Effects of Vetiveria zizanioides on the Restoration and Succession of Coal Gangue Mountain Plant Communities in Different Years

Abstract

The restoration of vegetation on coal gangue mountains has always been an area of concern, and therefore, an important area of research. Liupanshui city in Guizhou province, China, has a large number of coal gangue mountains, and for this reason, was chosen for studying vegetation succession on these sites. Vetiveria zizanioides is known to accelerate the restoration of vegetation on coal gangue mountains and to shorten community succession timeframes. Because of this, we investigated different successional stages after the planting of V. zizanioides on coal gangue mountains in the Dahe coal mine comprehensive environmental governance project area in Liupanshui city. Through field community surveys and model prediction, the effects of planted V. zizanioides on the species composition, species diversity, and community succession of gangue mountains 3, 6, 8, 10, and 13 years after planting were explored. In total, 35 plant species belonging to 17 families and 32 genera were recorded across the five different coal gangue mountains. With more time after planting, the height, coverage, density, and biomass of V. zizanioides all decreased, but increased for Miscanthus floridulus. The Simpson diversity index, Shannon–Wiener diversity index, and Pielou evenness index all first increased before decreasing over time; maximum values were recorded for the coal gangue mountain 8 years after planting of V. zizanioides. According to different similarity and dissimilarity indices, the successional stages became more similar with increasing time after planting. According to biomass fitting and prediction curves, the succession process of coal gangue mountain plant communities could be divided into a V. zizanioides community stage, a M. floridulus community stage, and a woody plant stage, that starts to approach the natural community of evergreen broad-leaved forests, with durations of 0–5.62 years, 5.62–17.48 years, and over 17.48 years, respectively.

1. Introduction

Coal mining usually leads to a series of environmental problems such as land damage, vegetation destruction, and air pollution, causing serious damage to the ecosystem [1,2]. As the main solid waste from coal mining and processing, coal gangue accounts for 15% to 30% of the raw coal output. Due to its low calorific value and difficult utilization, it is generally stacked in the open air and discarded on the surface around the production areas, forming coal gangue mountains [3,4,5]. By 2020, China had accumulated more than 6 billion tons of coal gangue. Liupanshui is one of the main coal production cities in southwest China, with prospective reserves of coal resources of 84.4 billion tons. This area has the highest coal gangue emission and stack in the entire province, with average annual coal gangue emissions of more than 10 million tons and cumulative emissions of 120 million tons [6]. This has resulted in the formation of hundreds of coal gangue mountains that not only occupy a large amount of land, but also have become one of the main sources of environmental pollution and deterioration in the mining area. At the same time, heavy metals in coal gangue can be easily leached, leading to water pollution and posing a potential threat to the groundwater and farmland systems near coal gangue mountains [7].

Vegetation restoration and the establishment of stable artificial plant communities are fundamental pathways to the governance and ecological reconstruction of coal gangue mountain areas [8]. Due to their special site conditions, poor structure and porosity, lack of soil nutrients, low capacity for water and fertilizer conservation, and poor plant survival, restoration and reconstruction of vegetation are particularly difficult in these areas [9]; therefore, artificial planting has become the main strategy for restoration of these environments [10]. The principle of restoration and reconstruction selection is to select plants with fast growth, high adaptability, and positive effects on soil structure and porosity improvement. Vetiveria zizanioides is a perennial herbaceous plant of the Poaceae family, which has a developed root system, wide adaptability, strong tolerance to environmental conditions, and rapid growth. It is often used in the ecological restoration of coal mine wasteland, as it can not only grow in poor habitats such as tailings waste created during product separation [11,12,13], but can also absorb and accumulate copper, zinc, lead, chromium, and nickel at thresholds of 15 mg/kg, 880 mg/kg, 78 mg/kg, 5–18 mg/kg, and 347 mg/kg, respectively [14,15,16,17].

