Comprehensive Evaluation of Soil Quality Reconstruction in Agroforestry Ecosystems of High-Altitude Areas: A Case Study of the Jiangcang Mining Area, Qinghai–Tibet Plateau

Abstract

This study focuses on the alpine meadow ecosystem of the Qinghai–Tibet Plateau, which plays a vital role in carbon sequestration and water resource protection. However, mining activities have severely damaged the ecosystem, posing challenges for ecological restoration. The study selected the Jiangcang mining area and analyzed the physical, chemical, and carbon characteristics and heavy metal content of soil samples from the slag platforms and slopes (0–20 cm), which were restored in 2015 and 2020 to explore the effects of different soil reconstruction methods on soil function and ecological resilience. The results show that the minimum data set (MDS) can effectively replace the total data set (TDS) in assessing soil quality. The assessment indicates good restoration effects in 2020, with some areas rated high in soil quality. Although issues such as high bulk density, high electrical conductivity, low moisture content, nitrogen deficiency, and low organic matter limit ecological restoration, the carbon sequestration capacity of the restored soil is strong. This study provides scientific evidence for ecological restoration in cold mining areas, indicating that capping measures can enhance soil resistance to erosion, nutrient retention, and carbon sink functions.

1. Introduction

The alpine meadow ecosystems of the Qinghai–Tibet Plateau, serving as the core component of Earth’s “Third Pole,” play a pivotal role in the global ecosystem [1]. These ecosystems provide essential services, including soil conservation, biodiversity preservation [2], regional climate modulation, and carbon cycle regulation [3,4]. They not only store approximately 2.5% of the world’s soil organic carbon [5] but also form the headwaters of several major rivers in Asia. However, recent human activities, particularly mining operations, have posed severe threats to these ecosystems, with their impact on carbon loss significantly exceeding that of other anthropogenic disturbances [6,7]. The Muli mining area (37°20′–38°20′ N), the focal site of this study, exemplifies these ecological challenges.

The extreme environmental conditions—including low temperatures, intense radiation, hypoxia, thin and nutrient-poor soils, and short growing seasons [8]—create a unique “ecological bottleneck” effect. The natural formation of a 30 cm soil layer requires 3000–8000 years, while vegetation recovery proceeds even more slowly, with alpine meadow formation taking millennia [9,10]. Open-pit mining activities, through topsoil removal and microtopographic alteration, have led to a 40–60% reduction in surface soil carbon storage. This degradation not only severely threatens local herders’ livelihoods but also disrupts critical ecological processes in regional carbon–water coupling [11].

While significant progress has been made in ecological restoration research for mid- to low-altitude regions, systematic studies on soil quality reconstruction in high-altitude mining areas remain notably insufficient. Given soil’s crucial role as both a major carbon reservoir and a regulator of hydrological and nutrient cycles [12,13], a deeper understanding of reconstructed soil quality dynamics is of paramount scientific importance for optimizing restoration measures, enhancing ecosystem resilience, and regulating carbon–water coupling processes [14,15,16].

Therefore, this study selects the Muli Jiangcang mining area on the Qinghai–Tibet Plateau as its research site (Figure 1). Through the comprehensive analysis of reconstructed soil environmental quality, we elucidate the impacts of different soil reconstruction methods on soil functionality and ecological resilience. The findings will provide critical guidance for optimizing soil cover and vegetation restoration strategies in future ecological rehabilitation efforts, thereby enhancing ecosystem services such as carbon sequestration and water regulation in cold high-altitude mining regions. Furthermore, these discoveries offer valuable data support for understanding carbon–water coupling processes and promoting the sustainable management of fragile alpine ecosystems.

2. Materials and Methods

2.1. Study Area

The Muri mining area is located on the southern foothills of the Qilian Mountains, representing a typical alpine meadow ecosystem of the Qinghai–Tibet Plateau and serving as the largest coal mining region in Qinghai Province, with proven coal reserves of 3.339 billion tons. The mining area has an average altitude of 4000 m, characterized by a cold high-altitude environment. The soils are primarily alpine meadow soil and swamp meadow soil, featuring moderate pH levels and a shallow soil layer (average thickness of about 20 cm).

Due to the mountainous environment, the vegetation exhibits a vertical zonal distribution with relatively simple community structure. While trees are sparse, herbaceous plants thrive, resulting in an overall vegetation coverage of 70–90%. The main vegetation types include alpine swamp vegetation and alpine meadow vegetation. The former is dominated by Kobresia tibetica and Carex orbicularis, while the latter consists mainly of Kobresia pygmaea, Kobresia capillifolia, Kobresia humilis, and Puccinellia macranthera.

Since the 1970s, coal mining activities in the Juhugeng and Jiangcang areas have created 11 open-pit mines and 19 waste dumps, severely degrading the original Kobresia meadow ecosystem. To restore the ecosystem, Qinghai Province implemented different rehabilitation measures in 2015 and 2020. In 2015, a basic approach involving waste rock crushing, simple land reshaping, and direct seeding was adopted, but vegetation recovery was limited, with coverage remaining at around 50%. In 2020, an improved comprehensive process was introduced, including in situ tillage, waste rock crushing and screening, aggregate collection, organic fertilizer application, and shallow tillage for soil reconstruction. These measures significantly enhanced the restoration outcomes, increasing vegetation coverage from 54.37% to 72.75%.

2.2. Soil Sample Collection and Analysis

2.2.1. Sample Collection

This study focuses on the Jiangcang mining area. In August 2024, sampling points were set up on slopes and platforms at wells 2, 4, and 5 and at natural grasslands. Among these, well 2 was the slag restoration site in 2015, while wells 4 and 5 were restored in 2020. At each sampling point, soil samples were collected from a depth of 0–20 cm, with 1 kg of soil taken using the quartering method. A total of 10 soil samples were collected. The soil samples were brought back to the laboratory, where they were naturally dried, and debris such as tree branches, grass roots, and stones were removed. The samples were then crushed and sieved for subsequent analysis. The specific sampling points are shown in Figure 2 and

Table 1. Sampling point locations.

