Eco-Environmental Effects of Changes in Territorial Spatial Pattern and Their Driving Forces in Qinghai, China (1980–2020)

Abstract

As urbanization and industrialization have advanced in leaps and bounds, the territorial spatial pattern of Qinghai has experienced profound transformation and reconstruction, which has been directly reflected in land-use changes and affected the eco-environment. In this context, we constructed a functional classification system of “production-living-ecological” (PLE), used remote sensing data for six periods from 1980 to 2020, and employed the land transfer matrix, eco-environmental quality index, ecological contribution rate of land-use transformation and geographical detectors to analyze the changes in the territorial spatial patterns, eco-environmental effects and driving forces of eco-environmental quality. The results revealed that (1) the spatial distribution of the province was characterized by the relative agglomeration of the production and living spaces and the absolute dominance of ecological spaces; (2) The eco-environmental quality of the region portrayed a steady improvement, with a significant reduction in the medium–low and low-quality areas; and (3) the annual average precipitation, proportion of non-agricultural area, and socio-economic factors had a significant impact on the eco-environmental quality of the region, meanwhile, national economy and ecological policies are important indirect driving forces of eco-environmental quality. Our findings will provide guidelines for territorial spatial management and serve as a reference for eco-environmental protection in Qinghai.

1. Introduction

Territorial space is an important foundation for national survival and social and economic development, which is the carrier and place for human production and life. It is a composite dynamic system composed of natural ecological and socio-economic elements [1,2]. Changes in territorial spatial pa-terns are manifested as changes in the interaction and functional connections between different land use types in the region [3,4]. This, in turn, affects changes in the society, economy, and ecology of the region and is also an important link in studying regional systems of human–natural coupling relationships [5,6].

Notably, the concept of land use form was introduced by British geographer Alan Grainger in 1995 [7], Since then, scholars have studied the territorial spatial pattern, based on land use, while considering systematization, hierarchy, and diversification [8,9,10,11,12]. An increasing number of studies are analyzing the spatial patterns of land under the influence of land use change from different perspectives [13,14], while analyzing the evolution of different land types [14,15,16] and influencing factors and effect methods [17,18].

The land-use changes in the territorial spatial pattern of a region are affected by the natural background conditions and socio-economic development of the region [19,20]. Simultaneously, the changes also affect the regional climate and ecological environment. Foley [21] concluded that changing land use strategies can effectively improve the negative effects of food production, freshwater resources, forest resources, regional climate, and air quality (especially the spread of infectious diseases). Peters [22] analyzed the comprehensive effects of climate and land use on the biodiversity and eco-system functions in Kilimanjaro. Notably, previous studies have indicated that climate can regulate the impact of land use on biodiversity and ecosystem function. Disordered transition and irrational utilization will lead to a series of problems such as ecological environment deterioration and ecosystem service function decline [23,24]. To study the effect of land use on ecosystem service function, scholars have focused more on the areas where the national development strategy has high importance, or the ecological environment is fragile [18,25,26]. The eco-environmental effect of territorial spatial pattern change, and its driving force analysis, are one of the important ways to optimize environmental protection, ensure food security, and promote the economical and intensive utilization of land and resources [27,28,29].

Since 1978, China has experienced a reform; the rapid development of its economy has accompanied a series of issues related to sustainable development between the production and living activities in territorial spatial patterns and between human settlements and natural ecosystems [30,31,32,33]. Solving problems related to the rapid development and transformation of the social economy, such as those regarding land development order and the heavy costs associated with resources and the environment, has always been an important scientific aspect of the study of the human-natural coupling relationship in the field of regional sustainable development [34,35]. Qinghai is the ‘Asian water tower’ and contains the Three Rivers Source, which plays a very important role in the global ecological development. With the rapid development of urbanization and industrialization, the spatial form of land and the spatial pattern of production, life, and ecology in the Qinghai province are also changing [36,37]. In this context, through the construction of the functional classification system of production-living-ecological (PLE) space, it is used to analyze the relationship between macro spatial pattern and micro land use. Meanwhile, it will be helpful to explore the characteristics of territorial spatial pattern change at the macro scale and study the eco-environment quality and its driving factors caused by land use at the micro scale [37,38,39].

In the context of sustainable development, eco-environment quality changes based on territorial spatial patterns and factors affecting the changes have attracted much attention. It will be of great theoretical and practical significance to study the spatio-temporal pattern change of territorial spatial patterns, eco-environmental quality and its driving forces in the process of social and economic development. Specifically, our objectives were to (1) from the perspective of PLE land use function, quantify the spatio-temporal changes of Qinghai’s territorial space pattern from 1980 to 2020; (2) employing the eco-environmental quality index and the ecological contribution rate method of land use transformation, the dynamic change of eco-environmental quality and ecological contribution of land use transformation were analyzed; and (3) The driving factors affecting the change of ecological environment quality were analyzed and the driving mechanism was discussed in order to provide reference strategies for Qinghai Province to realize the rational utilization of land resources and spatial planning, and to formulate differentiated ecological protection policies for regions.

2. Materials and Methods

2.1. Study Area

Qinghai Province is located in the northeast of the Qinghai-Tibet Plateau (36°31′–39°19′ N latitude, 89°35′–103°04′ E longitude) (Figure 1). The region has a plateau continental climate, with an average annual temperature of 2–9 °C and an annual rainfall of 250–550 mm. Qinghai is the birthplace of the Yellow, Yangtze, and Lancang rivers and one of the most important ecological protection barriers in China. At present, due to the fragile ecological environment restricted by topography and resources, the exploitable land resources in Qinghai province are scarce and the land use structure is simple [40]. The above situation has led to the imbalance of population distribution and economic development layout in the province at different degrees, and the issues related to ecological environment protection have also become prominent.