The vegetation restoration of gangue mountains is actually a secondary succession process of the vegetation community, and the direction of vegetation succession is always towards stable community structure and function [18]. Different vegetation restoration times lead to different dominant species in the succession process. As restoration years pass, annual plants are typically dominant in the early stage (2–3 years), while perennial plants become the dominant species in the middle stage (5–10 years) and grass shrub communities are formed in the later stage (more than 15 years) [19]. The succession process of vegetation can differ greatly due to artificial assistance and the choice of vegetation restoration mode [20]. Vegetation restoration in artificial planting areas is significantly shorter than that in natural restoration areas [21].

At present, the research on vegetation restoration of coal gangue mountains mainly focuses on vegetation growth, soil matrix improvement, vegetation restoration mode, and vegetation restoration mechanisms; there are few studies on community succession after vegetation restoration. Based on field community investigation and biomass fitting prediction analysis, this paper explored the effects of planted V. zizanioides on plant communities of gangue mountains at five different time periods after its planting (3, 6, 8, 10, and 13 years after planting). We assessed the plant community composition and species diversity for the five successional stages, as well as the β-diversity between the stages. We aim to provide a theoretical basis for vegetation restoration and long-term vegetation maintenance of coal gangue mountains.

2. Materials and Methods

2.1. Overview of the Study Area

The study area was located in the Dahe coal mine comprehensive environmental governance project area, Dahe town, Zhongshan district, Liupanshui city, Guizhou Province, China (104°50′ E, 26°37′ N), which is 6 km away from downtown Liupanshui. The average altitude is 1600 m and the northwestern part of the area is higher than the southeast. It has a north subtropical monsoon climate, and the annual average temperature is about 13 °C. The average temperature in the coldest month is 2.9 °C in January, and that in the warmest month is 29.6 °C in July. The frost-free period is 242 d, the annual average sunshine duration is about 1253 h, and the annual average precipitation is 1234.7 mm (mainly from April to October) [22]. The zonal vegetation is subtropical evergreen broad-leaved forest. Dahe coal mine was built in 1970. The coal gangue mountain present here has a height difference of more than 60 m, a slope of 15–40°, and gravel and stones greater than 1 mm account for more than 80% of the soil. The structure is loose, the soil is deficient in nutrients, and the water holding capacity is poor [22]. V. zizanioides was planted to carry out comprehensive mine governance and ecological vegetation restoration. In April of the years 2002, 2005, 2007, 2009, and 2012, new coal gangue mountains were produced in the study area. In the same year of production, V. zizanioides was planted on these mountains at a row spacing of 50 cm × 20 cm with 5~7 tillers. In each planting hole, about 0.5 kg of cultivated soil was added before planting. Each plant was sufficiently watered to wet the soil after planting. After 1 month, plants that had not survived were replanted. No management was implemented for surviving plants, and V. zizanioides was allowed to grow naturally afterwards [23]. Thus, in the year 2015, five different coal gangue mountains (I to V) existed with an increasing succession time after V. zizanioides planting. Their geographical locations of the gangue mountains are shown in Figure 1, their basic situation in the year 2015 is summarized in

Table 1. Basic situation of five coal gangue mountains in the year 2015.

Coal Gangue Mountains	Planting Time (Year)	Years after Planting (Year)	Altitude (m)	Planting Density (cm × cm)	Dominant Species
I	2012	3	1860	50 × 20	Vetiveria zizanioides
II	2009	6	1879	50 × 20	Vetiveria zizanioides
III	2007	8	1880	50 × 20	Vetiveria zizanioides
IV	2005	10	1846	50 × 20	Miscanthus floridulus
V	2002	13	1793	50 × 20	Miscanthus floridulus

.

2.2. Plot Setting and Quadrat Investigation

In 2015, three plots of 25 m × 25 m were randomly set up in each of the five coal gangue mountains, and all species in the plots were recorded. In April, June, August, and October 2015, three survey quadrats of 1 m × 1 m herbaceous plants were set along the diagonals of each plot of the five coal gangue mountains, respectively, and a total of 180 herbaceous quadrats were investigated. The plant species, number of individuals, height, and coverage of each species in the quadrats were recorded. The aboveground parts of the plants were cut by species, taken and dried to constant weight, and the dry biomass of each species on the ground was measured.

