Research on Sustainable Land Use in Alpine Meadow Region Based on Coupled Coordination Degree Model—From Production–Living–Ecology Perspective

: Changes in land use types in alpine meadow areas have significant impacts on the ecological environment in alpine areas. Exploring land use change is crucial for land use management and optimization in alpine regions. Thus, it is necessary to analyze land use evolution and its drivers in alpine meadow regions from a production–living–ecology space (PLES) perspective by using remote sensing data. We first constructed the PLES evaluation system for Gannan. Then, we analyzed the spatial and temporal evolution characteristics and coupling degree of PLES in the study area. Finally, the driving factors affecting PLES were explored with geodetector. The conclusions of the study reveal that the distribution of productive and ecological spaces is large and concentrated, while the distribution of living spaces is more decentralized. The PLES was mainly concentrated in the area above 2500 m but below 4000 m and with a slope of 40 ◦ or less. During the study period, the area of production space showed a decreasing trend, while the areas of living and ecological space both showed increasing trends, primarily occurring at the expense of production space. DEM and GDP were the main factors affecting the distribution of PLES. The coupling level and the degree of coupling coordination were relatively stable in general, showing a pattern of “high in the east and low in the west”. The study provides technical support and a theoretical basis for the future planning of land space and ecological environment optimization in the alpine meadow regions


Introduction
Land is closely related to human production, life, ecology and other human activities [1].The spatial patterns and functions of land evolve because of the interaction of human behavior with the land [2].The accelerated industrialization and urbanization of the 20th and 21st centuries have dramatically altered land use patterns, resulting in an increased demand for production-living-ecology spaces (PLES) [3].Moreover, the imbalance in the utilization of space for PLES has caused environmental pollution, degradation of ecosystem functions, incomplete living space facilities and lack of vitality [4].Therefore, constructing a national land space utilization pattern and forming a coordinated PLES is an effective way to mitigate many of the problems caused by land use.
At present, in the context of research, scholars have carried out PLES research from different perspectives.For instance, some studies [5][6][7] related to "PLES" address functional classification, spatial identification, spatial optimization and so on.In terms of research scale, PLES mainly focuses on the provincial scale [8], city cluster scale [9,10], county scale [7] and watershed scale [11,12].Study regions have mainly involved mining and grain composite areas [13], the Poyang Lake area [14] and the Yangtze River Delta green integrated development demonstration area [15,16].The above studies have enriched the research content of PLES and provided theoretical and case support for subsequent studies.However, there are fewer studies on land use resources from a PLES perspective for alpine meadow regions.With the deepening of PLES research, some scholars have also carried out PLES driver analysis using principal component analysis [17], linear regression analysis [18], partial least squares regression models [19] and geographic detectors [20].However, the drivers of land use change are diverse and comprehensive.Geodetectors are a spatial analysis method that identifies spatial heterogeneity and its causes [21].It has been widely used to perform driver analysis and factor analysis.Some scholars have used geodetectors for PLES driver analysis, indicating the applicability of geodetectors in PLES.Therefore, we will utilize geodetectors to carry out PLES drive analysis in alpine regions.
Alpine meadows play an important role in regulating climate, conserving water and maintaining ecological balance, but are facing the threat of degradation.Exploring the spatial pattern of land use is crucial for implementing spatial land management and optimization in alpine areas.Changes in land use types are important for the protection of sensitive and fragile plateau ecosystems.The Gannan Tibetan Autonomous Prefecture (Gannan) is situated at the eastern fringe of the Qinghai-Tibetan Plateau, a typical alpine region.Gannan is a water source containment area and recharge area of the Yellow River and Yangtze River, which is China's ecological main function area and ecological civilization advance demonstration area.It is also rated as "China's Tourism Destination with the Most Ethnic Characteristics" by the United Nations Habitat Development Council and the World Chinese Federation.As the urbanization process continues to accelerate, it has led to expanding cities and increasing population.Eventually, there is a growing tension between the supply and demand for production space (PS), living space (LS) and ecological space (ES).Moreover, most of the present surveys on Gannan emphasize the analysis of grassland productivity and driving mechanisms [22,23].There are fewer studies on land use resources and their drivers using remotely sensed data from the PLES perspective.Thus, we propose to use geodetectors to dissect the migration pattern and drivers of PLES in Gannan.
Specifically, we aim to study the evolutionary trend of spatio-temporal dynamics of the Gannan PLES based on land use type data.On this basis, geodetectors were used to examine the drivers that influence spatial and temporal changes and the degree of coupling coordination of PLES in Gannan.This study can build a rationale and case support for the spatial distribution of high-quality development of land in alpine areas.