2.2. Methods

2.2.1. Construction of Production–Living–Ecological (PLE) Land Classification System

Territorial spatial classification uses the differences in land use types to integrate various elements in the whole area, and then coordinate the layout and utilization of various spatial resources [11,41]. Based on the obtained land type data (including six first-level types of cultivated land, woodland, grassland, water area, construction land, and unused land, and 25 second-level types, such as paddy field and dry land), from the perspective of PLE space, we analyzed the processes of land resources in terms of their quantity and space reallocation among the production, living, and ecological function. The dynamic economic and social development and transformation of the studied territorial space at each stage can be understood, using PLE space as a reference [42,43]. We considered the high ecological and environmental resolution of the secondary classification of land use, results of different global ecosystem services, measured by scholars from various countries, such as Costanza et al. [44], and the actual situation of ecosystem services in China (such as the distinction between paddy field and dry land). Then we employed the eco-environmental quality index obtained by Li et al. [45]. Meanwhile, because this index system is widely used in China and better conforms to the actual situation of ecological service function in China, we directly adopted this index as the background value of the eco-environmental quality index [46,47]. The area weighting method was used to assign the eco-environmental quality index values to various land categories in the PLE space. Finally, we calculated the eco-environmental quality index of land use types for the production, living and ecological functions (

Table 1. PLE land classification system and eco-environmental quality index of land use types in Qinghai province.

Territorial Land Space Classification Based on PLE Space		Corresponding Land Type		Eco-Environmental Quality Index (Land Use Types)
1st Level Classes	Subclasses	1st Level Classes	Background Value of Eco-Environmental Quality Index
PS	APS	Cultivated land	Dry cultivated land (0.25) paddy field (0.3)	0.250
	IPS	Urban and rural, industrial and mining, residential Land	Other construction land (0.15)	0.150
LS	ULS		Urban land (0.2)	0.200
	RLS		Rural residential land (0.2)	0.200
ES	GES	Grassland	High covered grassland (0.75) Medium coverage grassland (0.45) Low cover grassland (0.2)	0.334
	FES	Wood Land	Forest land (0.95) Shrub land (0.65) Sparse woodland (0.45) Other woodlands (0.4)	0.647
	WES	Water Area	Canal (0.55) Lake (0.75) Pond (0.55) Permanent glacier snow land (0.9) Tidal flat (0.45) Beach land (0.55)	0.659
	OES	Unused Land	Sand (0.01) Gobi (0.01) Saline alkali land (0.05) Swamp land (0.65) Bare land (0.05) Bare rock gravel (0.01)	0.056

PS: Production space; LS: Living space; ES: Ecological space; APS: Agricultural production space; IPS: Industrial and mining production space; ULS: Urban living space; RLS: Rural living space; FES: Forest ecological space; GES: Grass ecological space; WES: Water ecological space; OES: Other ecological space.

).

2.2.2. Territorial Spatial Transfer Matrix

The territorial space transfer matrix is an application of the Markov model commonly used to analyze land use change. In this method, according to the change relationship of land cover in different time and direction, two-dimensional matrix is used to analyze the specific situation of mutual transformation between different land use types, through quantitative data, e.g., the change of location and area and the initial and final land class transfer. Thus, the overall trend of land use change and the change of land use structure can be understood [48]. The mathematical formula of the transition matrix is as follows:

$${\mathrm{S}}_{\mathrm{ij}}=\left[\begin{array}{ccc}{\mathrm{S}}_{11}& {\mathrm{S}}_{12}& \cdots {\mathrm{S}}_{1\mathrm{n}}\\ {\mathrm{S}}_{21}& {\mathrm{S}}_{22}& \cdots {\mathrm{S}}_{2\mathrm{n}}\\ \cdots & \cdots & \cdots \\ {\mathrm{S}}_{\mathrm{n}1}& {\mathrm{S}}_{\mathrm{n}2}& \cdots {\mathrm{S}}_{\mathrm{nn}}\end{array}\right]$$

(1)

In Equation (1), S_ij is the total area of the territorial space of type i at the beginning of the study to type j at the end of the study. n is the number of land use types of territorial space utilization. The data of land use types in different periods were analyzed using ArcGIS 10.2 software, and the transfer matrix of the land types in each period was established.

2.2.3. Eco-Environmental Effect

• Unit eco-environmental quality index

The distribution law of territorial space is strongly dependent on the spatial scale, and the study of scale selection will greatly affect the conclusions obtained. To obtain the most appropriate scale, based on the results of Chen et al. [43,49], we adjusted the image of the study area. Finally, a 4 × 4 km scale was used to sample the study area, with equal spacing, and nearly 46,000 sample areas were generated. Comprehensively considering the proportion of the PLE space area in each ecological grid cell and the background value of the eco-environmental quality index, the eco-environmental quality status of each ecological grid cell in the study area was quantitatively expressed. The mathematical formula used for this analysis is shown below:

$${\mathrm{EV}}_{\mathrm{i}}={\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{N}}\raisebox{1ex}{${\mathrm{A}}_{\mathrm{ki}}$}\!\left/ \!\raisebox{-1ex}{${\mathrm{A}}_{\mathrm{k}}$}\right.{\mathrm{R}}_{\mathrm{i}}}$$

(2)

In Equation (2), EV_i is the eco-environmental quality index of i ecological units. R_i is the eco-environmental quality index of class i land use type. A_ki is the area of land use type i in the kth ecological unit. A_k is the area of the kth ecological unit. n is the number of land use types. Simultaneously, we applied the Kriging method to carry out spatial interpolation on the eco-environmental quality index of the study area, and it was divided into five levels (

Table 2. Eco-environmental quality index level.