2.3. Data Analysis

2.3.1. α Diversity

As the basic characteristics of the community, α diversity can be used to indicate the habitat status, composition structure, individual distribution pattern and other characteristics of the community [24].

The importance value was used to measure the dominance of species in the five coal gangue mountains, according to the following formula [25]:

$$P_i=\left(R{H}_i+R{C}_i+R{D}_i\right)∕3$$

(1)

The following equations were used to calculate α diversity:

Simpson diversity index (D)

$$D=1-{\displaystyle \sum }_{i=1}^S{P}_i{}^2$$

(2)

Shannon–Wiener diversity index (H)

$$H=-{\displaystyle \sum }_{i=1}^S{P}_i{ \mathit{lnP}}_i$$

(3)

Pielou evenness index (J)

$$J=\frac{H}{\mathit{lnS}}$$

(4)

Margalef richness index (R)

$$R=\left(S-1\right)∕\mathit{ln}N$$

(5)

where RH_i represents the relative height, RC_i represents the relative coverage, and RD_i represents the relative density; S represents the total number of species in the quadrat, and N represents the number of individual plants in the quadrat.

2.3.2. β Diversity

β diversity is used to indicate the degree of habitat separation by species and the habitat diversity of different regions, and can also measure the species composition of the five coal gangue mountains [26].

The following equations were used to calculate β diversity:

Cody index (β_c)

$${\beta }_c=\left[g\left(T\right)+1\left(T\right)\right]∕2$$

(6)

Whittaker index (β_ws)

$${\beta }_{ws}=S∕ma-1$$

(7)

Jaccard index (C_j)

$$C_j=j∕\left(a+b-j\right)$$

(8)

Sorenson index (C_s)

$$C_s=2j∕\left(a+b\right)$$

(9)

where β_c and β_ws represent dissimilarity measurements, C_j and C_s represent similarity measurements, S represents the total number of species recorded in the study system, ma represents the average number of species in each plot, g(T) represents the number of species increased along environmental gradient T, 1(T) represents the number of species decreased along environmental gradient T, a represents the number of species in stage A, b represents the number of species in stage B, and j represents the number of species in both stage A and B.

Statistical analysis was conducted with SPSS 16.0 software (Chicago, IL, USA). One-way ANOVA and ANOVA/Duncan multiple comparison tests were used to test the differences in height, coverage, density, importance, biomass, and diversity indices among the five coal gangue mountains, and the significance level a = 0.05. The regression method of curve estimation in SPSS 16.0 was used to determine the regression equation with the highest fitting degree, and the regression relationship between the dry biomass and planting years of V. zizanioides, Miscanthus floridulus, and woody plants were analyzed to predict the changes in dominant species in the future. SigmaPlot 14.0 (St. Louis, MO, USA) was used for plotting.

3. Results

3.1. Changes in Species Composition and the Importance Value in Different Coal Gangue Mountains

Table 2. Species composition of plots in different coal gangue mountains.

