Abstract
Production-living-ecological space (PLES) reflects the main function of land use types. It is one of the main directions that many scholars researched to evaluate, predict, and optimize the land space from the perspective of PLES. Yunnan Province is affected by such factors as economy, topography, and natural environment. The conflict of PLES is serious, and the problems of spatial planning development are prominent. This study aims at the current status of PLES, based on the establishment of restrictive constraints such as ecological red line, arable land minimum, and natural reserves. Meanwhile, these constraints were combined with the development planning of the Yunnan Province to forecast the quantitative structure change in the PLES in Yunnan Province in 2035 and 2050, coupling Markov and PLUS models to optimize the future space layout. This study can provide a scientific basis for the optimization of land space in Yunnan Province and other areas. The prediction accuracy of the Markov–PLUS model is 98.55%, which can be effectively used to simulate and predict the distribution of PLES in Yunnan in 2035 and 2050. From 2010 to 2015, the disordered layout of PLES in the Yunnan Province was obvious, and the ecological space (ES) seriously occupied the production space (PS) and living space (LS). In 2035 and 2050, the industrial production space (IPS) of Yunnan Province expands and presents distinct regional aggregation. LS and the water ecological space (WES) areas have increased. The layout of PLES in 2035 and 2050 of Yunnan Province mainly centers on PS. The orderly development of IPS promotes the regional economic growth, ensures that the agricultural production space (APS) will not be damaged and allocates the ES reasonably. It will also promote the overall optimization and coordinated development of PLES in Yunnan Province.
1 Introduction
Land space is the basic carrier of sustainable development of human economy and society. We must consider the optimization of land and space development pattern as the primary measure of ecological civilization construction. It was put forward in the 18th National Congress of the Communist Party of China. The outline of national land planning (2016–2030) [1], issued by the State Council of the People’s Republic of China (PRC) in 2017, points out the severe challenges we faced when planning the land space, such as the increasing resource constraints, the increasing pressure on the ecological environment, and the urgent need to optimize the land space development pattern. Meanwhile, it also puts forward the overall strategy which contains to combine the overall planning and optimization of the land space development pattern with regional coordinated development. In 2019, the opinions of the Communist Party of China (CPC) Central Committee and the State Council on establishing and supervising the implementation of the land space planning system further defined the main objectives of the land space planning [2]. This opinion put forward that “by 2035, the modernization level of land spatial governance system and governance capacity will be comprehensively improved. At the same time, intensive and efficient production space (PS), livable living space (LS), and beautiful ecological space (ES) with safe, harmonious, competitive and sustainable development will be basically formed.”
Production-living-ecological space (PLES) can be divided into PS, ES, and LS [3]. The evolution of PLES is affected by many factors, such as natural geography, resources, environment, social economy, and so on [4]. Under the complex correlation of these factors, it restricts or promotes the evolution of land space development pattern. In recent years,with the acceleration of urbanization and the increasing demand for PLES, especially for PS and LS. Which lead to the regional PLES conflict problem more and more prominent [4,5,6,7]. How to optimize the land spatial layout from the perspective of PLES, promote the balance of population, resources, and environment, and unify the economic, social, and ecological benefits have been some of the main research directions of many scholars [8].
Zhang et al. [9] used the system dynamics model (SD) and the FLUS model to build a scenario simulation model to stand the regional macroscopic land use demand and completed the spatial allocation. Meanwhile, combining the construction of China-Brazil economic corridor with regional ecological environment policy. They finally achieved the multi-scenario simulation of land use. Aburas et al. [10] based on the CA–Markov model predicted the growth of urban lands and their spatial trends in Seremban, Malaysia. And based on the CA transition rules and the transition area matrix produced from the Markov chain model-based calibration process, the future urban growth in Seremban for 2020 and 2030 was simulated. Lin et al. [11] put forward a land spatial pattern optimization allocation method based on the ecological security pattern by coupling MCR–FLUS–Markov model. They also constructed the scenario of LS expansion, PS development, ES protection, and comprehensive optimization scenario. Four models were used to predict the spatial structure of central Yunnan Urban Agglomeration in 2030. Although the above research provides technical support for PLES optimization to a certain extent, there are still many areas that need to be improved, for instance, improving the simulation accuracy of spatial optimization, increasing the reliability of the optimization results while taking the quantity structure and spatial layout into account.
This study established Markov–PLUS model. On the one hand, it highlighted the advantage of Markov chain which has no aftereffect in the process of quantity forecasting to ensure the relative accuracy. On the other hand, the RFC algorithm in the PLUS (a patch generated land use change simulation model with higher simulation accuracy, which can consider the quantitative structure objectives and the suitability of spatial function layout) model was used to excavate the developing potential of various space types, combined the multi-objective optimization algorithm, and the dynamic simulation patches were generated. It has high accuracy in predicting and optimizing the spatial distribution of various space types [12].