Level	Low-Quality Area	Medium-Low-Quality Area	Medium-Quality Area	Medium-High-Quality Area	High-Quality Area
Value	EV ≤ 0.15	0.15 < EV ≤ 0.25	0.25 < EV ≤ 0.35	0.35 < EV ≤ 0.45	EV > 0.45

).

2.

Ecological contribution rate of land use function transformation

The ecological contribution rate of land use function transformation refers to a certain type of land use change resulting from the change in the regional ecological quality. By calculating the ecological contribution rate of land use transformation, we can demonstrate the main type of the land ues transformation, which causes the increase or decrease [50], which can be expressed as follows:

$$\mathrm{LEI}=\left({\mathrm{LE}}_{\mathrm{t}+1}{-\mathrm{LE}}_{\mathrm{t}}\right)\mathrm{LA}/\mathrm{TA}$$

(3)

In Equation (3), LEI is the ecological contribution rate of land use function transformation. LE_t+1 and LE_t are the ecological quality index of the land use types at the beginning and end of the change, respectively, reflected by a certain land use change type. LA is the area of the change type; TA is the total area of the region.

2.2.4. Geographical Detector

Geographical detectors are a statistical tool that is used to detect the spatial differentiation of geographical phenomena and explain their driving forces [51]. The term was proposed by Wang Jingfeng [52]. In this study, we used the factor and interaction detection modules of a geographic detector to identify the main driving forces that affect the regional eco-environmental quality, and at the same time, try to explore the driving mechanism that affects the eco-environmental quality of Qinghai Province.

2.3. Data Sources and Pre-Processing

2.3.1. Data Sources

In this study, we used the six periods land use data of Qinghai Province for the years 1980, 1990, 2000, 2010, 2015, and 2020 at 30 m × 30 m spatial resolution. The digital elevation model (DEM) image data at 30 m spatial resolution and the datasets of the spatial differences of the precipitation and mean temperature of Qinghai were obtained from the Data Centre for Resources and Environmental Sciences, Chinese Academy of Sciences (https://www.resdc.cn, accessed on 3 March 2022). We used the human–computer interactive visual interpretation method, which was based on the data from Landsat multispectral scanner system (MSS), thematic mapper/enhanced thematic mapper (TM/ETM), and Landsat8, which was used to interpret the data. The comprehensive interpretation accuracy of the method was not less than 90% and could thus meet the needs of this study. We used ArcGIS to extract the elevation, slope, and relief from the DEM data. The socio-economic data were collected from the Qinghai Statistical Yearbook and National Economic and Social Development Statistical Bulletin of each city, county, and district.

2.3.2. Selection of Indexes

The evolution of regional ecological environmental governance is determined by several factors. Natural conditions determine the basis of eco-environmental quality, but socio-economic factors are also important to change the regional eco-environmental quality [16,53]. To explore the driving forces and evolutionary mechanism of the eco-environmental quality in Qinghai Province, we selected 10 indicators on the basis of the natural conditions and socio-economic influences of the region (

Table 3. The serial number and name of each factor.

Number	X1	X2	X3	X4	X5
Name	Altitude	Slope direction	Relief amplitude	Annual average temperature	Annual average precipitation
Number	X6	X7	X8	X9	X10
Name	Year-end total population	Population density	GDP	Non-Agricultural proportion	Road network density

GDP: gross domestic product.

).

3. Results

3.1. Overall Characteristics of Changes in Territorial Spatial Pattern

3.1.1. Horizontal Regional Differentiation of Territorial Spatial Pattern

Qinghai Province has obvious regional differentiation in territorial space level, portraying a strong agglomeration of the production and living spaces and the dominance of ecological spaces. In 2020, the proportion of production, living, and ecological spaces in the total area of Qinghai Province was 1.29, 0.13, and 98.57%, respectively. The production and living spaces are mainly concentrated in the eastern part of Qinghai (Xining City, Haidong City, south-eastern part of the Haibei Tibetan autonomous prefecture, Hainan Tibetan autonomous prefecture, and south-eastern part of the Haixi Mongolian Tibetan autonomous prefecture) having the largest proportion of ecological spaces that are mainly distributed in central and western Qinghai.

From the perspective of the spatial distribution of second-class places, the urban living spaces (ULS) and rural living spaces (RLS) in Qinghai Province are small and concentrated, and their spatial distribution is similar to those of agricultural production space (APS) and industrial production space (IPS). Grassland ecology space (GES) and other ecology space (OES) were the most widely distributed spaces. Notably, the GES were mainly distributed in the Haidong region and Qinghai-Tibet Plateau. The OES were mainly distributed in the Qaidam basin. The forestland ecological space (FES) is mainly distributed in the eastern margin of the Qinghai and Kunlun Mountains. The water ecological space (WES) is distributed in the whole region, but more concentrated in the western Qinghai-Tibet Plateau (upstream of the Three-River Headwaters region).