Family Name	Speices Name	Plots
		I	II	III	IV	V
Poaceae	Vetiveria zizanioides	1	1	1	1	1
	Miscanthus floridulus	-	1	1	1	1
	Cynodon dactylon	1	-	-	-	-
	Cymbopogon caesius	-	1	-	-	-
	Chrysopogon aciculatus	-	1	1	-	1
	Digitaria sanguinalis	-	-	1	-	-
	Zoysia japonica	1	-	1	1	1
Fabaceae	Vicia sepium	-	1	-	-	-
Asteraceae	Erigeron annuus	1	-	1	-	-
	Erigeron acris	-	-	-	1	1
	Artemisia argyi	1	1	-	-	-
	Artemisia sieversiana	-	-	1	1	1
	Sonchus arvensis	1	1	-	-	-
	Taraxacum mongolicum	-	1	-	-	-
	Senecio scandens	-	-	1	1	-
	Artemisia carvifolia	1	1	1	-	-
	Aster indicus	1	-	-	-	-
Oxalidaceae	Oxalis corniculata	-	1	1	1	1
Polygonaceae	Fallopia multiflora	-	1	-	-	-
Amaranthaceae	Alternanthera philoxeroides	1	-	-	-	-
Clusiaceae	Hypericum monogynum	-	-	1	1	1
Oleaceae	Ligustrum lucidum	-	-	-	1	1
Loganiaceae	Buddleja fallowiana	1	-	1	1	-
Lamiaceae	Prunella vulgaris	-	1	1	-	-
	Clinopodium chinense	-	1	1	-	-
Convolvulaceae	Convolvulus arvensis	-	-	-	1	-
Geraniaceae	Geranium wilfordii	-	1	1	-	-
Apiaceae	Cnidium monnieri	-	-	1	1	-
Rosaceae	Duchesnea indica	-	1	-	1	-
	Rubus idaeus	-	-	1	1	1
Equisetaceae	Equisetum arvense	1	-	1	-	-
Ranunculaceae	Anemone hupehensis	-	-	-	1	-
	Anemone vitifolia	1	1	1	-	-
Betulaceae	Betula luminifera	-	-	-	-	1
	Corylus heterophylla	-	-	-	-	1
Total: 17	35	12	16	19	15	12

Note: “1” means that this species appeared within plots of the five coal gangue mountains. “-” means that this species has not appeared within plots of the five coal gangue mountains. “I”, “II”, “III”, “IV”, and “V” correspond to the five coal gangue mountains 3, 6, 8, 10 and 13 years after planting, respectively.

shows that 35 plant species were found in the plots of the five coal gangue mountains 3, 6, 8, 10, and 13 years after planting, belonging to 17 families and 32 genera. Of these, Asteraceae had the most species (nine species), accounting for 25.71% of the species; Poaceae followed with seven species, accounting for 20% of total species. The total number of species in the five coal gangue mountains 3, 6, 8, 10, and 13 years after planting were 12, 16, 19, 15 and 12, respectively. V. zizanioides was the only species that appeared in all the five coal gangue mountains. There were three, three, one, two, and two species only appearing in the five coal gangue mountains 3, 6, 8, 10, and 13 years after planting, respectively. Two woody plants, Hypericum monogynum of Clusiaceae and Rubus idaeus of Rosaceae, appeared in the coal gangue mountain 8 years after planting. Ligustrum lucidum was also found in the coal gangue mountain 10 years after planting, which had three perennial woody plants. There were two more woody plants from Betulaceae found in the coal gangue mountain 13 years after planting, which had five woody plants.

As shown in

Table 3. Species importance value of quadrats in different coal gangue mountains.

Family Name	Speices Name	Quadrats
		I	II	III	IV	V
Poaceae	Vetiveria zizanioides	0.40	0.37	0.34	0.24	0.24
	Zoysia japonica	0.15	0.04	0.01	0.06	0.04
	Miscanthus floridulus	-	0.21	0.23	0.44	0.47
Lamiaceae	Clinopodium chinense	0.18	0.05	0.03	0.07	0.06
	Prunella vulgaris	-	0.03	0.03	-	-
Ranunculaceae	Anemone vitifolia	0.06	0.03	0.03	-	-
Asteraceae	Artemisia carvifolia	0.13	0.13	0.13	-	-
	Aster indicus	-	0.02	0.02	-	-
	Artemisia argyi	-	0.05	0.02	0.02	0.04
	Erigeron annuus	-	-	0.02	0.02	0.02
	Erigeron acer	-	-	-	0.04	0.04
	Bidens pilosa	-	0.03	-	0.03	0.02
Amaranthaceae	Alternanthera philoxeroides	0.08	-	-	-	-
Oxalidaceae	Oxalis corniculata	-	0.03	0.06	0.07	0.08
Apiaceae	Cnidium monnieri	-	-	0.08	-	-
Total: 7	15	6	11	12	9	9

Note: “-” means that this species has not appeared within the quadrat. “I”, “II”, “ III “, “IV”, and “V” correspond to the five coal gangue mountains 3, 6, 8, 10 and 13 years after planting, respectively.

