Z=\left\{\begin{array}{c}\frac{S-1}{\sqrt{Var\left(S\right)}},\:if\:S>0\\\:0,\:if\:S=0\\\:\frac{S+1}{\sqrt{Var\left(S\right)}},\:if\:S<0\end{array}\right.$$
(11)

Result

Function score of production-living-ecological space

Between 2000 and 2020, the functional scores of ecological space in Ganzhou were consistently high, whereas living space exhibited the lowest scores, with production space falling in between (Fig. 3). This indicates that ecological functions remained dominant across the region, while living space development was relatively limited. Notably, Nankang and Zhanggong districts experienced a significant increase in the functional scores of both production and living spaces, suggesting intensified urbanization and industrialization. However, this expansion occurred at the expense of ecological space, leading to a decline in its functional score in these areas.

Fig. 3
figure 3

Spatiotemporal distribution of the functional scores of production-living-ecological spaces. (Fig. 3 was generated by the manuscript author using ArcGIS 10.2 software. Software version number is 10.2, the link to https://desktop.arcgis.com/.)

Full size image

The frequency distribution of functional scores further highlights the disparities among the three spatial functions (Fig. 4). The majority of production space scores were concentrated below 0.4, reflecting moderate industrial and agricultural activity across the region. Living space scores were predominantly below 0.2, emphasizing the relatively small proportion of land dedicated to residential and infrastructural functions. In contrast, ecological space scores were generally above 0.6, reaffirming the region’s strong ecological foundation and extensive green coverage. These results illustrate the spatial imbalance among production, living, and ecological spaces, underscoring the trade-offs involved in land use transformations over the study period.

Fig. 4
figure 4

Frequency distribution histogram of production-living-ecological space function scores. The bars in the figure represent the number of grid cells, and the curve represents the Kernel Density Estimation (KDE).

Full size image

Spatial autocorrelation of production-living-ecological space

To investigate the spatial interdependencies among the three functional spaces, bivariate spatial autocorrelation analysis was performed using both global Moran’s I and LISA.

Table 3 shows the Moran’s I values for the spatial autocorrelation of P-L, P-E, and L-E spaces from 2000 to 2020 (p = 0.01). The results indicate that all three functional relationships exhibit significant spatial clustering, though with varying trends over time. The P-L correlation remained stable (0.54–0.55), reflecting a persistent spatial association between industrial and residential development. The P-E correlation declined from 0.59 in 2000 to 0.52 in 2020, suggesting increasing spatial separation as production activities expanded into ecological areas. The L-E correlation was the weakest and declined the most, dropping from 0.18 to 0.12, indicating a growing spatial decoupling between urban expansion and ecological land. Overall, while production and living spaces maintained strong clustering, the spatial interaction between ecological space and both production and living functions weakened, highlighting increasing land-use conflicts between development and ecological conservation.

Table 3 Moran’s I (p = 0.01) of Spatial autocorrelation in production-living-ecological space.
Full size table

While global Moran’s I provides an overall measure of spatial association, LISA analysis identifies the specific locations of statistically significant spatial clusters and outliers, offering deeper insights into local patterns (Fig. 5).

Fig. 5
figure 5

The LISA cluster map of production-living-ecological space derived from local spatial autocorrelation, where NS denotes non-significance (p > 0.05). (Fig. 5 was generated by the manuscript author using ArcGIS 10.2 software. Software version number is 10.2, the link to https://desktop.arcgis.com/.)

Full size image

The LISA map for P-L functions is dominated by HH and LL clusters. A prominent HH cluster is consistently observed in the central urban area encompassing the Nankang and Zhanggong districts. This geographically reflects the concentration of Ganzhou’s primary industrial zones, commercial centers, and dense residential areas, where high levels of economic production and urban living functions naturally co-locate due to agglomeration effects, infrastructure availability, and employment opportunities. Conversely, extensive LL clusters are found primarily in the peripheral, often mountainous or predominantly agricultural, counties of Ganzhou (e.g., parts of Huichang, Anyuan, Xunwu). These areas typically have lower levels of industrialization and urbanization, resulting in low functional scores for both production and living spaces.

The spatial relationship between production and ecological functions reveals more complex patterns indicative of potential land use conflicts. Low Production-High Ecological clusters are prevalent in large portions of the peripheral counties (e.g., Chongyi, Shangyou, northern Anyuan), corresponding to areas with extensive forest cover, mountainous terrain, and designated ecological protection zones where production activities are limited. Conversely, High Production-Low Ecological clusters are found near the expanding urban and industrial centers (like Nankang-Zhanggong periphery) and potentially areas of intensive agriculture. These clusters highlight zones where production activities likely encroach upon or degrade ecological spaces. LL clusters are also present in some peripheral areas, possibly representing regions with limited economic development and less dominant ecological features compared to core conservation areas. The declining global Moran’s I for P-E is thus manifested locally through these competing spatial arrangements.