Serial No.	Sample No.	Sampling Location	Ecological Restoration Time
1	24ML-01	South slope of Slag Heap at No. 5, Jiangcang Mining Area	2020
2	24ML-02	South platform of Slag Heap at No. 5, Jiangcang Mining Area	2020
3	24ML-03	North platform of Slag Heap at No. 5, Jiangcang Mining Area	2020
4	24ML-04	North slope of Slag Heap at No. 5, Jiangcang Mining Area	2020
5	24ML-05	North platform of Slag Heap at No. 4, Jiangcang Mining Area	2020
6	24ML-06	North slope of Slag Heap at No. 4, Jiangcang Mining Area	2020
7	24ML-07	South slope of Slag Heap at No. 4, Jiangcang Mining Area	2020
8	24ML-08	Platform of No. 2, Jiangcang Mining Area	2015
9	24ML-09	Slope of Waste Dump at No. 2, Jiangcang Mining Area	2015
10	24ML-10	Natural Grassland, Jiangcang Mining Area	Natural Grassland

.

2.2.2. Test Indicators and Methods

In this study, indicators that significantly affect vegetation growth were selected, as their interactions and synergistic effects can comprehensively reflect soil productivity and adaptability to environmental stress. These indicators include pH, soil bulk density, soil texture, soil moisture content, electrical conductivity, total nitrogen, total phosphorus, total potassium, ammonium nitrogen, nitrate nitrogen, available phosphorus, available potassium, organic matter, organic carbon, microbial carbon, easily oxidizable carbon, and heavy metals (Cd, Pb, Cu, Cr, Hg, and As) as the physical, chemical, and carbon characteristics and heavy metal indicators for the soil quality evaluations. Specific measurement methods are listed in

Table 2. Soil sample measurement indicators and methods.

Index Types	Measurement Indicator	Measurement Method
Physical index	Bulk Density	Cutting-ring method
	Soil Texture	Stokes’ law—pipette method
Chemical index	pH	Titration method (water-to-soil ratio 5:1)
	Moisture Content	Oven drying method
	Electrical Conductivity
	Total Nitrogen	Kjeldahl digestion, AA3 type continuous flow analyzer measurement
	Total Phosphorus	NaOH fusion, molybdenum–antimony anti-colorimetry
	Total Potassium	NaOH fusion, flame photometry method
	Ammonium Nitrogen	KCl extraction, AA3 type continuous flow analyzer measurement
	Nitrate Nitrogen	KCl extraction, AA3 type continuous flow analyzer measurement
	Available Phosphorus	NaHCO3 extraction, molybdenum–antimony anti-colorimetry
	Available Potassium	NH4OAc extraction, flame photometry method
Carbon characteristic index	Organic Matter	Potassium dichromate volumetric method (with heating)
	Organic Carbon	Potassium dichromate volumetric method (with heating)
	Microbial Carbon	Chloroform fumigation extraction method
	Easily Oxidizable Organic Carbon	Potassium permanganate oxidation—colorimetric method
Heavy metal index	Cd, Pb, Cu, Cr, Hg, As	Inductively coupled plasma mass spectrometry (ICP-MS)

[17,18].

2.3. Soil Quality Evaluation

2.3.1. Determination of Membership Function

Since different soil indicators have varying impacts on soil quality and the measurement units of each indicator differ, it is necessary to standardize the raw data before calculating the composite soil index to eliminate the influence of variable dimensions and ensure the objectivity and scientific accuracy of the results. Additionally, the variation in soil factors is continuous, so a continuous membership function is used to standardize the evaluation indicators for each soil factor. The soil membership function can be divided into S-shaped and parabolic types [19]. The S-shaped function includes both ascending and descending types. The soil pH and bulk density use the parabolic function to calculate the membership value, while heavy metals such as Cd, Pb, Cu, Cr, Hg, and As use the “S” descending function to calculate the membership value. Other soil indicators are evaluated using the “S” ascending function. The turning points used for the function types are taken from the second national soil survey grading standards [20]. The final membership function types and turning points for the study indicators are shown in

Table 3. Soil indicator membership function type and turning points.

Turning Point	Function Type	x1	x2	x3	x4
pH	Parabolic	4.5	6	7.5	8.5
Bulk Density (Mg/cm3)		1	1.1	1.2	1.4
Organic Matter (g/kg)	S-shaped	6	40
Total Nitrogen (g/kg)		0.5	2
Ammonium Nitrogen (mg/kg)		10	100
Nitrate Nitrogen (mg/kg)		3.5	35
Total Phosphorus (g/kg)		0.2	1
Available Phosphorus (mg/kg)		3	40
Total Potassium (g/kg)		6	30
Available Potassium (mg/kg)		30	200
Microbial Biomass Carbon (mg C/kg)		386	1611.6
Easily Oxidizable Organic Carbon (g C/kg)		5.5	41.2
Total Cr (mg/kg)	Descending S-shaped	53.2	520
Total Cu (mg/kg)		17.6	28.3
Total Cd (mg/kg)		0.1	0.6
Total Pb (mg/kg)		13.8	28
Total Hg (mg/kg)		0	0.1
Total As (mg/kg)		5.4	9.8

.

The Ascending S-type Membership Function (Equation (1)) is used to standardize the soil factor when the value lies between the lowest level (x₁) and the highest level (x₂) in soil classification. The specific formula is as follows:

$$f(x)=\left\{\begin{array}{cc}0.1& x\le x_1\\ 0.9(x-x_1)/(x_2-x_1)+0.1& x_1<x<x_2\\ 1.0& x\ge x_2\end{array}\right.$$

(1)

where x₁ and x₂ are the content values corresponding to the lowest and highest levels of the indicator in the soil classification, respectively.

The Descending S-type Membership Function (Equation (2)) is used for standardization when the soil factor lies between the lowest level (x₁) and the highest level (x₂). The specific formula is as follows:

$$f(x)=\left\{\begin{array}{cc}1& x\le x_1\\ 1-0.9(x-x_1)/(x_2-x_1)& x_1<x<x_2\\ 0.1& x\ge x_2\end{array}\right.$$

(2)

where x₁ and x₂ are the content values corresponding to the lowest and highest levels of the indicator in the soil classification.