, species composition and importance values of the five coal gangue mountains in the quadrats were different. With more time after planting, the number of species in the quadrats first increased and then decreased. The number of species was lowest in the coal gangue mountain 3 years after planting; there were only six species, belonging to five families and six genera, including Gramineae, Lamiaceae, Ranunculaceae, Asteraceae, and Amaranthaceae. Four species had importance values greater than 0.10, including V. zizanioides (0.40), Clinopodium chinense (0.18), Zoysia japonica (0.15), and Artemisia carvifolia (0.13), respectively. The sum of these four species made up more than 85% of the important value, which was dominated by V. zizanioides. There were eleven species in the coal gangue mountain 6 years after planting, belonging to five families and ten genera, including Gramineae, Asteraceae, and Lamiaceae, and three species with importance values greater than 0.10, including V. zizanioides (0.37), M. floridulus (0.21), and Artemisia carvifolia (0.13). The sum of these three species made up more than 70% of the importance value, which was jointly dominated by V. zizanioides and M. floridulus. The number of species was highest in the coal gangue mountain 8 years after planting; here, there were twelve species belonging to six families and eleven genera, and three species with importance values greater than 0.10, including V. zizanioides (0.34), M. floridulus (0.23), and A. carvifolia (0.13). The sum of these three species made up more than 70% of the importance value, which was jointly dominated by V. zizanioides and M. floridulus. The species composition in the coal gangue mountain 10 and 13 years after planting was the same, including nine species, belonging to four families and eight genera, such as Gramineae and Asteraceae. There were two species with importance values greater than 0.10 in both quadrats, including V. zizanioides and M. floridulus, and the sum of these two species jointly dominated and made up about 70% of the importance value.

3.2. Changes of Plant Community Diversity in Different Coal Gangue Mountains

As shown in Figure 2, except for Margalef richness index, the Simpson diversity, Shannon–Wiener diversity, and Pielou evenness indices of the five coal gangue mountains all showed significant differences. With more time after planting, the α-diversity firstly increased and then decreased, and the maximum values all appeared in the coal gangue mountains 8 years after planting.

From the perspective of β diversity (

Table 4. β diversity matrix of plant communities in different coal gangue mountains.

Measures	Coal Gangue Mountains	II	III	IV	V
βc	I	8.50	8.50	11.00	11.50
	II		7.00	10.50	11.00
	III			7.00	7.50
	IV				4.00
βws	I	0.49	0.49	0.64	0.67
	II		0.33	0.55	0.57
	III			0.35	0.41
	IV				0.26
Cj	I	0.35	0.35	0.18	0.20
	II		0.50	0.29	0.28
	III			0.48	0.42
	IV				0.59
Cs	I	0.51	0.51	0.30	0.33
	II		0.67	0.45	0.43
	III			0.65	0.59
	IV				0.74

Note: β_c: Cody index; β_ws: Whittaker index; C_j: Jaccard index; C_s: Sorenson index. “I”, “II”, “III”, “IV”, and “V” correspond to the five coal gangue mountains 3, 6, 8, 10, and 13 years after planting, respectively.