The L-E relationship is strongly characterized by spatial decoupling, particularly evident in the LISA map. Significant Low Living-High Ecological clusters dominate vast areas, especially in the western and southern mountainous regions (e.g., Chongyi, Dayu, Longnan). These represent areas rich in ecological assets (forests, water bodies) but with low population density and limited urban development. Conversely, High Living-Low Ecological clusters are concentrated in the main urban cores (Nankang, Zhanggong) and the county seats. This pattern clearly illustrates the classic urban development dynamic where the expansion of residential areas, infrastructure, and human settlements (high L) often occurs at the expense of natural or semi-natural ecological spaces (low E), leading to habitat fragmentation and loss of green space within and around developed zones. LL clusters are again found in peripheral rural areas with sparse population and less prominent ecological assets.

Spatiotemporal evolution of production-living-ecological Spatial coupling coordination

The spatiotemporal dynamics of the D for P-L-E spaces in Ganzhou from 2000 to 2020 were evaluated using the ternary coupling coordination model (Fig. 6).

Fig. 6
figure 6

Spatiotemporal evolution of the coupling coordination degree of production-living-ecological spaces. (Fig. 6 was generated by the manuscript author using ArcGIS 10.2 software. Software version number is 10.2, the link to https://desktop.arcgis.com/.)

Full size image

The results reveal distinct spatial patterns in coordination levels. A significant portion of Ganzhou consistently falls within the Slightly Coordinated category (0.4 ≤ D < 0.6). These areas, often transitional zones between urban centers and rural peripheries or regions with mixed agricultural and ecological landscapes, exhibit a basic level of interaction among the three functions, but lack strong synergistic development. The Severely Imbalanced category (0 ≤ D < 0.2) is predominantly concentrated in the extensive peripheral regions, particularly in the southern and western mountainous areas (e.g., parts of Chongyi, Dayu, Anyuan, Xunwu). This low coordination likely stems from the dominance of a single function (often high ecological value due to forests, but low production and living scores) and limited infrastructure or economic activity, hindering functional interaction and integration. These areas often correspond to ecologically sensitive zones or regions with challenging topography, limiting diversified development.

Conversely, the Moderately Coordinated category (0.6 ≤ D < 0.8) is primarily found in the northern part of Ganzhou, notably expanding over time in and around the central urban agglomeration of Nankang, Zhanggong, and Ganxian districts. This higher coordination reflects the influence of more intensive urbanization and industrialization, leading to a co-location and interaction of production facilities, residential areas, and associated infrastructure. The expansion of these moderately coordinated zones by 2020 suggests that development in these core areas, while potentially pressuring ecological spaces (as seen in the P-E and L-E autocorrelation), has also fostered a greater degree of functional interplay compared to the less developed peripheries, possibly reflecting municipal planning efforts aimed at balancing growth.

Figure 7 shows the frequency distribution of the D for P-L-E spaces over different periods. The distribution exhibits a bimodal pattern, with peaks near 0 and 0.5. This suggests that in earlier years, most areas had either low or moderate coordination between the three functional spaces. Over time, there has been a noticeable shift, with an increase in the frequency of D values greater than 0.4, indicating an overall improvement in the coordination level of P-L-E spaces in Ganzhou. This trend signifies a positive transformation towards a more integrated land use system, where ecological conservation is increasingly balanced with urban and industrial development. However, as of 2020, no areas have exceeded a D value of 0.8, which highlights the fact that, while progress has been made, the region has not yet reached a high level of coordination between these spatial functions on a broad scale.

Fig. 7
figure 7

Frequency distribution changes of coupling coordination degree in production-living-ecological spaces. The bars in the figure represent the number of grid cells, and the curve represents the Kernel Density Estimation (KDE).

Full size image

To assess the trends in spatial coupling coordination, we employed the Mann-Kendall trend test on the D values (Fig. 8). The results indicate that the vast majority of regions (84.3%) exhibit no significant trend. These stable areas are widespread, spanning both severely imbalanced peripheral regions, where coordination remained consistently low due to persistent geographical constraints or conservation mandates, and slightly coordinated zones, where the balance, though basic, showed minimal change. This pervasive stability implies that despite urbanization-driven land use shifts, the foundational coordination structure across much of Ganzhou persisted resiliently or became entrenched over the study period. Importantly, a notable portion of the study area (10.1%) demonstrates a statistically significant increasing trend in coupling coordination. These areas are primarily clustered around the expanding urban centers (Nankang, Zhanggong, Ganxian) and along major transportation corridors. This positive trend likely reflects successful integration efforts associated with planned urban expansion, infrastructure development, and potentially industrial upgrading, where economic and residential growth has occurred in ways that improved the functional balance with ecological considerations compared to previous states. Analyzing these improving areas could offer valuable lessons for promoting coordination elsewhere. In contrast, only 5.6% of regions show decreasing trends, reflecting areas where coordination between the functions has weakened over time. These regions are sparsely distributed, indicating that overall, the coupling coordination of production, living, and ecological spaces in Ganzhou has been improving.