The Parabolic Membership Function (Equation (3)) is applied when the soil factor has continuous variation, where turning points x₁ and x₄ correspond to the low and high content values at the lowest level in soil classification and turning points x₂ and x₃ correspond to the low and high content values at the highest level. The specific formula is as follows:

$$f(x)=\left\{\begin{array}{cc}1.0-0.9(x-x_3)/(x_4-x_3)& x_3<x<x_4\\ 1.0& x_2\le x\le x_3\\ 0.9(x-x_1)/(x_2-x_1)+0.1& x_1<x<x_2\\ 0.1& x\le x_1 or x\ge x_4\end{array}\right.$$

(3)

where x₁ and x₄ are the turning points corresponding to the low and high content values and x₂ and x₃ are the turning points corresponding to the low and high content values at the highest level in soil classification.

2.3.2. Establishment of the Minimum Data Set and Determination of Weights

By using principal component analysis (PCA), correlation analysis, and norm value comparison, the minimum data set (MDS) is selected from the total data set (TDS) of all soil factor indicators [20,21]. Principal components with eigenvalues greater than or equal to 1 are extracted through PCA [22]. In this study, considering the cumulative variance contribution of each principal component, only those with eigenvalues greater than or equal to 0.9 are considered. Evaluation indicators with loading values greater than or equal to 0.5 are grouped together [23]. If a specific indicator has loadings greater than or equal to 0.5 in different PCs, it is assigned to a group with other indicators that have a lower correlation. If an indicator has loadings lower than 0.5 across all PCs, it is assigned to the group with the highest loading values [24].

The norm value of evaluation indicators in each group is then calculated. The indicators with norm values within the top 10% range of the highest value in the group are selected [24]. The correlation of the selected indicators within each group is analyzed, and based on the Pearson correlation coefficient, a decision is made regarding whether to retain the indicator. If the correlation between indicators is low, all indicators in the group can be retained; if the indicators are highly correlated, the one with the highest norm value is selected for the MDS [25,26]. The norm value represents the ability of an indicator to explain the combined information. The calculation formula for the norm value is as follows (4), where N_ik represents the comprehensive loading of the i-th evaluation indicator on the first k principal components with eigenvalues greater than 1, u_ik is the loading of the i-th evaluation indicator on the k-th principal component, and λ_k is the eigenvalue of the k-th principal component.

$$N_{ik}=\sqrt{({\displaystyle \sum _{i=1}^k{u}_{ik}{}^2\lambda _k})}$$

(4)

2.3.3. Comprehensive Evaluation of Soil Quality

The comprehensive soil quality evaluation uses a fuzzy mathematics-based method, which involves determining the membership function and membership values based on soil factor parameters. The weights of the factors are then determined through the establishment of the minimum data set (MDS). The soil quality index (SQI) is calculated using the following Formula (5):

SQI = ∑(Wij × Nij)

(5)

where W_ij represents the weight coefficient of the j-th factor indicator for the i-th factor and N_ij represents the membership value of the j-th factor indicator for the i-th factor.

3. Results

3.1. Physical Properties of Reconstructed Soils in Slag Mountain

As shown in Figure 3 and

Table A1. Physical and chemical characteristics of reconstructed soil.

Sample No.	pH	Bulk Density (Mg/cm3)	Water Content (%)	Sand (%)	Silt (%)	Clay (%)	Total N (g/kg)	Total P (g/kg)	Total K (g/kg)	NH4+-N (mg/kg)	NO3−-N (mg/kg)	Available P (mg/kg)	Available K (mg/kg)
24ML-01-01	7.61	1.64	14.79	41.10	24.00	34.90	3.64	0.56	23.94	99.00	0.71	42.31	176
24ML-02-01	7.96	1.10	21.57	29.10	28.00	42.90	2.75	0.62	23.28	16.57	0.75	13.18	164
24ML-03-01	8.03	1.43	15.40	37.10	24.00	38.90	1.69	0.53	24.76	3.73	0.72	10.27	162
24ML-04-01	7.76	1.76	48.72	57.10	24.00	18.90	9.39	1.13	20.08	50.92	8.12	6.52	335
24ML-05-01	7.89	1.70	16.20	61.10	20.00	18.90	4.13	0.62	22.15	7.32	0.51	120.84	205
24ML-06-01	7.71	1.60	25.01	57.10	24.00	18.90	4.12	0.72	26.37	17.84	14.61	32.33	283
24ML-07-01	7.96	1.52	13.57	29.10	32.00	38.90	2.23	0.67	22.79	10.10	0.52	35.65	170
24ML-08-01	7.81	1.19	9.47	53.10	20.00	26.90	2.29	0.49	26.44	3.63	0.73	17.48	190
24ML-09-01	8.02	1.68	11.01	37.10	28.00	34.90	1.02	0.55	21.88	10.43	2.50	24.00	113
24ML-10-01	8.10	0.55	60.37	41.10	44.00	14.90	13.49	0.79	19.72	26.68	34.50	23.17	62

, the mean bulk density of the 2020 restored platform soils at No. 4 and 5 was 1.41 Mg/cm³, while the bulk density of the slopes was 1.63 Mg/cm³. The 2015 restored platform bulk density at No. 2 was 1.19 Mg/cm³, with the slope bulk density being 1.68 Mg/cm³, while the natural grassland had a bulk density of 0.55 Mg/cm³ (Figure 3a). The soil texture on both the restored slag heap slopes and platforms in 2020 and 2015 showed no significant differences, with textures mainly consisting of clay loam, sandy clay loam, and loam, while the natural grassland soil texture was primarily clay (Figure 3b). It can be seen that the bulk density of the restored platform and slope in the Jiangcang mining area is significantly higher than that of the natural grassland, which has a low bulk density of clay. Furthermore, although the differences in the soil bulk density and texture between the two restoration measures are not significant, the bulk density of the ecological restoration platform in both cases is relatively lower than that of the slope.