Yunnan Province is located in the southwest of China. The general land use planning outline of Yunnan Province (2006–2020) [13], issued by the Department of land and resources of Yunnan Province in 2009, pointed out the current situation and forms of land use in Yunnan Province. The planning also put forward that the total amount of land resources is large, but there are many constraints on land use. Land use types are rich, but the distribution is scattered. In addition to this, the structure and layout are unreasonable. The overall level of urban land use is better, but the intensity of land development is low. The level of economical and intensive land use is unbalanced among regions. These are the main problems faced by land space planning. At the same time, the task of spatial distribution and structural adjustment of land use is clarified, and put forward the strategic goal of “guiding the whole society to protect and utilize land resources reasonably, by optimizing the allocation of resources.” In January 2015, when general secretary Xi Jinping visited Yunnan, he put forward the strategic orientation of “Yunnan’s efforts to become the leader in the process of ecological civilization construction.”
Based on the above analysis, this study established a prediction and optimization model of PLES coupled with Markov and PLUS, which is a new land use change simulation model. It can effectively realize the dynamic simulation of PLES optimization at patch scale. Based on the current situation assessment and policy orientation, the quantitative structure of PLES in 2035 and 2050 of Yunnan Province will be predicted by setting restrictive conditions and objective constraints. At the same time, the PLUS model was introduced to optimize the spatial layout of PLES in 2035 and 2050 in Yunnan Province. The Markov–PLUS model takes the mutual influence, restriction, and promotion of multiple elements in the PLES into consideration, which makes the PLES in Yunnan Province reach the optimal layout from the perspective of quantity structure and spatial layout in 2035 and 2050. The results can guide the future land space development of Yunnan Province, whose methods can provide technical support for the overall planning and optimization of the PLES in Yunnan Province and other regions.
2 Geography and methods
2.1 Study area
Yunnan Province (21°8′N–29°15′N, 97°31′E–106°11′E) is located in the southwest of China, adjacent to Sichuan, Guizhou, and Guangxi in the north and Myanmar, Laos, and Vietnam in the south (Figure 1). The mountainous plateau terrain of Yunnan Province is complex and undulating. The terrain is in a ladder like downward trend from north to south and from west to East. The total land area of the province is about 394,100 km2. The mountainous area is about 349,300 km2, accounting for 88.64% of the total area of the province. Yunnan Province has subtropical and tropical monsoon climate. The rainfall is abundant, but the spatial and temporal distribution is extremely uneven, which is easy to form strong convective weather. Because of this, debris flow, landslides, and other disasters occur frequently. At the same time, affected by the plate movement, the earthquake disaster rate in Yunnan ranks first among all provinces and cities in China. According to the statistics of the National Bureau of Statistics, the GDP of Yunnan Province in 2019 was CNY 2,322.38 billion, ranking 17th in China. According to the 7th national census bulletin, Yunnan Province had a population of 4.72 million, ranking 12th in the country. Due to the large population, backward economic situation, unique natural geography, resources, and environmental conditions, the problems that exist in the spatial structure and layout of PLES in Yunnan Province have become more and more prominent. This has seriously hindered the construction of ecological civilization and the development of land space in Yunnan Province. For instance, space types are rich, but the overall layout is scattered, and the level of efficient and intensive use of space is low. The ES is too large and the terrain condition is complex, which limit the development of PS and LS. With the development of urbanization and industrialization, the existing agricultural production space (APS) is squeezed out. On the basis of protecting the ES and APS, we should spare no efforts to coordinate the development of PS and LS, improve the overall utilization level of PLES in Yunnan Province and ensure the effective transformation of its land space development pattern and regional sustainable development. These have become one of the key problems that remain to be solved in Yunnan Province.
Location and topography of the study area.
2.2 Research framework
The following steps have been taken to develop the research framework of the special prediction and optimization of PLES in Yunnan Province in 2035 and 2050:
Step 1: Establish the classification system of PLES. Evaluate the present distribution of PLES. Diagnose the conflict mechanism of PLES. Analyze the evolution trends of PLES.
Step 2: Based on the current situation assessment and policy orientation, set up the urban development boundary, permanent basic farmland protection red line, ecological protection red line and other restricted areas, and the land space planning development goal constraints of Yunnan Province. Meanwhile, use Markov model to predict the quantitative structure of PLES in Yunnan Province in 2035 and 2050.
Step 3: coupling the PLUS model with the Markov model to simulate and analyze the detailed patch evolution characteristics of PLES, and make the distribution of PLES in Yunnan Province in 2035 and 2050 reasonable.
2.3 PLUS model
Liang et al. [12] proposed the PLUS model. It is a patch-generating land use simulation based on the patch of grid data, which uses a new analysis strategy – land expansion analysis strategy (LEAS). By extracting and sampling the expansion part of each land use type between the two periods of land use change. They employed a random forest classification (RFC) algorithm to explore the relationships between the growth in each land use type and the multiple driving factors, so as to obtain the development probability of each land type and the contribution of each driving factor to all kinds of land expansion in a specific period. LEAS integrates the advantages of transition analysis strategy and pattern analysis strategy, and avoids the analysis of transformation types with the exponential growth of category number. Meanwhile, it has retained the ability of the model to analyze the mechanism of land use change over a period of time. In addition, it also includes a new type of CARS (CA based on Multiple Random Seeds) model. It combines the mechanism of random seed generation and threshold decrement to make the model automatically generate dynamic simulation patches under the constraint of development probability. Meanwhile, it can be coupled with multi-objective optimization algorithm to make the simulation results have strong robustness.