Study Areas
Gannan is located in western China, in the southwestern part of Gansu Province, and belongs to the transitional zone between the Qinghai-Tibetan Plateau and the Loess Plateau (Figure 1).The region consists of one city and seven counties, with a gross territory of 45,000 km 2 .It is also a region of water conservation and recharge of the Yellow River and Yangtze River.Gannan is also a nationally determined ecological main function area, and an ecological civilization advance demonstration area.At the end of 2022, the permanent population was approximately 683,700.Gannan was named "China's Tourism Destination with the Most Ethnic Characteristics" by the United Nations Habitat Development Council.The regional GDP in 2022 was 24.51 billion yuan, an increment of 4.0% in relation to the previous year.

Data Sources and Pre-Processing
The data used in this research include land use data, digital elevation model (DEM) data and socio-economic data.The land use datasets were obtained from GlobeLand30 (https://data.casearth.cn/thematic/glc_fcs30(accessed on 22 July 2023)) with a resolution of 30 m × 30 m.The land use types are divided into ten categories: cropland, forest, grassland, shrubland, wetland, water bodies, tundra, construction land, unused land, glacier and snow and ice.The DEM data were derived from the Geospatial Data Cloud (https://www.gscloud.cn/(accessed on 20 July 2023)), with a resolution of 90 m.The land use data and the DEM data were masked by ArcGIS10.8software using the vector extent data to obtain the extent of the study area.The altitude and slope were extracted based on DEM data by ArcGIS10.8software.According to the average altitude of Gannan and the customary division of slopes, the vertical difference is divided into segments of 500 m.The altitude classification is listed in Table 1.The slope difference is every 10 degrees.Socio-economic data such as GDP, population density and total number of people at the year-end are from the Gansu Statistical Yearbook and Gannan Statistical Yearbook.Specific data sources are listed in Table 2 [24].

Methodology
To explore the spatio-temporal dynamic characteristics of PLES and its driving force analysis in Gannan, we constructed the evaluation system of PLES based on land use data, and the specific process is shown in Figure 2.

Classification and Evaluation System for PLES in Gannan
The diversified characteristics of land use led to differences in its functions.In this study, with reference to the available literature and situation on the ground in the region under study, the land use categories were classified into production land, living land and ecological land.The classification and evaluation system of PLES is the basis for constructing a reasonable spatial configuration of PLES [25].On this basis, following the principles of systematicity, representativeness and scientificity, the evaluation scheme of PLES is based on existing research [7,26,27] and the current land use status of the region was comprehensively studied.Each type of land was split into four classes according to its functional strength and integrity and was given a score of 5, 3, 1 or 0, respectively.These classes are based on variations in the primary and secondary functions of land production, living and ecology [28].This led to the construction of the Gannon PLES scale (Table 3).

Coupling Coordination Degree (CCD) Model
There are interconnections, mutual influences and constraints among production functions, living functions and ecological functions of land [29].The coupling is a quantitative measure of coordination between system components [30,31].Based on the literature [32], we built a coupling degree model that included PS, ES and LS: where C is the coupling degree between PLES functions, and it takes a value between 0 and 1. P i , R i and E i represent the integrated evaluation value of PS, LS and ES, respectively.The degree of coupling can be used to interpret the extent of engagement between PS, LS and ES.However, it cannot reveal the coordination features of the three spatial types with respect to each other.Consequently, according to the coupling degree, the CCD model was constructed to reveal the coordination characteristics among the three space types.The CCD model is as follows: where CC d is the coupling coordination degree of PLES, and α, β and γ are the coefficients to be determined for the production function, living function and ecological function, respectively.Based on the opinions of relevant experts, the pending coefficients are determined as α = 0.