), the Cody and Whittaker dissimilarity indices showed a gradual decreasing trend from the coal gangue mountains I to V; the Cody dissimilarity index decreased from 8.50 comparing the coal gangue mountains 3 years and 6 years after planting to 4.00 comparing the coal gangue mountains 10 years and 13 years after planting. The Whittaker dissimilarity index decreased from 0.49 to 0.26 contrasting these coal gangue mountains at different years after planting, respectively. Compared with the other four coal gangue mountains different years after planting, the Cody and Whittaker dissimilarity indices increased gradually with post-planting years at the coal gangue mountain 13 years after planting; The Cody dissimilarity index increased from 4.00 comparing the coal gangue mountains 10 years and 13 years after planting to 11.50 comparing the coal gangue mountains 3 years and 13 years after planting, the Whittaker dissimilarity index increased from 0.26 to 0.67 contrasting these coal gangue mountains at different years after planting, respectively. The Jaccard and Sorenson similarity indices showed an opposite trend to the Cody and Whittaker dissimilarity indices. From the coal gangue mountains I to V, the Jaccard and Sorenson similarity indices showed an increasing trend in adjacent communities; the Jaccard similarity index increased from 0.35 comparing the coal gangue mountains 3 years and 6 years after planting to 0.59 comparing the coal gangue mountains 10 years and 13 years after planting, the Sorenson similarity index increased from 0.51 to 0.74 contrasting these coal gangue mountains at different years after planting, respectively. Compared with the other four coal gangue mountains, the Jaccard and Sorenson similarity indices decreased gradually with post-planting years at the coal gangue mountain 13 years after planting; the Jaccard similarity index decreased from 0.59 comparing the coal gangue mountains 10 years and 13 years after planting to 0.20 comparing the coal gangue mountains 3 years and 13 years after planting, the Sorenson similarity index decreased from 0.74 to 0.33 contrasting these coal gangue mountains at different years after planting, respectively.

3.3. Dynamic Changes of Community Succession in Different Coal Gangue Mountains

As can be seen from Figure 3, the height, coverage, density, and biomass of V. zizanioides in the above five coal gangue mountains showed a significant decreasing trend with more time after planting. The height, coverage, density, and biomass of V. zizanioides in the coal gangue mountain 13 years after planting decreased by 40.31%, 64.59%, 37.83%, and 80.05%, respectively, compared with those in the coal gangue mountain 3 years after planting. In contrast to V. zizanioides, the coverage, density, and biomass of M. floridulus significantly increased with more time after planting after its appearance in the coal gangue mountain 6 years after planting; among the above three indicators, plants in the coal gangue mountain 13 years after planting; increased by 56.48%, 55.81%, and 63.25% compared with those in the coal gangue mountain 6 years after planting, respectively.

According to the fitting relationship between the biomass of V. zizanioides, M. floridulus, and woody plants with each planting year (Figure 4), the dry biomass of V. zizanioides decreased sharply first in the visible period before gradually slowing. At 13.67 years after planting, the dry biomass of V. zizanioides decreased to the minimum of 29.42 g/m². The dry biomass of M. floridulus increased firstly and then decreased with the planting time; when the planting time was 11.59 years, the dry biomass of M. floridulus reached the maximum of 319.10 g/m², while when the planting time was 29.42 years, it decreased to 0 g/m². The dry biomass of woody plants increased linearly. According to the dry biomass fitting curve, the dry biomass of V. zizanioides was always the highest before the intersection point K₁ of biomass curves of V. zizanioides and M. floridulus (5.62, 173.77); this could be classified as the first succession stage (V. zizanioides community stage), with a duration of 0–5.62 years. The dry biomass of M. floridulus was always the highest from K₁ point to the intersection point K₂ of biomass curves of M. floridulus and woody plants (17.48, 178.40); this was the second succession stage (M. floridulus community stage), and the duration of this stage was 5.62–17.48 years. After K₂ (17.48, 178.40), the dry biomass of woody plants was always the highest, which was the third succession stage (woody plant stage).