The RFC algorithm is a decision tree-based ensemble classifier, which is able to process high-dimension data as well as deal with multicollinearity among variables, and finally output the growth probability
The value of s is either 0 or 1, a value of 1 indicates that there were other land use types that changed to land use type i, while 0 represents other transitions. x is a vector that consists of multiple driving factors. I(·) is the indicative function of the decision tree set. h n (x) is the prediction type of the n-th decision tree for vector x. N is the total count of decision trees.
2.4 Markov chain
Markov is a kind of model which is suitable for modeling stochastic system. In the modeling process, it is assumed that the future state only depends on the current state, not affected by any previous state. This feature makes the model play an important role in the field of prediction modeling and probability prediction. Markov chain is a random variable sequence X n with Markov property. By using the method of probability theory, the present state and development trend of a variable are used to predict the future state and trend, and the state t + 1 is only related to the state t, which has no aftereffect [14].
State equation:
State probability:
Transition probability:
Markov chain can calculate the transition probability of spatial area in a random time series. It is based on the current spatial state of data distribution and the change in spatial structure. The transition probability matrix is arranged as follows:
where P ij shows the transferring probability from i state to j state.
Through the transition probability matrix of A → B state, to calculate the development direction of B in the process of B → C state, the development status of C state is obtained.
3 Establishment of Markov–PLUS model
3.1 Select drivers
The layout of PLES is the consequence driven by multiple factors. According to the actual situation of Yunnan Province, the feasibility of relevant data acquisition, and comprehensive domestic and foreign research conclusions of relevant driving factors [15,16,17,18,19,20,21], three factors were selected in this study: terrain factor, location factor, and natural factor. They include eight types of driving factors for PLES evolution in Yunnan Province (Table 1).
The driving factors of Markov–PLUS model
| Order number | Driver type | Drivers | Data source/processing |
|---|---|---|---|
| 1 | Terrain factor | DEM | DEM data come from the data center of resources and environment science, Chinese Academy of Sciences (https://www.resdc.cn) |
| 2 | Slope | Based on the DEM, using ArcGIS (Slope extraction) to generate | |
| 3 | Aspect | Based on the DEM, using ArcGIS (Aspect extraction) to generate | |
| 4 | Location factor | Distance from road | Based on the highway data, using ArcGIS (Euclidean distance) to generate. The data are from National Geographic Data Center (https://www. geodata.cn) |
| 5 | Distance from school and hospital | Based on the POI data of schools and hospitals, using ArcGIS (Euclidean distance) to generate. The data is encoded by Python and crawled on the Internet | |
| 6 | Natural factor | Temperature | The data are from the database of the world climate data (https://www.worldclim.org/data/index.html) |
| 7 | Precipitation | The data source is WorldClim database of global weather and climate data (https://www.worldclim.org/data/index.html) | |
| 8 | NDVI | NDVI data are from moderate-resolution imaging spectrometer (MODIS, https://neo.sci.gsfc.nasa.gov/view.php?datasetId=MOD_NDVI_M) |
Terrain factors determine the types of PLES in essence. To a certain extent, altitude, relief, aspect, and other terrain factors affect climate condition, soil, and water resources distribution, and then affect the layout of PLES. In this study, Digital Terrain Model (DEM), slope, and aspect of terrain factors were selected as the driving factors of the PLES evolution.
Location factors promote the transfer and aggregation of urban PS and LS. The area with serried road network and high traffic accessibility has complete infrastructure. Meanwhile, the development of LS and PS is faster than other areas. The area close to school and hospital will produce the phenomena of LS agglomeration. In this study, we selected the spatial location data of roads, hospitals, and schools. These position factor data were obtained by calculating the Euclidean distance of the position factors.
Among the natural factors, climate, precipitation, and other factors have complex effects on the layout of PLES, especially on the remote mountainous areas of Yunnan Province. In these areas, the frequent occurrence of extreme weather such as rainstorm and blizzard inhibits the development of PS and LS, and threatens the safety of human life and property. In this study, the temperature, precipitation, and normalized vegetation index (NDVI) were selected as the natural driving factors for the evolution of PLES in Yunnan Province.
The ArcGIS was used to unify spatial resolution and coordinate system after obtaining the driving data. And then we constructed the Markov–PLUS model with these data to simulate the distribution of PLES in Yunnan Province in 2035 and 2050.
3.2 Identify restricted conversion areas
Government departments regulate and control the distribution of various spatial types through land spatial planning and related policies. Based on the principles of ecological priority and food security, the control of land space use was implemented, which has always been an important way of land space planning. In Yunnan province, the ecological environment is fragile, the anti-interference ability is poor, and the ecosystem service function is seriously declined. In 2019, the Delimitation scheme of ecological protection red line in Yunnan Province delimits the ecological protection red line with an area of 118,400 km2, accounting for 30.90% of the total land area of the province [22]. The development restriction of ecological protection red line is particularly important in the optimization of PLES layout in Yunnan Province.