Geodetector
Geodetector is a mathematical method for detecting spatial dissimilarities and explaining the driving factors underlying it.It includes four types of detectors: factor detector, interaction detector, risk zone detector and ecological detector [21].In this research, we primarily capitalized on the factor detector and interaction detector in the geodetector to unveil the main impact factors of PLES evolution in Gannan.The details of the geodetector have been described in the previous literature.
The influencing factor of PLES needs to consider the development status of the study area, the relative stability of the indicators and the accessibility of the data.Based on the existing literature [36,37], combined with the PLES theory [38], we selected 8 different indicators to reveal the influencers of PLES changes.The specific indicators are listed in Table 5. Figure 3 displays that the differences in the structure of the PLES in Gannan are more obvious.It can be noted that from 2000 to 2020, the area of productive space decreased from 26,512.62 km 2 to 22,188.48km 2 .This represents a total reduction of 4324.14 km 2 .The area of LS changed from 42.83 km 2 to 170.55 km 2 , with a total increase of 127.71 km 2 .In contrast, from 2000 to 2020, there was a general upward trend in ES, with an increase of 4610.68 km 2 .To summarize, for the years 2000 to 2020, PS was dominant in Gannan, but its size declined.LS and ES exhibited upward trends.The main characteristics of PS and ES are large area distributions and concentrations, and that of LS is piecemeal distribution.The PS is concentrated in the west and north.The LS is mainly clustered in the urban districts of various counties and districts, signaling a continuous expansion trend.The ES is mainly concentrated in the southeast, and it has gradually expanded towards the north from 2000 to 2020.

Characterization of the Vertical Distribution of PLES
According to the average altitude of Gannan and the customary division of slopes, the vertical difference is divided into segments of 500 m.The slope difference is every 10 degrees.Figure 4 shows that there are significant differences in the vertical gradient of the PLES in Gannan.The peak value of the PS area is at 3500~4000 m above sea level.The peak of value of the LS area is at 3000~3500 m.The peak value of the ES area occurs at 3000~3500 m above sea level.In general, PLES types were more diverse at altitudes of 2500~4000 m during the study period (Figure 4a).In the period 2000-2020, PLES types were more diverse in areas with slopes below 40 • .The percentage of ES is the largest in the spatial distribution of areas with slopes above 40 • (Figure 4b).

Characterization of Changes for the PLES 3.2.1. Characterization of the Distribution of Area Changes in PLES
As illustrated in Figure 5, the areas of LS and ES that transformed into other spaces are relatively minor.By comparing the area of transformation of each space, it can be noticed that the area of PS transformed into other space is the largest, amounting to 5370 km 2 .The area transferred out of the other spaces is larger than the area transferred in, with 97.52% converted into ES and 2.5% converted into LS.The expansion of LS and ES leads to the retreat of PS.Secondly, the transition between ES and PS is more frequent than in other countries.This was manifested in two phases: a significant increase from 2000 to 2010 and a gradual transition from 2010 to 2020.

Spatial Changes in PLES from 2000 to 2020
Figure 6 illustrates the main manifestations of spatial changes in the PLES in Gannan from 2000 to 2020.The results show that productive land in Gannan was mainly converted to ecological land between 2000 and 2010.The total area converted was 3834.72 km 2 , mainly distributed in Lintan, Zhuoni and Hezuo in the northeast.Of the ecological land that has been converted into other uses, most has become productive land.The area converted is 341.90 km 2 , mainly in western Zhuoni and eastern Luqu.The outcome suggests that from 2010 to 2020, the productive land in Gannan was mainly converted into ecological land with an area of 1402.12 km 2 .The conversion of other land types is small and insignificant.6).Natural geographical factors are the fundamental conditions that influence the emergence and transformation of space structures in Gannan.From the factor detection results, it can be seen that the order of explanatory power for PLES in descending order is: X3 > X1 > X2 > X6 > X5 > X7 > X4 > X8.This indicates that DEM is the main natural factor influencing the change in PLES in the region.8 demonstrates that the interaction between two factors is more powerful than a single factor in interpreting spatial correlations.These two-factor interactions are nonlinearly and interactively enhanced, indicating a combined effect of several factors.The outcomes of the analysis of the interacting factors suggested that the interaction of the slope factor with the other factors was significantly higher compared to the single factor effect.