3.1.2. Difference in Vertical Gradient of Territorial Spatial Pattern

The vertical gradient differentiation of territorial space in Qinghai Province was obvious (Figure 2). The widest and narrowest areas of the territorial space were located in areas having altitudes of 4500–5000 m and >5000 m, respectively. The area of the production and living spaces was inversely proportional to the altitude. There was no distribution of the production and living spaces above the altitude of 5000 m. The maximum area of the production and living spaces was at altitudes above 3000 m (5670.01 km² and 689.23 km², respectively). The ecological spaces fluctuated with the increase in the altitude; the maximum area of ecological spaces (205,502.94 km²) occurred at the altitude of 4500–5000 m above sea level.

In terms of the spatial distribution of second-level land classes, the land areas of the APS, IPS, ULS, and RLS portrayed a decreasing trend with increasing altitude. Notably, the maximum numbers of APS, IPS, ULS, and RLS all appeared in the areas having altitude less than 3000 m. The areas of GES and OES portrayed a trend of initial decrease, followed by an increase. This was followed by a decrease with increasing elevation; notably, the maximum numbers of GES and OES appeared at altitudes of 4500–5000 m and <3000 m, respectively. The FES portrayed an initial increasing trend with increasing elevation, followed by a decreasing trend; the maximum number appeared at an elevation of 3500–4000 m above the sea level. The WES area portrayed a trend of fluctuation; the maximum number of WES occurred at an altitude of 4500–5000 m.

3.2. Characteristics of Changes in Spatio-Temporal Patterns of Territorial Space

From 1980 to 2020, the territorial space of Qinghai Province in the horizontal region portrayed the pattern characteristics of increasing production and living spaces and shrinking ecological spaces (

Table 4. Structural changes of production-living-ecological space types in Qinghai Province from 1980–2020 (km²/%).

Space Type	Secondary Space Type	1980		1990		2000		2010		2015		2020
		Area	Scale	Area	Scale	Area	Scale	Area	Scale	Area	Scale	Area	Scale
PS	APS	7849.59	1.12	7982.88	1.14	8245.08	1.18	8571.28	1.23	8523.68	1.22	8604.42	1.24
	IPS	201.59	0.03	265.75	0.04	308.19	0.04	762.41	0.11	1065.99	0.15	657.80	0.09
LS	ULS	106.39	0.02	111.64	0.02	129.43	0.02	178.79	0.03	193.70	0.03	262.41	0.04
	RLS	543.77	0.08	544.59	0.08	569.82	0.08	614.85	0.09	681.54	0.10	685.65	0.10
ES	GES	373,144.71	53.56	372,763.25	53.51	372,329.20	53.44	392,372.13	56.32	392,205.73	56.30	391,414.40	56.18
	FES	28,718.82	4.12	28,724.06	4.12	28,709.43	4.12	28,563.01	4.10	28,585.00	4.10	28,540.55	4.10
	WES	27,436.21	3.94	27,434.20	3.94	27,292.10	3.92	29,749.50	4.27	30,055.67	4.32	32,700.94	4.69
	OES	258,665.66	37.13	258,840.37	37.15	259,083.43	37.20	235,855.88	33.85	235,356.54	33.78	233,801.68	33.56
Space Type	Secondary Space Type	1980–1990		1990–2000		2000–2010		2010–2015		2015–2020		1980–2020
		Area Change	Change Rate	Area Change	Change Rate	Area Change	Change Rate	Area Change	Change Rate	Area Change	Change Rate	Area Change	Change Rate
PS	APS	133.29	0.17	262.20	0.33	326.20	0.40	−47.60	−0.11	80.74	−0.19	754.83	0.24
	IPS	64.16	3.18	42.44	1.60	454.22	14.74	303.58	7.96	−408.19	7.66	456.21	5.66
LS	ULS	5.25	0.49	17.79	1.59	49.36	3.81	14.91	1.67	68.71	−7.09	156.02	3.67
	RLS	0.82	0.02	25.23	0.46	45.03	0.79	66.69	2.17	4.11	−0.12	141.88	0.65
ES	GES	−381.46	−0.01	−434.05	−0.01	20,042.93	0.54	−166.40	−0.01	−791.33	0.04	18,269.69	0.12
	FES	5.24	0.002	−14.63	−0.01	−146.42	−0.05	21.99	0.02	−44.45	0.03	−178.27	−0.02
	WES	−2.01	−0.001	−142.10	−0.05	2457.40	0.90	306.17	0.21	2645.27	−1.76	5264.73	0.48
	OES	174.71	0.01	243.06	0.01	−23,227.55	−0.90	−499.34	−0.04	−1554.86	0.13	−24,863.98	−0.24

). The area of production spaces increased from 8051.18 km² in 1980 to 9262.22 km² in 2020. The area of living spaces increased from 650.16 km² in 1980 to 948.06 km² in 2020. However, the area of ecological land decreased from 687,965.40 km² in 1980 to 686,457.57 km² in 2020.

From the perspective of the change degree of the secondary spatial structure (Figure 3), the proportion of the secondary spatial area in Qinghai Province in 2020 (from large to small) was: GES > OES > WES > FES > APS > RLS > IPS > ULS. In addition to the decrease in the FES and OES ratio (the areas decreased by 178.27 and 24863.98 km², respectively), the APS, IPS, ULS, RLS, GES, and WES increased significantly (754.83, 456.21, 156.02, 141.88, 18269.69, and 5264.73 km², respectively). The ULS and RLS areas portrayed an increasing trend; the GES and WES portrayed an initially increasing trend followed by a decreasing trend. The APS area portrayed a trend of fluctuation, IPS and OES portrayed an initial increasing followed by a decreasing trend, and FES portrayed a trend of a small drop.