Based on the basic national conditions of less cultivated land per capita, China has been implementing a strict cultivated land protection system. Meanwhile, the basic farmland protection zones in each area are defined according to the principle of “the total amount is not reduced, the use is not changed, the quality has been improved, and the overall layout is stable.” The protection of basic farmland is an important goal of the current land space planning. In Yunnan Province, the APS is relatively large, but the overall quality of the cultivated land is poor. With the development of urbanization and industrialization, the APS is gradually occupied by other spaces. In the future development planning, strictly preventing the red line of the cultivated land, protecting permanent basic farmland and improving the utilization efficiency of APS are also important for optimizing the layout of PLES in Yunnan Province. In addition, the important control lines and protected areas of land space planning such as natural reserves and urban development boundaries are also the restricted transformation areas which are set in this study.
3.3 Suitability probability analysis of different space types
This study is based on the current situation data of PLES in 2010 and 2015 and the binary data of restricted transformation area when using the PLUS model to simulate the layout of PLES in the future. The PLUS model was used to extract the expansion area of PLES in Yunnan Province and calculate the development probability of each space type by LEAS. After many experiments, we set the random sampling decision tree, rate, and training characteristic number to 20, 0.15, and 8, respectively, and got the suitability development probability of each space type (Figure 2).
Suitability development probability of PLES in Yunnan Province. (a) Represents agricultural production space, (b) represents industrial production space, (c) represents urban living space, (d) represents rural living space, (e) represents forestland ecological space, (f) represents grassland ecological space, (g) represents water ecological space, (h) represents other ecological space.
As shown in Figure 2, the suitable development areas of APS and rural living space (RLS) occupy most areas of the Yunnan Province. The suitable areas for the development of industrial production space (IPS) are mainly concentrated in the eastern and northern areas of the Yunnan Province. The areas with high suitability for urban living space (ULS) development are mainly in the western and eastern regions, especially in the junction area of Kunming and Yuxi. The suitable areas for the development of forestland ecological space (FES), grassland ecological space (GES), and water ecological space (WES) are concentrated in a small part of the northwest, northeast, and south. In addition, the development of WES is also distributed in the central part of Yunnan Province, where the APS is concentrated. The areas with high development suitability are expanded from the original space types to the surrounding areas, showing radioactive development. With the diffusion of distribution, the suitability is gradually reduced.
3.4 Setting neighborhood weights
The neighborhood factor parameter represents the expansion intensity of each space type, it is the expansion capacity of each type under the influence of driving factors, and the threshold value is [0–1]. The closer the value to 1, the stronger the expansion capacity of the space types [11]. In order to tally or accord with the expression of the expansion law of each space type, this study determines the neighborhood weight parameters by calculating the dimensionless amount of the historical total area change (△X i ) of each space type [15] (Table 2). Then, the data should be translated to avoid the occurrence of zero value, which can ensure the objectivity of neighborhood weight setting and meet the needs of data structure (Table 3).
Area change of different space types
| Type | APS | IPS | ULS | RLS | FES | GES | WES | OES |
|---|---|---|---|---|---|---|---|---|
| ΔX i | −459 | 530 | 47 | 35 | −359 | −310 | 503 | 13 |
Note: APS represents agricultural production space, IPS represents industrial production space, ULS represents urban living space, RLS represents rural living space, FES represents forestland ecological space, GES represents grassland ecological space, WES represents water ecological space, OES represents other ecological space.
Neighborhood weight coefficient of Markov–PLUS model in PLES of Yunnan Province
| Type | APS | IPS | ULS | RLS | FES | GES | WES | OES |
|---|---|---|---|---|---|---|---|---|
| Standardized value | 0.01 | 1 | 0.5149 | 0.4951 | 0.1189 | 0.1684 | 0.9703 | 0.4753 |
Note: APS represents agricultural production space, IPS represents industrial production space, ULS represents urban living space, RLS represents rural living space, FES represents forestland ecological space, GES represents grassland ecological space, WES represents water ecological space, OES represents other ecological space.
Through the calculations of expansion and standardized value, it can be seen that the expansion capacity of APS, FES, and GES is weak, while that of LS, IPS, WES and other ecological space (OES) is strong. Combining the transfer probability matrix of space types in the Yunnan Province from 2010 to 2015, and referring to the relevant research works [11,15,20], the parameter combination was debugged repeatedly to improve the simulation accuracy in the simulation process, and the final set of neighborhood factor parameters is shown in Table 4.
Neighborhood factor parameter
| Type | APS | IPS | ULS | RLS | FES | GES | WES | OES |
|---|---|---|---|---|---|---|---|---|
| Weight | 0.3 | 0.7 | 0.8 | 0.6 | 0.3 | 0.3 | 0.7 | 0.4 |
Note: APS represents agricultural production space, IPS represents industrial production space, ULS represents urban living space, RLS represents rural living space, FES represents forestland ecological space, GES represents grassland ecological space, WES represents water ecological space, OES represents other ecological space.
3.5 Precision test of Markov–PLUS
Based on the above analysis, the spatial distribution of PLES in Yunnan Province in 2015 was simulated according to the principle of “ecological priority and food security.” Compared with the current distribution of PLES in Yunnan Province in 2015, the results show that the kappa coefficient of Markov–PLUS model is 0.9855, and the overall simulation accuracy is 99.14%. According to the kappa coefficient evaluation standard, kappa coefficient is greater than 0.8, based on which it can be considered that the model has a good fitting effect on spatial distribution. It is suitable to simulate the change in the PLES in Yunnan Province in the future.