3.2. Chemical Properties of Reconstructed Soils in Slag Mountain

As shown in Figure 4 and Table A1, the mean pH value of the 2020 restored platform soils at No. 4 and 5 was 7.96, while the mean pH value of the slopes was 7.81 (Figure 4a). In 2015, the pH of the No. 2 platform was 7.81, while the slope pH was 8.02, and the natural grassland had a pH of 8.10 (Figure 4a). In 2020, the electrical conductivity on the platforms at No. 4 and 5 averaged 106.3 us/cm, while the slope conductivity averaged 311.7 us/cm (Figure 4b). In 2015, the electrical conductivity of the No. 2 platform was 144.4 us/cm, the slope conductivity was 87.7 us/cm, and the natural grassland conductivity was 283.8 us/cm (Figure 4b). In 2020, the average moisture content on the platforms of No. 4 and 5 was 17.72%, while the slope moisture content averaged 25.52% (Figure 4c). In 2015, the moisture content of the No. 2 platform was 9.46%, with a slope moisture content of 11.01%, and the natural grassland had a moisture content of 60.37% (Figure 4c). These data show that the pH values of the restored slag mountain reconstructions and the natural grassland are both weakly alkaline, with no significant differences, while the electrical conductivity of the restored slag mountain reconstructions is significantly higher than that of the natural grassland and the moisture content is significantly lower than in the natural grassland. The natural grassland is characterized by weak alkalinity, low electrical conductivity, and high moisture content, while the restored slag mountain reconstructions are weakly alkaline with high electrical conductivity and low moisture content.

As shown in Figure 5a,d, the total nitrogen, and nitrate nitrogen contents in the natural grassland were 13.49 g/kg, and 34.50 mg/kg, respectively, which were significantly higher than those in the 2015 and 2020 restored slag mountain reconstructions. The total phosphorus content in the natural grassland was 0.79 g/kg, slightly higher than in the 2020 and 2015 restored slag mountain reconstructions (Figure 5b). The available phosphorus content in the natural grassland was 23.17 mg/kg, slightly higher than in the 2015 restored slag mountain reconstructions but lower than in the 2020 restored slag mountain reconstructions (Figure 5e). The total potassium content in the natural grassland was 19.72 g/kg, slightly lower than in the 2020 and 2015 restored slag mountain reconstructions (Figure 5c). The available potassium content in the natural grassland was 62.00 mg/kg, significantly lower than in the restored slag mountain reconstructions (Figure 5f). The total nitrogen, total phosphorus, total potassium, ammonium nitrogen, nitrate nitrogen, available phosphorus, and available potassium contents in the 2020 restored slag mountain reconstructions were all higher than those in the 2015 restored slag mountain reconstructions, with the contents in the slope areas of both years being higher than in the platform areas.

3.3. Carbon Characteristics of Reconstructed Soils in Slag Mountain

As shown in Figure 6 and

Table A2. Carbon and heavy metal content of reconstructed soil.

Sample No.	Organic Matter (g/kg)	Organic Carbon (g/kg)	Microbial Biomass Carbon (mg C/kg)	Readily Oxidizable Carbon (g C/kg)	Cr (mg/kg)	Cu (mg/kg)	Cd (mg/kg)	Pb (mg/kg)	Hg (mg/kg)	As (mg/kg)
24ML-01	114.13	65.90	970.79	14.50	520	28.3	0.63	28	0.093	6.3
24ML-02	93.36	53.91	625.34	12.02	73.9	26.5	0.18	27.4	0.039	9.8
24ML-03	58.11	33.55	481.05	7.20	78	21.1	0.31	23.7	0.053	6.6
24ML-04	203.71	117.61	1166.01	31.36	54.6	19.1	0.16	16.2	0.036	9.2
24ML-05	96.76	55.87	701.30	16.51	75.2	26.1	0.23	23.1	0.041	5.4
24ML-06	160.42	92.62	1389.01	15.72	82	22.2	0.19	25	0.042	6.3
24ML-07	55.80	32.22	522.50	8.04	70.5	23.4	0.16	22.7	0.02	7.8
24ML-08	110.85	64.00	830.39	11.80	86.8	24.1	0.23	27	0.033	5.5
24ML-09	56.53	32.64	386.00	5.49	62.4	17.6	0.18	19.4	0.027	6.4
24ML-10	247.86	143.11	1611.62	41.20	53.2	18.9	0.10	13.8	0.042	8.9

, the soil organic matter (SOM), soil organic carbon (SOC), microbial biomass carbon (MBC), and easily oxidizable organic carbon (EOOC) contents in the natural grassland were 247.86 g/kg (Figure 6a), 143.11 g/kg, 1611.62 mg C/kg (Figure 6c), and 41.20 g C/kg (Figure 6d), respectively. In comparison, these values in 2015 were 110.85/56.53 g/kg, 64.00/32.64 g/kg, 830.39/386.00 mg/kg, and 11.80/5.49 g/kg, respectively. For the natural grassland, the corresponding values were 247.86 g/kg for organic matter, 143.11 g/kg for organic carbon, 1611.62 mg/kg for MBC, and 41.20 g/kg for EOOC. It is evident that the organic matter, organic carbon, MBC, and EOOC in the restored slag mountain platform and slope were significantly lower than those in the natural grassland. In 2020, the organic matter, organic carbon, MBC, and EOOC content in the restored slope were slightly higher than those in the platform; however, in 2015, these values in the restored slope were lower than those in the platform.

3.4. Heavy Metal Characteristics of Reconstructed Soils in Slag Mountain

Using the single-factor pollution index method, the single-factor pollution index for heavy metals in the reconstructed slag mountain reconstructions and natural grassland at the Jiangcang mining area was calculated. According to the data shown in

Table 4. Single-factor pollution index for heavy metals in reconstructed soils.