Based on the change rate of the secondary spatial structure in Qinghai Province, we could deduce that, during 1980–2020, the growth rates of the areas of the IPS, ULS, and RLS were 5.66, 3.67, and 0.65%, respectively, while those of the FES and OES portrayed a decreasing trend (−0.02 and −0.24%, respectively).

3.3. Spatial-Temporal Transformation Characteristics of Land Use Types in Territorial Space in Qinghai

During 1980–1990, the area transformed from GES was the largest (453.49 km²), accounting for 39.62% of the total area transformed (Figure 4a), among which 42.32% was converted into APS, 38.39% into OES, and 19.12% into other spaces. The total area of all the types of spaces converted to GES was only 72.03 km², and thus the total area of GES decreased.

From 1990 to 2000, the number of GES areas transferred to other spaces was the largest (which was 897.11 km²), accounting for 49.46% of the total converted area (Figure 4b), among which 40.46% was converted to OES, 31.85% to APS, and 27.68% to other spaces. However, the area of various spaces converted to GES was only 463.08 km², and thus, the total area of GES continued to decline.

From 2000 to 2010, the area of OES converted to other spaces was the largest (28,669.66 km²), accounting for 79.09% of the total amount (Figure 4c), among which 90.57% was converted into GES, 8.14% into WES, and 1.29% into other spaces. The area of all the types of space converted to OES was only 5441.88 km². Notably, compared with the previous 20 years, the total area of the OES decreased, whereas that of GES increased.

During 2010–2015, the area of OES converted to other spaces was 1585.31 km², accounting for 43.20% of the total amount (Figure 4d), among which 44.28% of OES was converted to WES, 38.37% to GES, and 16.85% to other spaces. The area of all types of space was converted to 1085.79 km², and the total area of OES continued to decrease.

From 2015 to 2020, the area of OES converted to other spaces was 3407.57 km², accounting for 41.19% of the total amount (Figure 4e), of which 62.58% was converted to WES, 31.28% to GES, and 6.15% to other spaces. The total area converted to OES was only 1866.97 km²; notably, the total area of OES continued to decrease during this period.

3.4. Comprehensive Quality of the Ecological Environment

3.4.1. Spatial-Temporal Evolution Characteristics of Ecological Environment Comprehensive Quality

The overall eco-environmental quality index of Qinghai province for 1980, 1990, 2000, 2010, 2015, and 2020 was 0.2557, 0.2562, 0.2561, 0.2653, 0.2655, and 0.2667, respectively. Except for a slight decrease in 2000, the overall ecological environment portrayed a significant improvement. Additionally, there were significant differences in the ecological and environmental quality grades (Figure 5). The area of high-quality regions continued to increase, whereas that of medium high-quality regions portrayed an initial decrease followed by an increase. The area of high-quality regions was the smallest in each period, accounting for less than 20% of the total area of the study area. Notably, the changes in the medium-quality and low-quality areas portrayed a wave-state potential. The area of medium low-quality regions portrayed an initial increase followed by a decrease, and the area of medium–low- and low-quality regions exceeded 55% of the total area, constituting the main body of eco-environmental quality (

Table 5. Distribution of ecological environment quality grades of Qinghai (km²/%).

Level	1980		1990		2000
	Area	Scale	Area	Scale	Area	Scale
<0.15	165,996	23.83	166,254.66	23.86	166,456.17	23.89
0.15–0.25	15,4681.38	22.20	154,807.65	22.22	155,386.63	22.30
0.25–0.35	255,745.53	36.71	255,276.09	36.64	254,679.48	36.56
0.35–0.45	80,180.28	11.51	80,118.64	11.50	79,888.23	11.47
>0.45	40,067.91	5.75	40,214.16	5.77	40,260.96	5.78
Level	2010		2015		2020
	Area	Scale	Area	Scale	Area	Scale
<0.15	156,510.63	22.47	156,629.88	22.48	153,724.32	22.07
0.15–0.25	133,162.01	19.11	132,982.11	19.09	132,836.13	19.07
0.25–0.35	276,066.81	39.63	275,342.31	39.52	275,437.8	39.54
0.35–0.45	88,509.33	12.70	88,748.46	12.74	89,530.61	12.85
>0.45	42,422.76	6.09	42,969.06	6.17	45,142.75	6.48

). As shown in Figure 5, the high-quality and the medium–high-quality areas were mainly distributed in the east and northwest of Qinghai. The medium-quality regions were distributed in the south and east of Qinghai and gradually expanded to the north; the low and medium–low-quality areas were distributed in most parts of the north and central Qinghai, but portrayed a decreasing trend.

3.4.2. Main Land Use Conversion Affecting Eco-Environmental Quality

We observed two types of ecological quality trends, namely, improvement and deterioration, which offset each other, ensuring stability. From 1980 to 2020, the trend of ecological environment improvement in Qinghai Province was much higher than that of ecological environment deterioration; notably, the degree of ecological environment improvement continued to increase. As shown in

Table 6. Major land use transformations influencing ecological environment quality and contribution rates.