Fig. 8
figure 8

Mann-Kendall trend test of coupling coordination degree (p < 0.1). (Fig. 8 was generated by the manuscript author using ArcGIS 10.2 software. Software version number is 10.2, the link to https://desktop.arcgis.com/.)

Full size image

Discussion

Spatial and Temporal patterns of coupling coordination

The results of this study reveal significant spatial and temporal patterns in the coupling and coordination of production, living, and ecological spaces in Ganzhou. The analysis of functional scores indicates that ecological spaces generally exhibit higher functional scores compared to living and production spaces, suggesting that the region retains a strong ecological foundation. However, However, this ecological dominance faces increasing pressure, evidenced by the growing spatial decoupling between developmental functions (production and living) and ecological functions33,34. The declining Moran’s I for P-E and L-E relationships, coupled with the prevalence of high production/living-low ecological clusters near urban centers and low production/living-high ecological clusters in peripheral conservation areas (Fig. 5), vividly illustrates this tension. This aligns with numerous studies in rapidly urbanizing regions of China and elsewhere, which report conflicts between expanding construction land and shrinking ecological spaces​35,36.

The analysis of D further highlighted significant spatial heterogeneity. While core urban areas like Nankang-Zhanggong exhibited moderate coordination (D ≈ 0.6–0.8), reflecting a degree of functional integration driven by development, vast peripheral areas remained in states of imbalance (D < 0.4), often characterized by strong ecological functions but weak production and living integration. This spatial disparity between urban cores and rural hinterlands is a common feature in regional development studies37,38.

A key contribution of this study is the application of the ternary coupling coordination model over a 20-year period, offering a more holistic view than bivariate analyses and revealing long-term dynamics. Interestingly, the Mann-Kendall trend test indicated that despite significant land use changes, the coordination level remained statistically stable across the majority (84.3%) of Ganzhou. This suggests a degree of path dependency or structural inertia in the P-L-E relationships across large parts of the landscape39,40. However, the emergence of statistically significant increasing coordination trends (10.1% of areas), primarily around the urban core, provides encouraging evidence that targeted development and potentially improved planning can foster better functional integration, even amidst urbanization pressures. The absence of highly coordinated areas (D > 0.8) anywhere in Ganzhou, however, underscores the persistent challenge of achieving optimal synergy.

Interpretation of findings and potential driving mechanisms

The spatiotemporal patterns observed in this study are the result of a complex interplay of multiple driving forces. Urbanization and industrialization emerge as primary influences, as evidenced by the aggressive expansion of urban areas with high living scores and the proliferation of industrial zones characterized by elevated production scores, particularly in the Nankang-Zhanggong-Ganxian nexus. This concentration has produced a strong coupling between production and living functions, observable through HH clusters, while simultaneously inducing a decoupling in the P-E and L-E relationships, manifesting as HL clusters. These dynamics mirror the principles of agglomeration economics and the well-documented trajectory of urban growth encroaching on peripheral natural landscapes41,42.

Compounding these socio-economic forces are the inherent topographical constraints and resource endowments of the region43. Ganzhou’s extensive mountainous terrain naturally restricts intensive agricultural and construction activities in outlying areas, thereby preserving high ecological scores and encouraging the emergence of LH clusters in both the production-ecological and living-ecological pairings. The resulting overall coordination index, which falls below 0.4, should be interpreted not as a failure of management but as a reflection of the intrinsic physical limitations that predispose these zones to reduced production and living functions.

Policy and planning considerations further nuance these spatial dynamics. Although not explicitly modeled in the current analysis, the presence of ecological protection zones and forest reserves likely contributes to the stability of high ecological function scores in designated areas. In contrast, the moderate coordination and gradual improvements observed around urban centers may indicate the early impacts of municipal planning strategies that aim to concentrate development while integrating environmental concerns. Pervasive stability in coordination across the region could also point to a historical shortfall in integrated spatial planning that fully harmonizes production, living, and ecological functions44.

In addition, broader socio-economic dynamics—including population migration towards urban centers, shifts in economic structure from primary to secondary or tertiary sectors, and the enhancement of transportation infrastructure—appear to underpin the intensification of production and living functions in core areas. This multifaceted evolution suggests that the relative stability in overall coordination might either indicate a resilient existing spatial structure or reflect a lag wherein the full impacts of recent developmental changes have not yet been realized45,46. Notably, the emerging localized trends toward improved coordination signal that more positive transitions are feasible if integrative planning approaches can be effectively implemented.