Sample	Cr		Cu	As		Cd		Pb		Hg
	Measured Value /Screening Value	Measured Value /Control Value	Measured Value /Screening Value	Measured Value /Screening Value	Measured Value /Control Value	Measured Value /Screening Value	Measured Value /Control Value	Measured Value /Screening Value	Measured Value /Control Value	Measured Value /Screening Value	Measured Value /Control Value
24ML-01	2.08	0.40	0.28	0.25	0.06	1.05	0.16	0.16	0.03	0.03	0.02
24ML-02	0.30	0.06	0.27	0.39	0.10	0.30	0.05	0.16	0.03	0.01	0.01
24ML-03	0.31	0.06	0.21	0.26	0.07	0.52	0.08	0.14	0.02	0.02	0.01
24ML-04	0.22	0.04	0.19	0.37	0.09	0.27	0.04	0.10	0.02	0.01	0.01
24ML-05	0.30	0.06	0.26	0.22	0.05	0.38	0.06	0.14	0.02	0.01	0.01
24ML-06	0.33	0.06	0.22	0.25	0.06	0.32	0.05	0.15	0.03	0.01	0.01
24ML-07	0.28	0.05	0.23	0.31	0.08	0.27	0.04	0.13	0.02	0.01	0.00
24ML-08	0.35	0.07	0.24	0.22	0.06	0.38	0.06	0.16	0.03	0.01	0.01
24ML-09	0.25	0.05	0.18	0.26	0.06	0.30	0.05	0.11	0.02	0.01	0.00
24ML-10	0.21	0.04	0.19	0.36	0.09	0.17	0.03	0.08	0.01	0.01	0.01

, Figure 7, and Table A2, the ratio of chromium (Cr) measurements to the screening value at the north slag platform of well 5 is 2.08, while the ratio of chromium (Cr) measurements to the control value is 0.40; the ratio of cadmium (Cd) measurements to the screening value is 1.05, and the ratio of cadmium (Cr) measurements to the risk control value is 0.16. Therefore, except for the Cr and Cd contents on the south slope of No. 5, which exceeded the soil environmental quality risk screening values, the heavy metals at all other points were below the risk screening values. Although the Cr and Cd contents on the slope of the south slag at well No. 5 exceeded the soil environmental quality risk screening values, they were still below the soil environmental quality risk control values.

3.5. Soil Quality Comprehensive Evaluation

3.5.1. Determination of the Minimum Data Set for Soil Quality Comprehensive Evaluation

Principal component analysis (PCA) was conducted on 18 soil indicators, including soil pH, bulk density, organic matter, total nitrogen, N-NH₄⁺, N-NO₃⁻, total phosphorus, available phosphorus, total potassium, available potassium, microbial biomass carbon (MBC), easily oxidizable organic carbon (EOOC), total Cr, total Cu, total Cd, total Pb, total Hg, and total As. The analysis revealed five principal components with eigenvalues greater than 1, contributing to a cumulative variance of 93.08%. The principal component loading matrix and norm values are shown in

Table 5. Principal component loading matrix and norm values.

Index	Principal Component					Norm Value
	PC1	PC2	PC3	PC4	PC5
Microbial Biomass Carbon	0.974	0.110	−0.039	0.134	−0.050	2.65
Organic Matter	0.959	0.022	0.194	0.126	−0.131	2.62
Easily Oxidizable Organic Carbon	0.898	0.012	0.403	0.005	−0.058	2.50
N-NO3−	0.886	−0.165	0.179	−0.277	−0.091	2.46
Total Nitrogen	0.883	−0.009	0.444	−0.044	−0.080	2.48
Total Cr	−0.063	0.977	−0.107	0.012	0.068	2.08
Total Hg	0.114	0.933	−0.129	−0.040	0.063	2.01
Total Cd	−0.237	0.922	−0.196	0.036	0.118	2.08
N-NH4+	0.223	0.897	0.228	0.238	−0.134	2.04
Total Cu	−0.208	0.568	−0.436	0.025	0.276	1.51
Total Potassium	−0.262	0.069	−0.942	0.123	0.003	1.59
Total Pb	−0.469	0.377	−0.743	0.074	0.057	1.87
Available Potassium	0.071	−0.025	−0.111	0.977	−0.030	1.19
Bulk Density	−0.533	0.154	0.120	0.733	0.280	1.75
pH	−0.105	−0.581	0.339	−0.702	−0.088	1.60
Total Phosphorus	0.524	−0.144	0.556	0.559	−0.255	1.82
Available Phosphorus	−0.051	0.079	0.069	0.036	0.956	1.13
Total As	0.273	−0.099	0.464	−0.055	−0.673	1.30
Eigenvalue	7.305	4.462	2.245	1.410	1.332
Principal Component Contribution (%)	29.71	24.38	15.82	13.86	9.31
Cumulative Contribution (%)	29.71	54.08	69.91	83.77	93.08

. Evaluation indicators with loading values greater than or equal to 0.5 were grouped together. If the loading values of an indicator were all below 0.5, it was assigned to the group with the highest loading value. The norm values within the top 10% of the highest value in each group were selected for the minimum data set (MDS). To avoid the duplication and redundancy of selected indicators, correlation analysis was conducted.

The final selected indicators for the minimum data set are as follows: In principal component 1, the contribution rate of the eigenvalue was 29.71%, with microbial biomass carbon (MBC), organic matter, easily oxidizable organic carbon (EOOC), N-NO₃⁻, total nitrogen, bulk density, and total phosphorus showing absolute loading values greater than 0.5. Based on the highest norm value, microbial biomass carbon (MBC) (norm value of 90%) was selected, and organic matter, easily oxidizable organic carbon (EOOC), N-NO₃⁻, and total nitrogen also met the criteria. Correlation analysis indicated that microbial biomass carbon (MBC), organic matter, easily oxidizable organic carbon (EOOC), N-NO₃⁻, and total nitrogen had significant correlations (Figure 8). To avoid redundancy, only microbial biomass carbon (MBC), with the highest norm value, was retained in the MDS.

In principal component 2, the loading values of total Cr, total Hg, total Cd, N-NH₄⁺, total Cu, and pH were all greater than 0.5. Total Cr, with a norm value of 90%, was selected as the indicator with the highest norm value. Total Cr, total Hg, total Cd, and N-NH₄⁺ met the conditions, and correlation analysis showed that these indicators are significantly correlated in pairs (Figure 8). Therefore, total Cr, exhibiting both the maximum loading value and the highest norm, was selected for the minimum data set.

In principal component 3, the contribution rate of the eigenvalue was 15.83%, with total K, total Pb, and total phosphorus having loading values greater than 0.5. The indicator with the highest norm value was total Pb. Further analysis indicates that the total phosphorus indicator is within 90% of the norm value of total Pb. Furthermore, correlation analysis reveals a significant negative correlation between total Pb and total phosphorus (Figure 8), leading to the selection of total Pb for the minimum data set.