1980–2020
Change Trend	Land Use Function Transformation	Index Movement	Contribution Proportion (%)
Improvement of Eco-environment	OES-GES	0.010699	67.13
	OES-WES	0.003989	25.03
	GES-WES	0.000724	4.54
	GES-FES	0.000260	1.63
	APS-WES	0.000074	0.47
	OES-FES	0.000054	0.34
	IPS-WES	0.000032	0.20
Deterioration of Eco-environment	GES-OES	−0.002535	62.87
	WES-OES	−0.000635	15.75
	FES-GES	−0.000304	7.54
	WES-GES	−0.000183	4.54
	GES-APS	−0.000140	3.47
	GES-IPS	−0.000073	1.82
	FES-OES	−0.000063	1.56

, the conversion of OES into GES and WES and that of GES into WES and FES were the main factors for environment improvement. The conversion of GES into OES, APS, and IPS, and that of WES into OES and GES, were the main factors for environmental deterioration. The land function types that led to the improvement of the ecological environment were relatively concentrated, and the first seven land function transformations that contributed to the improvement/deterioration of ecological quality accounted for 99.34% and 97.56%, respectively.

3.5. Driving Force Analysis of Eco-Environmental Quality

The results indicated that the eco-environmental quality of Qinghai province was jointly affected by multiple factors, and different influencing factors had varied effects on the eco-environmental quality of the region (Figure 6). All factors passed the significance test at the 0.05 level, and the factor contribution rate q value was used to measure the impact degree of each factor on the spatial differentiation of the eco-environmental quality of the region (q ≥ 0.100 was the factor, with a great impact on the eco-environmental quality of the study area). From 1980 to 2020, X5 (0.294), X4 (0.074) and X1 (0.061) contributed more to the natural factors, and X9 (0.223), X10 (0.199), and X8 (0.195) contributed more to the socio-economic factors. In general, socio-economic factors had a greater impact on the quality of the ecological environment.

3.5.1. Analysis of Natural Factors

From the perspective of factor interpretation, we analyzed the trends for the following factors: X1 (altitude), X2 (slope direction), X3 (relief amplitude), X4 (annual average temperature) and X5 (annual average precipitation). X1, X2, and X3 had little influence on the eco-environmental quality, and the changes were relatively stable. However, the q values of X1 and X3 in 1980 decreased slightly compared with those in 2020, indicating that the influence of X1 and X3 on the eco-environmental quality of the study area weakened. Compared with 1980, the advantage of q value of X2 in 2020 increased, indicating that the influence of X2 on the eco-environmental quality of the study area gradually enhanced, but its overall influence was less, compared to that of other factors. The influence of the q value of X4 on the ecological and environmental quality fluctuated in different years, but it had a greater impact on the ecological and environmental quality in 2020 (up to 0.1), indicating that the influence on ecological and environmental quality in 2020 strengthened. Finally, X5 had a great impact on the ecological and environmental quality. Although we observed a trend of fluctuation, the q value in each year was greater than 0.25, which indicated that X5 was the main driving force of the ecological and environmental quality. The q value increased from 0.31 to 0.34 from 1980 to 2020, indicating that the influence of X5 on the ecological and environmental quality of the study area increased significantly.

3.5.2. Analysis of Socio-Economic Factors

Additionally, we analyzed the following factors: X6 (year-end total population), X7 (population density), X8 (gross domestic product, GDP), X9 (non-agricultural proportion), and X10 (road network density). From the perspective of factor interpretation, X6–X10 had a great impact on the eco-environmental quality of the study area, which was the main driving force. The impact of X6 and X7 on the ecological environment of the study area fluctuated, but the overall level remained above 0.100, indicating that X6 and X7 had a great impact on the ecological environment of the study area. However, the q value of X7 decreased to 0.089 in 2020, indicating that the influence of X7 on the ecological and environmental quality weakened in 2020. Notably, X8, X9, and X10 portrayed an initial increasing trend, followed by a decreasing trend from 1980 to 2020; however, the values increased to different degrees, compared with 1980, indicating that the influence of X8–X10 on the eco-environmental quality of the study area was increasing.

3.5.3. Human–Natural Coupling Interaction Detection Results

Different factors have different effects on ecological and environmental quality; notably, there are complex interaction relationships among these factors, leading to differences in the magnitude, intensity, and direction of their effects. The interaction between factors may increase the impact on the ecological and environmental quality. From 1980 to 2020, the interaction between the natural and human factors in Qinghai portrayed two modes of non-linear and double factor enhancements; notably, there was no independent or weakening relationship, indicating that the influence of the interaction between the two factors was greater than the influence of each single factor. According to the results of factor detection and interaction detection, X5 and X1 (among natural factors) and X8 and X10 (among socio-economic factors) were the factors that portrayed the greatest influence on human-natural coupling interaction and factor detection (

Table 7. The interaction of natural and human factors on the driving force of eco-environmental quality evolution in Qinghai Province.