Analysis of the Distribution of the PLES
The PLES is mainly concentrated in the areas above 2500 m, below 4000 m and below 40 • .This indicates that PS and ES are mainly characterized by distribution over a large area and concentration, while LS is distributed sporadically.One possible reason is the significant effect of altitude on the vertical distribution of the PLES.The altitude of Gannan ranges from 1100 to 4900 m above sea level, and the overall altitude is on the high side.At lower altitudes, the terrain is more open and suitable for large-scale farming activities.The range of human activities is larger, and the area occupied by PS and LS is also larger.Within a given range, the hillier the elevation, the less the population distribution and the smaller the scope of human activities.The smaller the damage to the ecological environment, the larger the area of ES.When the altitude exceeds 4000 m, the climate conditions are more complicated, and the ecological environment is extremely harsh.So, the PLES is mainly concentrated in the altitude between 2500 and 4000 m.There have been studies that have proved that, within a certain range, climate and environmental conditions are relatively favorable at lower altitudes.Such areas are favorable to carrying out human production activities and the self-recovery of spatial environmental quality [39].
The second possible reason is the influence of slope.The main reason why most of the PLES is located in slopes below 40 • is that the slope affects the gathering of soil fertility [40].The steeper the slope, the more pronounced the erosion of the soil by flowing water, and high slopes cannot gather fertile soil.The lack of soil fertility will lead to a reduction in crop yield, i.e., low crop production efficiency and a reduction in agricultural production area [41].At the same time, the greater the average regional slope, the more complex the local climate change.In turn, this affects crop growth, and production efficiency is not obvious in high slope areas [42,43].That is why the PS is mostly distributed in slopes below 40 • .Transportation conditions are more complex in areas with steep slopes than in areas with gentle slopes.Population distribution is to a large extent restricted, with smaller populations living on areas with steep slopes [44].So, the proportion of LS on slopes above 40 • is very small.Places with large population distributions need more LS and PS, and may even occupy a certain area of ES.Therefore, the proportion of ES in areas with gentler slopes is smaller than that of LS and PS.Through the subsequent analysis of driving factors, it is found that, in 2020, DEM and slope are the factors that significantly influence the distribution of the PLES.This further confirms that the vertical distribution of PLES is mainly related to elevation and slope.The conclusion from Shi's research on related issues also confirms this conclusion [45].
As far as the horizontal distribution is concerned, one possible reason is related to the topography.The northwestern part of Gannan is a vast meadow grassland, which is the main pastoral area of the province and ideal for mass production activities.Therefore, PS is mainly located in the western and northern parts of the study area.The southern part is the Mindie mountainous area, with large mountains and deep ravines and a relatively mild climate.It is one of the important forest areas in the province.The eastern part is hilly and mountainous, with agriculture, forestry and animal husbandry, which is conducive to the expansion of ES.So the ES is mainly dispersed in the southeast of the study area.Influenced by the topographic conditions, the distribution type of towns and cities is mostly group or strip type, which is more scattered and narrower.The development of LS cannot be separated from PS and ES.Therefore, the LS is scattered in the PS and ES.This distribution situation also confirms the influence of topographic conditions on the distribution of the PLES.
We also found that for the duration of the survey, the PLES in the study area was differentiated.This is manifested in a decreasing trend in the area of PS.While the space of LS and ES showed an increasing trend, the increased area mainly came from the PS.One possible reason is related to local policies.The 11th Party Congress of the Prefecture put forward the strategic idea of "ecological statehood" and the construction of "ecological Gannan".The Twelfth Party Congress put forward the overall deployment of taking the lead in creating the "Qinghai-Tibet Plateau Green Modernization Advance Demonstration Zone, National Ecological Civilization Advance Demonstration Zone, National Regional Tourism Demonstration Zone, National Urban and Rural Environment Comprehensive Improvement Demonstration Zone", and focusing on fostering green industry, the development of eco-economy and construction of an ecological civilization.Guided by policy, corporate governance more directly and accurately understands the government guidance.Instead of blindly pursuing economic production benefits, it is conducive to further improving ecological benefits and even transforming in advance.In the meantime, to enhance the ecological environment in the western region, a strong ecological security barrier has been constructed.Comprehensive ecological management and precise pollution control are carried out, and ecological protection and restoration projects are continuously promoted.Therefore, the area of ES is on an increasing trend.
The second possible cause is related to the local industrial structure.Overall, the industrial structure of Gannan reveals a trend of decreasing proportion of agriculture year by year.The proportion of industry is stable and has a slow upward trend, the service industry is expanding and the structure tends to be rationalized.In the Gannan service industry, tourism accounts for a relatively large proportion.Since 2005, the tourism industry in Gannan has entered a sustainable development period.The promotion of tourism cannot be separated from the ecological environment.The unique advantages of the pristine natural landscape and humanistic landscape in Gannan promote the development of ecotourism in Gannan.This also accounts for the further strengthening of the protection of ecosystems in Gannan.The slogan of "green water and green mountains are golden silver mountains" has been realized.As a result, ES is growing in size.Moreover, the results of Zhou Guangliang et al.'s study [46] on the spatio-temporal evolution characteristics of PLES and its drivers in the Yellow River Basin are more in line with the findings of this study.In addition, the changes in industrial structure influence land use patterns which ultimately have an impact on ecological land use [47].These findings further support the results of this study.
Another possible reason is the increase in population size.As of 2020, the population of Gannan is 691,808, which is an increase of nearly 50,000 persons from the 2000s.The growth in population size contradicts the limited LS.The higher the population density, the faster the rate of human growth, and the more demand there will be for LS [48].Furthermore, the high rate of urbanization and urban growth has resulted in high demand for urban land [49,50].These in turn compress the stock of ES and affect the quality of the ecosystem.As a result, the protection of ES is a top priority due to the influence of policies.Therefore, the purpose of expanding the LS is realized by compressing part of the PS.Therefore, the area of LS tends to increase and the area of PS decreases.