In principal component 4, the contribution rate of the eigenvalue was 13.86%, and the loading values of available potassium, bulk density, pH, and total phosphorus were all greater than 0.5. The indicator with the highest norm value was total phosphorus, and the bulk density norm value was within 90% of the total phosphorus. Combined with correlation analysis (Figure 8), it was determined that total phosphorus and bulk density both have a low degree of correlation; thus, both were included in the minimum data set.

In principal component 5, the contribution rate of the eigenvalue was 9.31%, with available phosphorus and total As having loading values greater than 0.5. The indicator with the highest norm value was total As, and the norm value of available phosphorus did not fall within 90% of total As’s norm value; therefore, only total As was included in the minimum data set.

In conclusion, the selected indicators for the minimum data set (MDS) for evaluating soil quality in the Muli Jiangcang mining area were microbial biomass carbon (MBC), total Cr, total Pb, bulk density, total phosphorus, and total As. Among these, bulk density represents a soil physical property indicator, total phosphorus represents a soil chemical fertility indicator, microbial biomass carbon (MBC) represents a soil carbon characteristic indicator, and total Cr, total Pb, and total As represent soil heavy metal property indicators.

3.5.2. Determination of Weights for Soil Environmental Quality Comprehensive Indicators

Principal component analysis was conducted separately for the total data set (TDS) and the minimum data set (MDS) to obtain the weight for each indicator, as shown in

Table 6. Weights of soil environmental quality comprehensive evaluation indicators.

Index	Total Data Set (TDS)	Minimum Data Set (MDS)
	Weights	Weights
pH	0.05
Bulk Density	0.06	0.07
Organic Matter	0.06
Total Nitrogen	0.06
N-NH4+	0.05
N-NO3−	0.06
Total Phosphorus	0.06	0.08
Available Phosphorus	0.04
Total Potassium	0.05
Available Potassium	0.04
Microbial Biomass Carbon	0.06	0.32
Easily Oxidizable Organic Carbon	0.06
Total Cr	0.06	0.26
Total Cu	0.05
Total Cd	0.06
Total Pb	0.06	0.17
Total Hg	0.06
Total As	0.05	0.10

.

3.5.3. Soil Environmental Quality Comprehensive Evaluation

Soil quality indexes (SQIs) were calculated using the membership functions and corresponding indicator weights from Table 6 for each sampling point, as shown in

Table 7. Soil environmental quality evaluation results for TDS and MDS.

Sampling Point	TDSSQI	MDSSQI	Q
South Slag Mountain Slope at No. 5	0.52	0.34	64.20
South Slag Mountain Platform at No. 5	0.56	0.49	69.14
North Slag Mountain Platform at No. 5	0.52	0.49	64.20
North Slag Mountain Slope at No. 5	0.71	0.73	87.65
North Slag Mountain Platform at No. 4	0.63	0.58	77.78
North Slag Mountain Slope at No. 4	0.70	0.71	86.42
South Slag Mountain Slope at No. 4	0.60	0.50	74.07
Platform at No. 2	0.64	0.61	79.01
Waste Dump Slope at No. 2	0.56	0.52	69.14
Natural Grassland	0.81	0.91	100.00

. Descriptive statistics for the TDS and MDS are provided in

Table 8. Statistical features of soil environmental quality evaluation results based on TDS and MDS.

Comprehensive Index	Min	Max	Mean	SD	CV(%)	R²
SQI (TDS)	0.52	0.81	0.62	0.09	14.51	0.92 *
SQI (MDS)	0.34	0.91	0.59	0.16	27.11	—

* p ≤ 0.05.

, and the fitting curve is shown in Figure 9. The results indicate that the mean values and variation coefficients of the SQI for both TDS and MDS are very similar, and the linear regression relationship between the two is y = 1.696x − 0.473, with a correlation coefficient of 0.92 (p < 0.01), which is highly significant (Figure 9). This suggests that the MDS-based indicator system constructed in this study can replace the TDS for evaluating soil environmental quality in the Jiangcang mining area.

A higher soil quality index (SQI) value indicates superior soil quality, using the SQI of natural grassland as a reference standard. Based on the principle of natural selection optimization, we linked remediation targets to regional native soil quality, using Q to represent the degree of soil environmental quality restoration. Following the functional equivalence principle, we classified the post-remediation soil environmental quality of slag heaps into three tiers: Q ≥ 0.8, 0.6 ≤ Q < 0.8, and Q < 0.6. And this classification is consistent with the findings of relevant scholars [27,28]. The evaluation results reveal that the comprehensive assessment of soil quality for the restoration of the northern slope of wells No. 5 and No. 4 are Q ≥ 0.8, indicating effective restoration outcomes. In contrast, the comprehensive assessment of soil quality for the southern slag hill of well No. 5, well No. 4, and well No. 2 are 0.6 ≤ Q < 0.8.

$$Q={\displaystyle \frac{{TDSSQI}_i}{{TDSSQI}_n}}$$

(6)

where i represents the sample number of the slag mountain and n for that of the natural grassland.