Year	1980					1990					2000
Interactive Items and Interaction Value	X1∩X6	X1∩X7	X1∩X8	X1∩X9	X1∩X10	X1∩X6	X1∩X7	X1∩X8	X1∩X9	X1∩X10	X1∩X6	X1∩X7	X1∩X8	X1∩X9	X1∩X10
	0.215	0.304	0.221	0.314	0.301	0.243	0.332	0.225	0.318	0.300	0.214	0.288	0.309	0.311	0.297
	X2∩X6	X2∩X7	X2∩X8	X2∩X9	X2∩X10	X2∩X6	X2∩X7	X2∩X8	X2∩X9	X2∩X10	X2∩X6	X2∩X7	X2∩X8	X2∩X9	X2∩X10
	0.132	0.172	0.130	0.205	0.173	0.187	0.216	0.135	0.205	0.173	0.132	0.171	0.193	0.211	0.172
	X3∩X6	X3∩X7	X3∩X8	X3∩X9	X3∩X10	X3∩X6	X3∩X7	X3∩X8	X3∩X9	X3∩X10	X3∩X6	X3∩X7	X3∩X8	X3∩X9	X3∩X10
	0.153	0.192	0.161	0.234	0.192	0.217	0.249	0.165	0.235	0.192	0.155	0.190	0.221	0.236	0.190
	X4∩X6	X4∩X7	X4∩X8	X4∩X9	X4∩X10	X4∩X6	X4∩X7	X4∩X8	X4∩X9	X4∩X10	X4∩X6	X4∩X7	X4∩X8	X4∩X9	X4∩X10
	0.212	0.268	0.230	0.331	0.268	0.249	0.349	0.233	0.329	0.265	0.202	0.255	0.317	0.326	0.259
	X5∩X6	X5∩X7	X5∩X8	X5∩X9	X5∩X10	X5∩X6	X5∩X7	X5∩X8	X5∩X9	X5∩X10	X5∩X6	X5∩X7	X5∩X8	X5∩X9	X5∩X10
	0.363	0.389	0.349	0.364	0.386	0.377	0.351	0.348	0.355	0.361	0.347	0.361	0.350	0.358	0.362
Year	2010					2015					2020
Interactive Items and Interaction Value	X1∩X6	X1∩X7	X1∩X8	X1∩X9	X1∩X10	X1∩X6	X1∩X7	X1∩X8	X1∩X9	X1∩X10	X1∩X6	X1∩X7	X1∩X8	X1∩X9	X1∩X10
	0.286	0.286	0.360	0.311	0.328	0.310	0.302	0.345	0.337	0.344	0.219	0.245	0.290	0.315	0.333
	X2∩X6	X2∩X7	X2∩X8	X2∩X9	X2∩X10	X2∩X6	X2∩X7	X2∩X8	X2∩X9	X2∩X10	X2∩X6	X2∩X7	X2∩X8	X2∩X9	X2∩X10
	0.166	0.166	0.304	0.223	0.240	0.173	0.174	0.235	0.289	0.231	0.127	0.093	0.195	0.225	0.223
	X3∩X6	X3∩X7	X3∩X8	X3∩X9	X3∩X10	X3∩X6	X3∩X7	X3∩X8	X3∩X9	X3∩X10	X3∩X6	X3∩X7	X3∩X8	X3∩X9	X3∩X10
	0.180	0.181	0.323	0.240	0.261	0.191	0.194	0.255	0.315	0.252	0.146	0.118	0.212	0.244	0.243
	X4∩X6	X4∩X7	X4∩X8	X4∩X9	X4∩X10	X4∩X6	X4∩X7	X4∩X8	X4∩X9	X4∩X10	X4∩X6	X4∩X7	X4∩X8	X4∩X9	X4∩X10
	0.255	0.256	0.384	0.328	0.356	0.275	0.272	0.346	0.360	0.335	0.221	0.227	0.295	0.326	0.339
	X5∩X6	X5∩X7	X5∩X8	X5∩X9	X5∩X10	X5∩X6	X5∩X7	X5∩X8	X5∩X9	X5∩X10	X5∩X6	X5∩X7	X5∩X8	X5∩X9	X5∩X10
	0.323	0.324	0.354	0.322	0.335	0.302	0.296	0.327	0.388	0.332	0.360	0.359	0.357	0.367	0.358

).

4. Discussion

Territorial space is an important carrier of regional human activities and ecological environments. The interaction between people and the natural environment in territorial space changes the function of regional territorial spaces and shapes the production, activity, and ecological spaces through the changes in the land use types in a region [32,54]. In 1999, the European Spatial Development Perspective (ESDP) clearly stated that spatial planning can promote sustainable and balanced development among regions [55]. As an effective means for the construction of ecological civilization and spatial planning, PLE space is classified on the basis of different utilization functions of territorial space to optimize the development pattern, control the development intensity, and adjust the spatial structure of territorial space [39,56]. In this study, we used the PLE functional space classification to merge and classify the land use spatial data, which accounted for the lack of the consideration of the ecological function in land use classification, and realized the connection between land function and land use classifications. Therefore, this method is widely used in land function regulation, determining the eco–environment effects, and other related fields [57,58,59].

According to relevant studies, the rapid development of social economy and urbanization is accompanied by the deterioration of eco-environmental quality to a certain extent [25,60,61]. This degradation is usually caused by the change of territorial spatial patterns due to land use transformation (Figure 7). Qinghai Province is located inland of north-west China, and most of its areas are restricted development zones (areas with weak resource and environment carrying capacities, with poor conditions for large-scale ag-glomeration of economy and population, but related to ecological security in a large area of the country) [62,63]. The causes of eco-environmental quality changes in Qinghai Province are consistent with those in regions and cities with rapid economic development and high urbanization rate in China, such as Yellow River Delta, Beijing-Tianjin-Hebei region, Yangtze River delta economic zone, etc. It is mainly due to economic expansion and urban occupation of ecological space that APS and ecological land are converted into ULS and RLS in territorial space. However, because most of the area in Qinghai is a restricted development zone, it also has a certain comprehensiveness and complexity, for example, unfavourable ecological space maintenance leads to grassland and forestland degradation and wetland atrophy.