4 Results
4.1 Evolution of PLES pattern in Yunnan Province
Figure 3 shows the quantity structure and spatial distribution of PLES in Yunnan Province in 2010 and 2015. The distribution of PLES in Yunnan Province showed obvious regional characteristics. The proportion of ES between 2010 and 2015 was over 80%, mainly FES and GES. The PS is less than 20%, most of which are APS, while the IPS is small, the distribution is scattered. Furthermore, the overall degree of intensification is low. The proportion of LS is very small (less than 1%), RLS mainly develops along the WES, and ULS is mostly concentrated in the south of Kunming.
Quantity structure and spatial distribution of PLES in Yunnan Province in 2010 (a) and 2015 (b).
From 2010 to 2015, the pattern of PLES in the Yunnan Province changed significantly. Among them, the ES is mainly transferred out, FES and GES are, respectively, transferred out of 359 and 310 km2, which are mainly transformed into IPS and ULS. Significantly, there is only a small amount of WES scattered in a large area of APS in the east and west of Yunnan Province. With the improvement of the rationality of land spatial planning, the proportion and distribution of water area in 2015 was significantly better than that in 2010, with an increase of 503 km2. The IPS has shifted to 530 km2, and the unreasonable expansion of IPS can promote the development of regional economy. However, this encroaches on the APS leads to 459 km2 reduction in APS. More seriously, it may threaten regional food security. Compared with the large area of ES, ULS and RLS increased slightly. It is still difficult to change the imbalance of PLES. The unbalanced development of PLES in the Yunnan Province is still prominent, which is mainly manifested in the improper spatial distribution, the low level of space intensive and efficient utilization.
4.2 Quantitative structure prediction
Table 5 shows the transition probability matrix of PLES in the Yunnan Province from 2010 to 2015. According to Table 5, IPS and ULS had the largest transfer area from 2010 to 2015, and the probability of transferring to other space types was 0.0344 and 0.0195. The main transfer out of IPS was FES (accounting for 50.00% of the total transfer out area). ULS was mainly converted to IPS (accounting for 69.23% of the total transferred out area). IPS occupies ULS seriously, and part of ULS was converted to APS. It can be seen that with the acceleration of urbanization, the distribution of IPS has been gradually transferred to residential areas, and most of the areas were not suitable for the development of IPS, which are transformed to WES and GES. OES and FES have the lowest transfer probability. OES may be transformed into WES due to soil erosion. If it is suitable for plant growth or through artificial transformation, those areas will be converted into FES in a period of time. Part of the FES will develop towards the PS according to the needs and actual situation. Combined with the overall situation of transfer in and out, the changes in LS and OES tend to be stable.
Spatial transfer probability matrix of PLES in Yunnan Province from 2010 to 2015
| 2015 | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| APS | IPS (%) | ULS (%) | RLS (%) | FES (%) | GES (%) | WES (%) | OES (%) | |||
| 2010 | APS | 99.28 | 0.35 | 0.06 | 0.05 | 0.01 | 0.02 | 0.23 | 0.00 | |
| IPS | 0.86 | 96.55 | 0.00 | 0.43 | 1.72 | 0.43 | 0.00 | 0.00 | ||
| ULS | 0.15 | 1.35 | 98.05 | 0.00 | 0.30 | 0.00 | 0.15 | 0.00 | ||
| RLS | 0.00 | 0.42 | 0.07 | 99.44 | 0.00 | 0.07 | 0.00 | 0.00 | ||
| FES | 0.01 | 0.06 | 0.00 | 0.00 | 99.82 | 0.02 | 0.09 | 0.00 | ||
| GES | 0.01 | 0.17 | 0.01 | 0.01 | 0.03 | 99.57 | 0.21 | 0.00 | ||
| WES | 0.42 | 0.07 | 0.10 | 0.00 | 0.00 | 0.42 | 98.47 | 0.52 | ||
| OES | 0.00 | 0.00 | 0.00 | 0.00 | 0.05 | 0.00 | 0.05 | 99.90 | ||
Note: APS represents agricultural production space, IPS represents industrial production space, ULS represents urban living space, RLS represents rural living space, FES represents forestland ecological space, GES represents grassland ecological space, WES represents water ecological space, OES represents other ecological space.
In addition, FES, GES, and APS in the Yunnan Province changed greatly from 2010 to 2015. However, the ecological red line area accounted for 30.90% of the total area in the Yunnan province, and the cost of transforming IPS and LS into OES was also large. Therefore, there are few reserve cultivated land resources for development in the Yunnan Province. It is necessary to further strictly set the red line of the cultivated land and scientifically plan the development of land space.