4. Discussion

4.1. Impact of Soil Physical Properties on Soil Quality

Soil texture and bulk density are the core indicators reflecting soil environmental quality, directly influencing ecosystem health and restoration capacity. In the Jiangcang mining area, natural grasslands are predominantly silty loam (41.10% sand, 44.00% silt, and 14.90% clay), with a relatively low bulk density (0.55 Mg/cm³). In contrast, reconstructed soils on waste dumps, due to factors such as mechanical compaction, engineering sieving, and organic matter loss, are primarily composed of sand and clay, with a lower silt content, resulting in textures such as clay loam or sandy clay loam. The bulk density of these reconstructed soils is 2–3 times higher than that of natural grasslands. Located on the northeastern edge of the Qinghai–Tibet Plateau, the Jiangcang mining area receives an average annual rainfall of 500 mm, concentrated mainly in July and August. This means that during months with less rainfall, sandy soils exhibit stronger drought sensitivity, with higher evaporation rates, while clayey soils are prone to cracking under drought conditions, significantly accelerating water loss [29]. These factors exacerbate plant water stress, directly limiting photosynthesis and nutrient uptake, thereby inhibiting growth. Additionally, studies have shown that when bulk density exceeds 1.6 Mg/cm³, soil aggregates break down, macropores are compressed, and pores >30 μm are significantly reduced, impairing water retention and decreasing plant-available water while increasing root penetration resistance [30]. High bulk density also reduces the diffusion rates of O₂ and CO₂ while increasing N₂O emissions, intensifying anaerobic conditions [31] and further hindering plant growth. Principal component analysis (PCA) in this study further confirmed that bulk density is a key indicator (weight: 0.07) in the minimum data set (MDS) for soil quality. Its variation explains the significant differences in texture and bulk density between reconstructed waste dump soils and natural grasslands, resulting in lower soil quality in the reconstructed areas. Therefore, during the mechanical crushing of reconstructed soils in this region, it is essential to employ graded crushing to increase sand content while incorporating lignin-modified biochar to reduce bulk density.

4.2. Impact of Soil Chemical Properties on Soil Quality

The pH levels of both the reconstructed soil and natural grassland in the Jiangcang mining area are comparable and relatively high, ranging from 7.8 to 8.1, which indicates a weak alkaline nature. A high soil pH predominates negative charges, thereby reducing the solubility of trace elements and resulting in lower concentrations of these elements in the soil solution [32]. Additionally, at a higher soil pH, the mineralizable fractions of C and N increase because the bond between organic constituents and clays is broken [33].

Electrical conductivity refers to the ability of various cations and anions in soil leachate to conduct electricity. This property is primarily influenced by the salt content, moisture, and texture of the soil, and it is positively correlated with soil salinity [34,35]. In the Jiangcang mining area, the electrical conductivity of reconstructed slag heaps increased by 25% to 290% compared to natural grasslands. This increase in electrical conductivity results in higher soil salinity, which reduces the osmotic potential of the soil solution, thereby limiting the ability of soil microorganisms to access water [36]. Consequently, this leads to the cell desiccation and lysis of microorganisms [37], alters the community structure of microbial bacteria and fungi [38,39], causes ion-specific toxicity, and inhibits microbial activity [40]. As a result, microbial activity and diversity decrease, along with a diminished capacity of soil microorganisms to decompose soil organic matter (SOM) [41,42,43].

Furthermore, most plants can only tolerate low salinity levels. Exceeding this threshold can lead to adverse effects, such as slowed growth and/or death, which subsequently reduces plant productivity and results in the decreased input of soil organic matter (including carbon) and lower soil organic carbon content [44]. Existing research indicates that in arid and semi-arid regions, low groundwater content in the shallow soil layer intensifies capillary action, leading to the significant accumulation of salts in the soil. This finding aligns with observations that the moisture content of reconstructed slag heaps in the Jiangcang mining area is lower than that of natural grasslands, yet their electrical conductivity is significantly higher. Therefore, the reconstructed soil in the slag heaps exhibits characteristics of low moisture content and high electrical conductivity, both of which are detrimental to vegetation growth and ecological restoration.

4.3. Impact of Soil Nutrients Characteristics on Soil Quality

The nutrient content of the soil plays a crucial role in plant growth, having significant importance in nutrient cycling, energy flow, and the maintenance of ecosystem functions [45], and is an important indicator for assessing soil quality. Total soil nutrients represent the maximum potential of nutrient supply, while available nutrients indicate the portion that can be directly absorbed by plants. In addition, studies have shown that soil quality is not only related to the content and form of soil nutrients, such as nitrogen (N), phosphorus (P), and potassium (K), but also to the nutrient content ratios [46]. Nitrogen in soil exists in both inorganic and organic forms, with organic nitrogen being the primary form in the soil nitrogen pool [47], playing an important role in soil fertility, nitrogen cycling, and environmental protection [48]. Inorganic nitrogen, including ammonium and nitrate nitrogen, is the effective nitrogen form that plants directly utilize for growth [49].

The C/N ratio reflects the decomposition of organic matter and nitrogen supply capacity in soil. When 15 < C/N < 30, the amount of effective nitrogen provided by the soil is less than the microbial assimilation capacity, causing nitrogen deficiency in plants and affecting their growth [50]. When the C/N ratio < 15, the soil provides more effective nitrogen than microorganisms can assimilate, allowing plants to obtain the nitrogen needed for growth from organic matter mineralization. The C/P ratio indicates phosphorus availability, and a lower C/P ratio favors the microbial decomposition of organic matter and the release of nutrients, thus increasing soil phosphorus availability [50]. Soil N/P is typically used to analyze limiting nutrients in soil. When the soil N/P ratio is below 14, nitrogen is considered the limiting factor; when N/P exceeds 16, phosphorus is considered the limiting factor [51,52].

As shown in Figure 5, the total nitrogen, nitrate nitrogen, and ammonium nitrogen in the restored waste rock piles of the Jiangcang mining area are significantly lower than those in natural grasslands, with the restoration platform being lower than the restoration slope. In 2015, the C/N, C/P, and N/P ratios for the restored slag platform and slope were 28.00 and 31.98, 129.75 and 59.68, and 4.63 and 1.87 (Figure 10), respectively. In 2020, the C/N, C/P, and N/P ratios for the restored slag platform and slope were 17.69 and 16.88, 80.18 and 99.91, and 4.76 and 5.98 (Figure 9), respectively. The C/N, C/P, and N/P ratios for natural grasslands were 10.60, 182.25, and 17.19 (Figure 10), respectively. It can be seen that the C/N and C/P ratios of the restored waste rock and natural grasslands in the Muli mining area are higher than the national averages of 12 and 61, while the C/P ratio is lower than the national average of 5.2 [53]. These results indicate that the soils of the restored spoil heap in both 2015 and 2020 exhibit characteristics of low nitrogen, medium phosphorus, and high potassium content, with the soil quality levels being primarily influenced by nitrogen stress. Notably, nitrogen deficiency in the 2015 restored spoil heap is more pronounced than in the 2020 restoration, with the platform experiencing greater nitrogen deficiency than the slope.