Policy implications and recommendations

The findings underscore the need for more integrated and spatially explicit land use planning in Ganzhou to navigate the trade-offs between development and conservation effectively. Policy recommendations should be tailored to the distinct coordination characteristics of different areas:

  1. (1)

    For Core Urban Areas (Moderately Coordinated / Increasing Trend): Focus should be on optimizing coordination. This involves embedding green infrastructure (parks, greenways) within the urban fabric, promoting compact urban development to reduce sprawl, upgrading industrial activities towards cleaner production, and strengthening pollution control to mitigate the negative impacts of high P and L density on residual ecological functions47,48. Enhancing connectivity between urban green spaces and peripheral ecological areas is crucial.

  2. (2)

    For Peripheral Areas (Severely Imbalanced / Stable Low Trend): Strategies must be context-specific. In areas dominated by high ecological value, policies should prioritize strict ecological protection, potentially combined with eco-compensation mechanisms or the development of compatible livelihoods like sustainable forestry or eco-tourism49. In peripheral areas with low scores across all functions, interventions might focus on targeted poverty alleviation, sustainable agriculture, or basic infrastructure improvements, carefully planned to avoid future P-L-E conflicts. A one-size-fits-all development approach is unsuitable.

  3. (3)

    For Transitional Zones (Slightly Coordinated / Stable Mid Trend): These areas require proactive planning to guide future development towards higher coordination levels. Key strategies include establishing clear urban growth boundaries, creating ecological corridors to maintain connectivity between core ecological areas50promoting mixed land use that integrates living and working spaces with accessible green areas, and implementing sustainable agricultural practices51,52.

Limitations and prospects

While this study provides valuable insights into the coupling and coordination of land use functions, there are several avenues for future research. One limitation of this study is its reliance on land use data from only five time points (2000, 2005, 2010, 2015, and 2020), which provides a limited temporal scope. Future research could benefit from incorporating more frequent time points, enabling a finer-grained analysis of temporal changes and trends in land use coordination. Second, the study relies on a fixed land use classification, functional scoring system, and equal weighting in the coupling coordination model. These parameters influence the results. Future research would benefit significantly from sensitivity analyses exploring how varying the scoring criteria or weighting coefficients (reflecting different policy priorities or development stages) impacts the assessed coordination levels, thereby testing the robustness of the findings. Third, while we discussed potential drivers, the study did not explicitly incorporate socio-economic variables (e.g., GDP, population density, investment) or policy variables directly into the quantitative models. Future research integrating these factors using methods like geographically weighted regression or spatial regression models could provide deeper quantitative insights into the specific mechanisms driving P-L-E coordination changes. Finally, extending this methodological framework to other resource-based cities or different geographical contexts would enhance the generalizability of the findings and allow for comparative analysis of P-L-E coordination dynamics under varying conditions. Investigating the specific impacts of major policy interventions (e.g., implementation of ecological civilization policies) on coordination trends would also be a fruitful direction.

Conclusions

This study examined the coupling and coordination of production, living, and ecological spaces in Ganzhou City, China, over a 20-year period (2000–2020), using spatial analysis and the ternary coupling coordination model. The findings reveal several important insights into the spatial and temporal dynamics of land use coordination in a rapidly urbanizing region. Our findings reveal the persistent functional importance of ecological spaces, yet highlight increasing pressure from urban and industrial growth leading to spatial decoupling, particularly in peripheral regions where coordination remains low. While overall coordination levels are moderate and heterogeneous, a striking finding was the widespread temporal stability in coordination across most areas, contrasting with localized improvements near urban centers, suggesting complex dynamics beneath the surface of land use change.

The study demonstrates the value of the integrated ternary coupling coordination framework for holistically assessing P-L-E interactions over extended periods, offering insights beyond simpler bivariate or static analyses. The methodological approach employed—combining grid-based functional assessment, the ternary coupling coordination model, and trend analysis—provides a robust and adaptable framework that holds potential for generalizability. It can be applied to other rapidly urbanizing regions or resource-based cities facing similar challenges in balancing development and environmental protection, allowing for comparative assessments.

However, achieving highly coordinated and sustainable development remains a challenge in Ganzhou. Future research should aim to enrich this framework by explicitly incorporating socio-economic drivers (e.g., GDP, population dynamics, investment flows) and specific policy variables into quantitative analyses, potentially using spatial regression models to better understand the mechanisms influencing coordination patterns. Furthermore, conducting sensitivity analyses on the functional scoring and weighting parameters would enhance model robustness. Investigating these dynamics with higher temporal resolution data and comparing findings across diverse regional contexts would also yield valuable insights. Ultimately, this research underscores the critical need for adaptive, spatially explicit planning strategies to foster synergistic P-L-E relationships and advance regional sustainability goals.