Analysis of the Evolution of Spatial and Temporal Patterns of CCD
The level of Gannan CCD was relatively stable overall for the duration of the research.It demonstrates a wave-like evolution characteristic from low-level coupling to high-level coupling.One reason for this may be the different stages of development, with different major development conflicts and development goals in each county and city in Gannan.The spatial development of the PLES has gradually diverged, and the spatial development relationship has gradually changed.Some counties and cities generate development conflicts among the three major systems of nature, economy and society.Mapping in the carrier space is manifested in the instability of the PLES development; it is difficult to enter a long cycle of sustainable development.Therefore, its CCD is characterized by wave-like evolution.
Another reason may be the changes in the ideology and social cognition of the residents with the development of the society.At the beginning, at a poor level of financial development, the people's survival leads to the serious phenomenon of clearing land on steep slopes and indiscriminate cutting and logging.The stress on cropland is greater, and the distribution of cropland expands to upper elevations and high-gradient regions [51].The phenomenon of "cultivation in and forest out" emerges [52].The PS and ES are distributed at higher altitudes and higher slopes.The ES is compressed by the production space, and the spatial distribution of the PLES forms is not coordinated.Therefore, the coupling coordination level at this time is low.In the later stage, the income of the farmers increases as they go to the cities to work.They no longer rely on the steep slopes to cultivate the land to maintain their life.The pressure on cultivated land decreases, and the phenomenon of "forest in and cultivation out" appears.The change in human-terrestrial relations prompts the extension of ES to the low-altitude and low-slope areas.The proportion of ES increases, and PS and LS are further optimized.Therefore, the coordination level of the PLES space coupling increases, and gradually evolves from low-level coupling to high-level coupling.
Spatially, the high CCD areas in Gannan are mainly dispersed in the northeast, southwest and southeast parts of Gannan.There is a pattern of "high in the east and low in the west".It gradually spread to the neighboring districts and counties during the study period.One possible reason is that the eight counties and cities in Gannan have various stages of economic exploitation and urbanization.It makes the high and low values of coupling harmonization different in different areas.The eastern part of Gannan is more economically developed than the western part and has a higher level of urbanization.As of 2020, by comparing the urbanization levels of counties and cities, Hezuo in the east has the highest urbanization level, reaching 65.89%.Zhouqu in the west has the lowest urbanization level, at 29.78%.The more economically developed and the higher the level of urbanization, the higher the degree of completeness of public facilities.This can provide a higher level of life security and life service functions, and the degree of coupling coordination is higher [53].In economically backward regions, the degree of development is lower, and the degree of coupling coordination is lower.The regions of high value are mainly located in the northeast, southwest and southeast part of Gannan.This possible reason can also be confirmed by the conclusions drawn from the research on related issues by Li Na et al. [54].In the relatively economically backward regions, the disposable income per capita of rural residents plays a limiting role in the coordination of the PLES functional coupling.This may be due to the fact that the increase in rural residents' income has accelerated the construction of agricultural sites [55].The lack of planning for the use of rural residential land has led to a lag in the construction of roads and greening in villages.The present situation is dirty and messy, which leads to the reduction in the coordination of the "PLES functions" [56].
The second possible reason is that the transportation network is not synchronized with the construction of infrastructure.Transportation networks occupy ecological space by driving the expansion of construction land.High-grade roads have a strong radiationdriven effect on the economy and society [57], promoting the shift of land utilization in the surrounding areas to construction land use and occupying ES.Low-grade roads are directly related to regional economic and construction activities.Compared with high-grade roads, they more directly affect the surrounding land use, and their role in promoting the construction of land is also more obvious.Therefore, the construction of the transportation network will lead to mutual encroachment of various spaces.There are many environmentally sensitive sites such as nature reserves, water source protection zones, forest parks, geological parks, etc. in the whole state.A number of key projects planned to be implemented and under construction, such as the Lanhe Railway and Xicheng Railway, involve nature reserves.The application for adjusting the avoidance cycle is long, or even impossible to avoid.Some projects have not been able to start on schedule, and they are difficult to implement on the ground.This has resulted in some of the planned land being left idle for a long time.This further affects the level of coordinated development between production-living-ecology spaces, making the degree of coupled coordination within Gannan unbalanced.