Table 6 shows the change in space types in the Yunnan Province in 2035 and 2050, which is based on the change transition probability matrix of PLES in the Yunnan Province from 2010 to 2015 (Table 6 and Figure 4). Table 6 shows that compared with the current development situation, the areas of APS, FES, and GES will gradually decrease in 2035 and 2050, and the areas of IPS, ULS, RLS, and WES will increase year by year, among which the IPS and WES will increase greatly. Based on the above development, the future trend of land space use makes up for the lack of industrial development, and the imbalance between the development of WES and APS. It can be seen from Table 6 and Figure 4 that in the future development of Yunnan Province, the proportion of ES is large, the PS and LS are small, the spatial distribution is unbalanced, the LS and PS are seriously occupied by ES, and the problems of low degree of urban development still exist. Therefore, Yunnan Province should increase the investment in the PS and LS, reasonably plan the types of space utilization, and ensure the harmonious development of human and nature.
Quantity structure distribution of PLES in Yunnan Province (km2)
| Actual | Forecast | Increment (compared with 2015) | ||||||
|---|---|---|---|---|---|---|---|---|
| Type | 2010 | 2015 | 2015 | 2035 | 2050 | 2035 | 2050 | |
| PS | APS | 68,355 | 67,896 | 67,896 | 66,157 | 64,948 | −1,739 | −2,948 |
| IPS | 232 | 762 | 762 | 2,694 | 3,964 | 1,932 | 3,202 | |
| LS | ULS | 666 | 713 | 713 | 894 | 1,023 | 181 | 310 |
| RLS | 1,419 | 1,454 | 1,371 | 1,612 | 1,746 | 158 | 292 | |
| ES | FES | 218,555 | 218,196 | 219,059 | 216,859 | 215,946 | −1,337 | −2,250 |
| GES | 88,190 | 87,880 | 87,880 | 86,694 | 85,858 | −1,186 | −2,022 | |
| WES | 2,884 | 3,387 | 3,387 | 5,300 | 6,641 | 1,913 | 3,254 | |
| OES | 2,097 | 2,110 | 1,330 | 2,187 | 2,271 | 77 | 161 | |
Note: APS represents agricultural production space, IPS represents industrial production space, ULS represents urban living space, RLS represents rural living space, FES represents forestland ecological space, GES represents grassland ecological space, WES represents water ecological space, OES represents other ecological space.
Proportion of quantity structure of PLES in Yunnan Province.
4.3 Spatial layout optimization results
Figure 5 shows the spatial distribution of Yunnan Province in 2035 and 2050, which is based on the optimization results of the quantitative structure of Yunnan Province in 2035 and 2050. It can be seen from Figure 5 that compared with the distribution of historical space use, the WES and IPS are significantly increased. Most of the newly increased WES are distributed in the area of APS concentration, which makes up for the imbalance of APS and WES in the east and west of Yunnan Province to a certain extent. The increase in IPS is mainly around the ULS and WES, and the regional distribution shows a trend of centralized development, which is consistent with the intensive development advocated by the “14th 5-year plan,” and also confirms the reliability of the simulation results. The area of FES in Northwest of Yunnan province increased greatly. This phenomenon shows that the biodiversity conservation and management area in Northwest Yunnan and Sanjiang shelterbelt management area set up by the Medium-Long Term Development Plan of Yunnan National Forest (2019–2035) have a good development [23]. A large area of newly increased WES and IPS are converted from FES, and located in the area with small slope fluctuation. This transformation phenomenon alleviates the problem of ES occupying PS and LS. However, the LS of some small areas is occupied by IPS in the simulation process. The main reason is that the small patch land and the patch located at the boundary of a certain space type are not stable, which are easily disturbed by many factors in the development process, thus leading to the great randomness of its evolution direction [24].
Spatial distribution of Yunnan Province in 2035 (a) and 2050 (b).
5 Discussion and policy implications
5.1 Discussion
Under the background of devoting to create ES with mountains and rivers, PS with plateau characteristics, intensive and efficient LS, and creating beautiful Yunnan, we analyzed and evaluated the evolution pattern of PLES from 2010 to 2015 in the Yunnan Province, established an overall PLES optimization model which coupled Markov model and PLUS model, and considered the evolution law of PLES along with the coupling of terrain factors, location factors, and environmental factors. Meanwhile, it also sets up restricted transformation areas. Finally, it forecasts and optimizes the PLES in Yunnan Province in 2035 and 2050 from the perspective of quantity structure and spatial layout, the prediction accuracy of the model is 98.55%.
The results show that there is obvious irrationality in the quantity structure of PLES in the Yunnan Province, and the area of ES is much higher than that of LS and PS. It is mainly the FES and GES that are difficult to develop because of the constraints of the terrain, which also ensures that the ecological red line of the Yunnan Province will not be damaged in some degrees. The prediction results show that WES, as a necessary condition of PS and LS, will increase significantly in the future, and mainly focus on the development of APS. With the rapid development of economy, the area of PS is gradually increasing, especially the IPS, which mainly presents the trend of intensive development around LS. The prediction results are consistent with the 14th 5-year Plan and the Medium-Long Term Development Plan of Yunnan National Forest (2019–2035), which reflected the reliability of the prediction model established in this study, and showed that the prediction results can provide a reliable reference for PLES optimization in the Yunnan Province in the future.