4. Discussion

4.1. Influence of V. zizanioides on Community Characteristics in Different Coal Gangue Mountains

V. zizanioides is considered one of the most suitable pioneer plants for the reclamation of wasteland in mining areas [27]. It can quickly create a so-called “community environment” after establishment that is more suitable for the invasion and growth of other species. Studies have shown that V. zizanioides can improve the physical and chemical properties of tailings and generate large biomass, which is key in the pioneer community stage of reclamation of tailings wasteland [28]. V. zizanioides can reduce the erosion of soil by rain through interception of its aboveground parts, and improve the physical structure of soil through root elongation and fixation to conserve water and soil [29]. The plants improve the nutrient level of soil through their self-fertilization abilities, and improve the nutrient supply of soil to plants, forming a benign substance cycle [30], while the unplanted coal gangue mountain produced in 2002 is relatively fragile and almost bare, with few herbaceous plants (less than five species) and low coverage (less than 10%). Another example, it took 35~50 years for a few shrubs to emerge in Jixi natural coal gangue mountain in Heilongjiang province [31,32]. V. zizanioides was used in this study for reclamation and ecological restoration of a mining area, and was found to make vegetation rapidly restore the area and enter rapid succession after colonization. Species composition and structure of vegetation communities were optimized gradually [33], and the dominant species of communities was altered from V. zizanioides to M. floridulus over time, before a gradual community succession to woody plants as the dominant species. This is a result of the comprehensive effect of species diffusion and environmental screening [34], indicating that the substitution and change of community dominant species accelerated the process of vegetation restoration in the initial stage of vegetation restoration in the coal gangue mountain compared with natural succession [35].

The dynamics of diversity in coal gangue mountains is closely related to the restoration process of vegetation, and the temporal dynamics of species diversity can reflect the characteristics of the vegetation succession process to a certain extent [36]. With the extension of time past planting of V. zizanioides, its importance value showed a decreasing trend at temporal scale, while that of M. floridulus increased over time; this indicated that the dominance degree of communities in the coal gangue mountain was gradually occupied by the M. floridulus population, and the community structure has gradually changed. The plant community structure of the coal gangue mountain became relatively complex and gradually developed to a relatively stable stage of M. floridulus after planting of V. zizanioides. The α diversity index continuously increased during the 8 years after V. zizanioides planting, and then decreased as M. floridulus became the dominant species. This is probably because in the initial stage of V. zizanioides establishment, its density and coverage are both smaller, which cannot inhibit the immigration and growth of other herbaceous plants. The α diversity index reached its peak after 8 years of V. zizanioides establishment, as the dominant species were replaced by M. floridulus; shade plants and heliophilous plants also simultaneously appeared in coal gangue mountains, and mesophytic herbs gradually increased, therefore leading to the maximal diversity index. With community succession, the competitiveness of M. floridulus was stronger than that of V. zizanioides, and its density, coverage and biomass all increased rapidly, leading to the decrease or even disappearance of other herbaceous species in symbiosis with V. zizanioides and the subsequent significant decrease of the diversity. β diversity is mainly used to indicate the degree of habitat separated by species and habitat diversity in different regions [37]. The study on diversity changes in the restoration process of vegetation reclamation of land in the Antaibao mining area by Guo et al. [38] demonstrated that directly planting vegetation greatly accelerated the process of original vegetation succession, but that the community and natural vegetation succession showed convergence after three years; the main factors restricting the change of species diversity in the mining area were soil moisture and the succession time. In this study, the Cody and Whittaker dissimilarity indices of two adjacent coal gangue mountains showed a gradual decreasing trend from 3–13 years after planting, mainly because the immigration rate of species was fast and the succession rate of V. zizanioides community was also fast in the early stage of planting; with increasing time after planting, the species competition intensified and species immigration rate slowed down, and thus the succession rate also gradually decreased. The α diversity showed same pattern due to the fast species immigration rate and low species emigration rate, which resulted in more common species in the species composition, higher similarity among coal gangue mountains, and larger Jaccard and Sorenson similarity indices in communities in the late succession stage. This result was consistent with study of Zhang et al. showing that communities at different stages of vegetation restoration in Horqin, sandy land showed gradual increasing similarity among communities, and β diversity decreased gradually as succession progressed [39].