The change of ecological environment quality reflects the interaction between natural environment and human society in territorial space, and its change is complex and dynamic. Such changes are caused by the natural constraints force provided by natural factors, the human driving force provided by socioeconomic factors, and the coupling interaction force between humans and nature. The effect of natural factors on the change of eco-environment quality is smaller than that of socioeconomic factors, but it creates the basic conditions of ecological environment quality. Socioeconomic factors have a more direct impact on eco-environmental quality and play a leading role in the change of it. The human-nature coupling interaction force has a strengthening effect on the eco-environment quality, which is often accompanied by a guiding and decision-making power. Guiding and decision-making power refer to the influence of local political environment [25]. The direction and speed of the evolution of territorial spatial pattern and eco-environment quality are determined by the joint participation of these forces.

4.1. Natural Factors Are the Natural Constraints Force Affecting the Eco-Environment Quality

Due to the characteristics of natural environments, the territorial space of Qinghai Province portrayed significant horizontal regional differentiation and vertical gradient differences. Simultaneously, due to the characteristics of small population density and the relatively concentrated population, the territorial spatial pattern indicated the characteristics of relative agglomeration of the production and living spaces and absolute dominance of ecological spaces. The production and living spaces portrayed significant convergences and were mainly distributed in Qinghai city (prefecture) in the county administrative centre. The eco-environmental quality of Qinghai Province portrayed an overall steady increase, but due to the fragile ecological environment, although the high-quality areas expanded, the overall area remained small. The low-quality area and the low-medium quality area continued to shrink, but the proportion of the areas were still large, indicating that the eco-environmental quality needed to be improved further.

4.2. Socioeconomic Factors Are the Human Driving Forces Affecting Eco-Environment Quality

From 1980 to 2020, the living space continued to expand due to urbanization and population growth, and the growth range and speed of the ULS was much greater than that of RLS in Qinghai. Both the production and ecological spaces portrayed fluctuations. The former space increased, while the latter slightly decreased. The eco-environmental quality of Qinghai Province portrayed an overall steady increase, albeit a slight decrease in 2000; this was partly because the eco-logical civilization concept was in its infancy stage in China and because in 1992, China officially proposed to establish the goal of the socialist market economy, which resulted from the people expecting high economic benefits, with little consideration for the protection of the ecological environment.

4.3. Human-Nature Coupling Interaction Force Are the Crucial Guiding Forces Affecting Eco-Environment Quality

At the national level, in terms of the vast spatial areas of Qinghai Province, the topography is generally complex; notably, the region is also one of the multi-ethnic populated provinces in mainland China and has a unique natural geographic and socio-economic structure. Relevant national policies, such as Western development, returning farmland to forest (grass-land), ecological civilization construction and high-quality development of the Yellow River Basin, all have a significant impact on socioeconomic factors by combining the characteristics of regional natural environment and form a crucial guiding force for the change of eco-environment quality. The increase in the production space from 1980 to 2000 was mainly caused by crowding out the GES, which led to the continuous decline of GES during this period. At the beginning of the 21st century, China put forward the construction of ecological civilization, and with the vigorous promotion of ecological civilization construction, the ecological spaces of grassland, woodland, and water area improved to different degrees, through the conversion of other ecological spaces (unused land). Qinghai Province is located in the upper reaches of the Yellow River, and its location has important political, ecological, economic and social significance. The implementation of relevant regional policies makes Qinghai’s overall environmental quality portray a continuous upward trend. Additionally, we observed non-linear enhancement and double enhancement effects among the factors, indicating that human-nature coupling interaction force are the crucial guiding forces affecting eco-environment quality.

5. Conclusions

Based on the “PLE” spatial classification, we employed the land transfer matrix, eco-environmental quality index, and ecological contribution rate of land use transformation to quantitatively analyze the changes of territorial spatial pattern and eco-environmental effects in Qinghai Province, and used geographic detectors to explain the driving forces of eco-environmental quality evolution. (1) the spatial distribution of the province was characterized by the relative agglomeration of the production and living spaces and the absolute dominance of ecological spaces. It shows that there is a trend of expansion of production and living space and contraction of ecological space. (2) The eco-environmental quality of the region portrayed a steady improvement, with a significant reduction in the medium–low and low-quality areas. Spatially, the medium quality areas are mainly distributed in most of Haixi Prefecture, while the high-quality and medium-high-quality Area areas are mainly distributed in the eastern part of Qinghai and the southern part of Three Rivers Source region. (3) The annual average precipitation, proportion of non-agricultural area, and socio-economic factors had a significant impact on the eco-environmental quality of the region; meanwhile, national economy and ecological policies are important indirect driving forces of eco-environmental quality. Although the influence of natural factors on the eco-environmental quality of Qinghai Province is less than that of human factors, the support and constraint of natural geographical basis on the ecological environment cannot be ignored. Additionally, we observed non-linear enhancement and double enhancement effects among the factors, indicating that the human-nature coupling interaction force had a strengthening effect on the changes in the eco-environmental quality.

In the future development of Qinghai province, the key to the continuous improvement of ecological environment quality in the area are to optimize the layout of the PLE spaces, construct a reasonable territorial space protection pattern, and promote the sustainable development of human-natural coupling.