It is worth noting that Yunnan Province is located in the Yunnan–Guizhou Plateau, with frequent landslides, debris flows, earthquakes, and other geological disasters, all of which seriously threaten the safety of human life and property. This study considered the influence of many factors on the development of PLES. However, limited by data, the high incidence area of geological disasters is not considered as one of the influencing factors to guide the land space planning, which limits the development of PS and LS in this area. In the future study, the high incidence area of geological disasters in the Yunnan Province needs to be further regarded as the main restricted area, and the influence scope of different disasters should be fully considered to evaluate the current situation of PS and LS in the scope and the economic benefits of transforming the current situation to the low incidence area of geological disasters [25,26].
5.2 Policy implications
Combing the above analysis and the existing land use planning of Yunnan Province, this article put forward the following policy implications to promote the rational layout and sustainable development of PLES in the Yunnan Province:
First, strive to adjust the spatial distribution, reasonably plan the spatial type, and increase the investment in PS and LS. More importantly, increase the water area distribution in the APS and improve the current situation of low agricultural yield caused by uneven water area distribution. Second, promote the development of state-owned forest farms in Northwest of Yunnan. And promote the healthy development of biodiversity conservation management area and Sanjiang shelterbelt management area. Third, focus on economic development, pay attention to the development of areas suitable for industrial development, improve the current situation of insufficient economic development in the Yunnan Province, and promote the orderly implementation of the task of poverty alleviation.
6 Conclusion
In this study, the Markov–PLUS model was established to analyze the current situation evolution and predict future development trend from the quantitative structure and spatial distribution of PLES in the Yunnan Province. According to the quantity structure of PLES in the Yunnan Province from 2010 to 2015, the level of economic development in Yunnan Province has improved. The APS has basically implemented the “balance of occupation and compensation.” The forecast results of PLES in 2035 and 2050 show that the industrial development will be remarkable in the future. The development benefits from major national strategies and policies are converged in Yunnan. The WES increases, which made up for the low quality of agricultural development. But the development style of Yunnan Province is still extensive. At the same time, the county economy is not strong, the level of urbanization is low, and the foundation for the continuous improvement of ecological environment quality is not solid. These problems remain to be the focus of future development. Furthermore, the method used in this study can also be used to evaluate the current situation of PLES and predict the future development trend, so that the decision makers formulate targeted land spatial planning, which is worth promoting.
Acknowledgments
This research was supported by the National Natural Science Foundation of China (Grant No. 42064001).
-
Conflict of interest: Authors state no conflict of interest.
-
Author contributions: L.W. and S.Z. contributed to all aspects of this work; S.O. conducted data analysis and L.W. wrote the main manuscript text. All authors reviewed the manuscript.
References
[1] Central People’s Government of the People’s Republic of China; 2017 (in Chinese). http://www.gov.cn/zhengce/content/2017-02/04/content_5165309.htm. 4/2/2017.Search in Google Scholar
[2] Central People’s Government of the People’s Republic of China 2019 (in Chinese) http://www.gov.cn/zhengce/2019-05/23/content_5394187.htm 29/5/2019Search in Google Scholar
[3] Zhang XS, Xu ZJ. Functional coupling degree and human activity intensity of production–living–ecological space in underdeveloped regions in China: case study of Guizhou Province. Land. 2021;10:1–13. 10.3390/land10010056.Search in Google Scholar
[4] Cai EX, Jing Y, Liu YL, Yin CH, Gao Y, Wei JQ. Spatial-temporal patterns and driving forces of ecological-living-production land in Hubei Province, Central China. Sustainability. 2018;10:1–15.10.3390/su10010066Search in Google Scholar
[5] Dong ZH, Zhang JQ, Si A, Tong ZJ, Na L. Multidimensional analysis of the spatiotemporal variations in ecological, production and living spaces of inner mongolia and an identification of driving forces. Sustainability. 2020;12(19):1–15. 10.3390/su12197964.Search in Google Scholar
[6] Yang YY, Bao WK, Li YH, Wang YS, Chen ZF. Land use transition and its eco-environmental effects in the Beijing–Tianjin–Hebei urban agglomeration: a production–living–ecological perspective. Land. 2020;9(9):1–24. 10.3390/land9090285.Search in Google Scholar
[7] Lin. G, Jiang D, Fu J, Cao C, Zhang. D. Spatial conflict of production–living–ecological space and sustainable-development scenario simulation in Yangtze River delta agglomerations. Sustainability. 2020;12:1–11. 10.3390/su12062175.Search in Google Scholar