Driver Analysis for the PLES
Taking 2020 into account, for instance, this research found that DEM and GDP were the most important elements affecting space allocation of PLES among natural and socioeconomic factors, respectively.One possible reason is that, among natural factors, the complexity of the terrain affects the distribution of the PLES.The more complex the terrain, the more difficult it is for human beings to determine the distribution of the various geographical spaces.Therefore, it has a hindering effect on human beings ability to carry out productive and living activities [58].The more undulating the terrain is, the more mountainous/hilly and gully areas there are, and complex landscape adds to the difficulty of ecological protection [59].The change in land type in the event of external disturbance leads to the degradation of the ecological environment.The vegetation cover and the height of vegetation growth are relatively low.Especially in the hills or mountain tops, most of the distribution is sparse grassland or bare land, the ecological environment quality is low and the ecological benefit is low.So, the complexity of the terrain affects the scalability of ES.This reason is further supported by Ke Liu et al. on the spatial conflict evolution and differentiation mechanism of the "PLES" in river valley cities [60].
A second possible reason is the different growth requirements of crops and vegetation types in different regions [61].This can affect the proportion of space occupied by PLES.Gannan has an alpine climate and is influenced by the monsoon.Most areas have long winters without summer, spring and fall, and a frost-free period of 85 to 180 days.Light is abundant, with annual sunshine hours of 1800 to 2600.Heat is insufficient, with a mean annual temperature of 1.1 to 12.7 • C and large vertical differences.Annual precipitation ranges from 400 to 800 mm, with significant differences in geographic distribution, and annual evaporation ranges from 1137 to 1973 mm.The growing conditions of crops and vegetation types are different in different areas.If the topographic climate conditions are poor, the region is subject to a greater influence of precipitation all year round.Plant growth is more sensitive to moisture, which is one of the reasons why there is less ES in the zone.So, the different growth requirements of crops and vegetation affect the spatial type of the site.This in turn affects the distribution of the PLES.
The impact of slope on the ecological environment is mainly realized through influencing climate and human activities.On the one hand, the greater the regional average slope, the more complex the local climate change.It tends to form more precipitation, which in turn affects the growth of vegetation.Therefore, the slope affects the climate of the ecological environment.In addition, in terms of human development history, the plains with gentle slopes are more utilized by human development to form production and living land, and the ecological effect is relatively low.On the other hand, mountainous and hilly areas with larger slopes are less disturbed by anthropogenic maneuvers, reducing the severity of environmental damage and impact.
Among the socio-economic factors, the total number of people at the year-end, population density and GDP have the greatest influence on the spatial allocation of the PLES.The possible reason for this is that as the year goes on, the number of people increases and the population density becomes higher.The development needs of human beings become higher and higher.However, land space is limited, and the demand for PS and LS can only be satisfied by compressing the ES.This causes changes in the distribution of the PLES spatial pattern.So, the total number of people at the year-end and population density play an essential role in impacting the spread of PLES.The growth of GDP has led to the rapid development of economy and urbanization in Gannan.It not only causes the continuous expansion of PS and LS in Gannan, but also leads to the enhancement of human disturbance to the ecological environment.The landscape pattern tends to be fragmented and complicated and causes changes in the vertical band spectrum of the PLES.Therefore, GDP is also one of the socio-economic factors affecting the spatial distribution of the PLES.