4.2. Influence of V. zizanioides on Community Succession in the Coal Gangue Mountain

The essence of community succession is the substitution of species, which is the result of the combined action of ecological factors such as environmental adaptation of species, evolution of community environment, and interspecific competition [40]. As for the succession law of coal mine wastelands, most scholars believe that the succession speed gradually slows down during the succession process, and that the restoration and succession of vegetation follows the general law of community succession. The species community composition trends from simple to complex and different species become dominant in each stage of succession; finally, the composition gradually tends to dynamic balance [41]. In this study, V. zizanioides was planted to accelerate the process of community succession in the early stage of reclamation. As the species composition in the study area was still dominated by herbaceous plants at this time and the diversity of herbaceous plants was higher than that of trees and shrubs, this indicated that the V. zizanioides community sequence was still in the early stage of succession. However, V. zizanioides was replaced as the dominant species by M. floridulus as time progressed, and shrubs of Rosaceae and trees of Betulaceae and Oleaceae also appeared; this indicated that coal gangue mountains in the study area were on a succession track towards zonal vegetation evergreen broad-leaved forest. It was predicted that this coal gangue mountain would progress to the third community stage (woody plant stage) after 17.5 years. The natural process of vegetation restoration can last for decades or even centuries [42]. In contrast to this, in the present study area, artificial restoration changed the direction and speed of community succession, shortening its restoration cycle.

According to the results of this study, the similarity in species composition increased with time after planting showing strong continuity and progression and indicating that the community succession rate is slowing down after the first decade of fast initial succession. Succession speed is fast in the initial stages but becomes slower as time progresses. In the succession sequence seen here, a relatively stable state with M. floridulus as the dominant species was reached. After duration for a period at this steady state, higher-level dominant plants (woody plants) immigrated into the community and disrupted the steady state; thus, the succession entered into the next stage (woody plant stage). In the succession process, plants change from annual to perennial, from herb to shrub, and the life history of community component species in the succession process becomes longer; the time for plant species to colonize, reproduce, and renew is also longer, and thus the change of vegetation is highest in the early stage but then relatively stable in the later stages [43]. The variation trend of the α diversity also showed that each index increased rapidly in the first eight years, but decreased significantly over the last five years. It is generally believed that the initial succession rate is faster, but then tends to slow down until a temporary stable state is reached. According to the appearance order of dominant species, the natural succession of vegetation in the study area went through the following stages: V. zizanioides stage, M. floridulus stage, and lastly, a woody plant stage.

5. Conclusions

In this study, the species composition, species diversity, species importance values, species height, coverage, and biomass of five successional stages after planting of V. zizanioides on coal gangue mountains were investigated. In total, we found 35 plant species belonging to 17 families and 32 genera across the five different coal gangue mountains representing different successional stages. With increasing time after planting of V. zizanioides on coal gangue mountains, its height, coverage, density, and biomass decreased, while M. floridulus increased. Alpha diversity indices increased in early successional stages, but decreased with increasing time after planting of V. zizanioides. In general, the successional stages became more similar with increasing time after planting. The succession process of coal gangue mountains could be divided into a V. zizanioides community stage (0–5.62 years), a M. floridulus community stage (5.62–17.48 years), and a woody plant stage (over 17.48 years). The results also imply that V. zizanioides is well adapted to the habitat of coal gangue mountain, and can occupy a dominant position in the initial succession stage. After about 3 years of planting, V. zizanioides community accelerated the process of vegetation restoration, M. floridulus gradually became the dominant species until 10 years after planting. In addition, the results imply that V. zizanioides is suitable to be a pioneer plant in the initial stage of ecological restoration of coal gangue mountains. In order to achieve a better effect of vegetation restoration and community succession, M. floridulus can be chosen to fill the void between the row spacing of V. zizanioides after 3 years. This method will accelerate the progress of artificial vegetation restoration and community succession in coal gangue mountains. By this, the successional stage of evergreen broad-leaved forests can be reached earlier.