The possible reasons for this phenomenon include the following: (1) the total nitrogen in the restored slag mountain soils is significantly lower than in the natural grassland, and the microbial biomass in the restored soils is also lower, leading to the insufficient conversion of organic nitrogen to inorganic nitrogen; (2) the high pH of the Jiangcang mining area facilitates the processes of dissimilatory nitrate reduction to ammonium and denitrification, which increases ammonia volatilization and nitrogen loss; (3) in platforms, nitrate nitrogen is more easily leached by rainfall and migrates, accumulating in the slopes, leading to significantly higher nitrate nitrogen levels in the slopes compared to the platforms and more severe nitrogen deficiency in the platforms.

4.4. Impact of Soil Carbon Processes on Soil Quality

Soil organic carbon (SOC) refers to the total amount of carbon-containing organic matter present in the soil, which is twice that of atmospheric carbon and three times that of the total carbon found in the Earth’s vegetation. SOC plays a crucial role in enhancing soil fertility, promoting plant growth, and stimulating soil microbial activity, and it serves as an important indicator for soil quality assessment [54]. However, many components of soil SOC possess stable physical and chemical properties, and their response to changes in external environmental conditions occurs over extended periods, making it challenging to reflect ecosystem changes in a timely and efficient manner [55]. In contrast, soil microbial biomass carbon (SMBC) and easily oxidizable organic carbon (EOOC), as vital components of active soil carbon, exhibit high sensitivity to alterations in the surrounding environment. They can swiftly respond to the health and vitality of the ecosystem, thus indicating whether the ecosystem has been compromised and serving as recognized early warning indicators [56,57]. The highly significant correlation between soil microbial biomass carbon (SMBC) and soil organic carbon (SOC) (R² = 0.88) (Figure 11a) indicates that microbial biomass carbon can serve as a sensitive indicator for the restoration of carbon pools in alpine mining areas [58].

The organic matter content in natural grasslands is 247.86 g/kg, with annual litter input from alpine meadow vegetation ranging from 1.2 to 1.8 t/ha [59]. In contrast, the vegetation coverage and litter amount in the restoration area are lower, with organic matter only representing 20–50% of those found in natural grasslands. As a result, open-pit mining has resulted in a significant loss of SOC in the Jiangcang mining area, with the SOC content in the restored area declining by 50% to 80% compared to natural grasslands. This change in carbon stocks aligns with the carbon loss patterns observed in the high-altitude mining regions of the Tibetan Plateau [60].

Although the total SOC in the restored area is markedly lower than that in natural grasslands, the proportion of SMBC in the restored area is significantly higher (Figure 11c), suggesting that increased microbial activity may accelerate the organic carbon turnover [61]. However, the proportion of EOOC is lower than that in natural grasslands (Figure 11d), indicating a reduction in the easily decomposable carbon pool and an increase in carbon stability. This characteristic of “high activity–low decomposability” may arise from the mineral protection effect of added soil, which suppresses the mineralization of easily decomposable carbon [62,63], or from low temperatures (average annual temperature −4 °C) that slow microbial metabolic rates and prolong the carbon sequestration period. However, if the EOOC/SOC ratio remains below 25%, it may indicate that carbon stability is overly reliant on physical protection rather than biological activity, necessitating adjustments to organic amendment strategies (such as the addition of lignin-like slow-decomposing materials) to improve soil quality.

5. Conclusions

This study focuses on the Jiangcang mining area of the Qinghai–Tibet Plateau, analyzing the physical, chemical, carbon characteristics, and heavy metal indicators of soil samples from the restoration slag platforms and slopes (0–20 cm) in 2015 and 2020, revealing the impact of different soil reconstruction methods on soil function and ecological restoration capacity.

The results indicate that the comprehensive evaluation of soil environmental quality for the total data set (TDS) and the minimum data set (MDS) in the Jiangcang mining area shows a highly significant linear regression relationship, suggesting that MDS parameters such as bulk density, total phosphorus, microbial biomass carbon, total Cr, and total Pb content can replace TDS to assess the soil recovery quality in the mining area. The evaluation in 2020 shows that the adopted techniques resulted in good soil recovery, with the comprehensive evaluation grade of soil quality on the northern slopes of the slag hill at well No. 5 and well No. 4 being Q ≥ 0.8, indicating effective restoration.

However, due to the poor conditions of the original soil matrix and extreme environmental factors in the mining area, the restoration of the ecosystem still faces significant challenges. The bulk density of the restored soil was significantly higher than that of the native soil in natural grasslands, and the high bulk density inhibited soil aeration and water retention, thus reducing the movement and biological availability of gases and water in the soil. Additionally, the restored soil exhibited high electrical conductivity and low moisture content, further limiting the availability of water for soil microbes, leading to cell desiccation and lysis, suppressing microbial activity, and decreasing the soil’s organic matter decomposition capability. Furthermore, the C/N ratio of the restored soil was between 15 and 30, with the effective nitrogen available being less than the assimilation amount by microbes, resulting in nitrogen deficiency that adversely affects plant growth. Meanwhile, the restored soil showed an N/P ratio of less than 14, indicating that the soil is primarily nitrogen-limited, consistent with the significantly lower total nitrogen, nitrate, and ammonium nitrogen levels compared to natural grasslands.

The study further indicates that the organic matter content in the restored area is only 20–50% of that in the natural grasslands, leading to insufficient carbon input in the mining area and a significant decrease in total soil organic carbon (SOC) compared to natural grasslands. However, the restored soil exhibits high activity and low decomposability characteristics, indicating a good carbon sequestration capacity.

In summary, this study provides scientific evidence for ecological restoration in high-altitude mining areas. The overburden in the mining area not only enhances soil erosion resistance and nutrient retention but also significantly improves regional carbon sink capacity and water conservation functions. Moreover, the study highlights the importance of the water–carbon process coupling in ecological restoration in high-altitude mining areas, offering guidance for future ecological restoration practices. Future research should focus on the long-term effects of soil reconstruction in high-altitude mining areas and further explore more efficient ecological restoration technologies to address the dual pressures of climate change and human activities on high-altitude ecosystems.