[8] Sun XF, Yu C, Wang JB, Wang M. The intensity analysis of production living ecological land in Shandong Province, China. Sustainability. 2020;12:1–17. 10.3390/su12208326.Search in Google Scholar
[9] Zhang XR, Li AN, Nan X, Lei GB, Wang CB. Multi-scenario simulation of land use change along China-Pakistan economic corridor through coupling FLUS model with SD model. J Geo-Inform Sci. 2020;22(12):2393–409 (in Chinese with English abstract). 10.12082/dqxxkx.2020.190618.Search in Google Scholar
[10] Aburas MM, Ho YM, Pradhan B, Salleh AH. Spatio-temporal simulation of future urban growth trends using an integrated CA-Markov model. Arab J Geosci. 2021;14:1–12. 10.1007/s12517-021-06487-8.Search in Google Scholar
[11] Lin YL, Zhao JS, Chen GP, Zhang M. Territory space pattern optimization of the urban agglomeration in Central Yunnan Based on MCR-FLUS–Markov model. Trans Chin Soc Agric Machinery. 2021 (in Chinese with English abstract).Search in Google Scholar
[12] Liang X, Guan QF, Clarke KC, Liu SS, Wang BY, Yao Y. Understanding the drivers of sustainable land expansion using a patch-generating land use simulation (PLUS) model: A case study in Wuhan, China. Computers Environ Urban Syst. 2021;85:1–14. 10.1016/j.compenvurbsys.2020.101569.Search in Google Scholar
[13] Ministry of Natural Resources of the People’s Republic of China; 20 Jan 2011 (in Chinese). http://www.mnr.gov.cn/gk/ghjh/201811/t20181101_2324675.html.Search in Google Scholar
[14] Lin XD, Fan SL, Sun QY, Tang JH, Zhang ZZ. Spatial simulation and analysis of land use pattern based on GIS technology and Logistic regression model: a case study in Longhai City Journal of Fujian Agriculture and Forestry University (Natural Science Edition). 2017;46(4):468–173 (in Chinese with English abstract). 10.13323/j.cnki.j.fafu(nat.sci.).2017.04.018.Search in Google Scholar
[15] Wang BS, Liao JF, Zhu W, Qiu QY, Wang L, Tang LN. The weight of neighborhood setting of the FLUS model based on a historical Scenario: A case study of land use simulation of urban agglomeration of the Golden Triangle of Southern Fujian in 2030. Acta Ecol Sin. 2019;39(12):4284–98 (in Chinese with English abstract).Search in Google Scholar
[16] Zhang SW, Wei LY, Jin XX, Lu YQ. The land use simulation and delimitation of urban development boundary in county area based on FLUS-UGB. J Geo-Inform Sci. 2020;22:1848–59 (in Chinese with English abstract). 10.12082/dqxxkx.2020.190452.Search in Google Scholar
[17] Wang MC, Guo X, Wang FY, Zhang XY. Dynamic change and predictive analysis of land use type in Changchun city based on FLUS model. J Jilin Univ (Earth Sci Ed). 2019;49(6):1759–804 (in Chinese with English abstract). 10.13278/j.cnki.jjuese.20180312.Search in Google Scholar
[18] Ding QL, Chen Y, Bu LT, Ye YM. Multi-scenario analysis of habitat quality in the yellow river delta by coupling FLUS with InVEST model. Int J Env Res Public Health. 2021;18(5):1–19. 10.3390/ijerph18052389.Search in Google Scholar
[19] Feng DR, Bao WK, Fu MC, Zhang M, Sun YY. Current and future land use characters of a national central city in eco-fragile region—a case study in Xi’an city based on FLUS model. Land. 2021;10(3):1–25. 10.3390/land10030286.Search in Google Scholar
[20] Zhao X, Peng JD, Fan ZY, Yang C, Yang H. Land Use Simulation and Urban Growth Boundaries Delineation in Wuhan Metropolitan Area Based on FLUS Model and “Dual Environment Evaluation.” J Geo-Inform Sci. 2020;22(11):2212–26 (in Chinese with English abstract). 10.12082/dqxxkx.2020.200137.Search in Google Scholar
[21] Long Y, Li P, Tang YL, Duan JN. Simulation of land use change in changliuping hilly area under farmland protection: application of FLUS model. Acta Agric Jiangxi. 2020;32(5):126–34 (in Chinese with English abstract). CNKI:SUN:JXNY.0.2020-05-024.Search in Google Scholar
[22] People’s Government of Yunnan Province; 29 June 2018 (in Chinese). http://www.yn.gov.cn/zwgk/zfgb/2018/2018ndsyq/szfwj_1447/201806/t20180629_145923.html.Search in Google Scholar
[23] Zhejiang Forestry Bureau; 26 March 2019 (in Chinese). http://www.zjly.gov.cn/art/2019/3/26/art_1276367_31639669.html.Search in Google Scholar
[24] Zheng Y, Hao RM, Wu XG, Wang K. Research on the spatial pattern optimization of production-living-ecological spaces in village based on MCR model. Res Soil Water Conserv. 2021;28(5):1–6 (in Chinese with English abstract). 10.13869/j.cnki.rswc.20210412.001.Search in Google Scholar
[25] Yang YY, Bao WK, Liu YS. Coupling coordination analysis of rural production-living-ecological space in the Beijing-Tianjin-Hebei region. Ecol Indic. 2020;117:1–13. 10.1016/j.ecolind.2020.106512.Search in Google Scholar
[26] Zou LL, Liu YS, Yang JX, Yang SF, Hu XD. Quantitative identification and spatial analysis of land use ecological-production-living functions in rural areas on China’s southeast coast. Habitat Int. 2020;100:1–11. 10.1016/j.habitatint.2020.102182.Search in Google Scholar
© 2022 Lingling Wang et al., published by De Gruyter
This work is licensed under the Creative Commons Attribution 4.0 International License.