Limitations and Future Prospects
This study quantitatively analyzes the spatiotemporal evolution of the PLES in Gannan from multiple dimensions.The distribution of space-time modes of PLES and the motivating factors of its changes are revealed.This can provide meaningful empirical reference for ecological conservation and high-quality management of Gannan to a certain extent.This study could guide subsequent regional spatial planning in Gannan, including regional functional zoning and the identification of key general functional areas.For example, ecological function protection zones can strengthen restrictions on human activities, minimize ecological damage and promote the restoration and enhancement of ecological functions.In regional spatial planning, they can be controlled as "ecological function protection zones" and "ecological control zones".Large areas identified as basic ecological function zones may be designated as "basic ecological function zones".In brief, compared with previous studies, we have combined the actual situation of the study area with the applied functions of land types to comprehensively and objectively reveal the current land space status of the study area.The findings of this study provide case support for the policy formulation of optimizing the land space pattern and integrating land space resources.
However, there are still some shortcomings in this study.The spatial distribution of the PLES is influenced by the interaction of multiple factors, and there are certain limitations in evaluating the unilateral impact of multiple factors.In the future, the impact of the interaction of several elements should be thoroughly regarded.The essence of the PLES is the result of the continuous evolution of the human-land relationship regional system, influenced by many factors such as nature, society and culture.In the future, more in-depth research is needed to combine policy factors and other driving factors.This will provide more reference suggestions for the optimization of the PLES and regional sustainable development in Gannan.

Conclusions
Taking the Gannan Tibetan Autonomous Prefecture as the study area, this study analyzes the evolution characteristics, driving mechanisms and coupling coordination degree of the production-living-ecology spatial pattern from 2000 to 2020.The following main conclusions are drawn: (1) Spatially, the evolution of PLES exhibits clear regional differences.The ES is mainly concentrated in the east and south of Gannan and has expanded towards the periphery over time.However, the spatial distribution of the PS is the opposite of the ES, concentrating mainly in the west and north.The LS is more dispersed.Height and slope boundaries are more pronounced in the PS and in the LS, where the intensity of human activity is higher.Vertically, PLES types are more diverse in areas above 2500 m in elevation and below 40 degrees of slope.(2) We found that for the duration of the survey, the PLES in the study area was differentiated.This is manifested in a decreasing trend in the area of PS.While the area of LS and ES showed an increasing trend, the increased area mainly came from the PS.(3) The severe imbalance category is still the main type of coupling coordination over the research period.However, the proportion of basic coordination types has gradually increased over time.The moderate imbalance type has the least number of countylevel units.In general, the overall level of CCD in Gannan is relatively stable, showing a wave-like evolution characteristic from low-level coupling to high-level coupling.Spatially, the high CCD areas in Gannan are mainly located in the northeast, southwest and southeast of Gannan.It shows the pattern of "high in the east and low in the west".(4) The evolution of PLES is a complex interaction between factors such as the ecological environment, socio-economic development, human consumption needs and institutional policies.Among the natural and socio-economic factors, DEM and GDP has the greatest impact on the spatial distribution of PLES.

Figure 1 .
Figure 1.Overview of the study area: (a) the study area in China, and (b) the elevation of the study area.

Figure 4 .
Figure 4. Statistical area of PLES at different altitudes (a) and slopes (b) in Gannan 2020.

Figure 5 .
Figure 5. Transfer process of PLES in Gannan counties and cities from 2000 to 2020 (unit:km 2 ).

3. 3 .
Figure7shows that the spatial pattern and CCD of PLES functions in Gannan in 2000, 2010 and 2020 are relatively consistent.Most of the main urban areas in the eastern districts and counties of Gannan show a high degree of coordination and coupling.The trend is gradually expanding to the outer circle.The coupling coordination level of the center location of the southwestern Maqu is low, lower than the coupling coordination level of the Maqu periphery.The coupling coordination level of other regions is relatively stable.It can be seen that the severe imbalance category is still the main type of coupling coordination over the research period.However, with the development of society, the proportion of basic coordination types has gradually increased.The moderate imbalance type has the least number of county-level units.

3. 4 .
Geodetector-Based PLES Driving Force Analysis 3.4.1.Individual Factor Analysis It was found that DEM (natural factors) and GDP (socio-economic factors) had the most effect (Table

Figure 8 .
Figure 8. Interaction detection results of PLES evolution drivers in Gannan 2020.

Table 4 .
The classification criteria of CC d .

Table 5 .
Driving factors of spatial evolution for the PLES.

Table 6 .
Detected results of driving factors of spatial evolution for PLES.