Characteristics of Grassland Plant Community Change with Elevation and Its Relationship with Environmental Factors in the Burqin Forest Region of the Altai Mountains

Abstract

The change grassland plant communities demonstrate with elevation has been one of the hot issues in ecological research, and there remain many unsolved problems. In order to further elucidate the rules of grassland plant community change with elevation, this study took the Burqin forest area as a research object, using field survey, redundancy analysis and grey correlation analysis to comprehensively assess the characteristics of change in grassland plant communities with elevation and the relationship of this evolution with environmental factors. The results showed that (1) the numbers of species, community biomass, community cover and community densities of grassland plant communities showed an “M” pattern with the increase in elevation. There were significant changes in the importance values and dominance of plants at different elevations; with increasing elevation, grassland plants became primarily dominated by cold-tolerant and well-adapted perennials. (2) The similarity coefficients of grassland plant communities at different elevations ranged from 0.06 to 0.62, i.e., from very dissimilar to moderately similar. (3) As the elevation increased, the Margalef species richness index, Shannon–Wiener diversity index, Simpson dominance index and Alatalo evenness index all showed an “M” pattern trend. (4) The degrees of correlation between temperature and precipitation and community biomass and species diversity were at a high level, and these were the most important environmental factors affecting the biomass and species diversity of grassland plant communities in the Burqin forest area. The results of this study can provide a theoretical basis for the rational utilization of grassland resources and for the sustainable development of grassland ecosystems in the Burqin forest area.

1. Introduction

Grassland is the largest terrestrial ecosystem, with functions of climate regulation and water conservation and with high economic and ecological value [1]. The characteristics of grassland plant communities comprehensively reflect the balance and stability of grassland ecosystems, which is of great significance for establishing an in-depth understanding of the community composition, structure and sustainable development of grassland resources [2]. In a study of vegetation-community characteristics of alpine meadows in the East Qilian Mountains [3], it was found that the diversity and biomass of grassland plants were primarily affected by elevation. When summarizing the characteristics of community species-diversity with elevation, the change in species diversity with elevation mainly showed five forms [4], namely, positive correlation, negative correlation, irrelevance, intermediate elevation expansion, and “V” pattern. On the north slope of Motianling in Gansu [5], researchers found the biomass and species diversity of grassland to differ at varying elevations. With the change in elevation, the environmental factors also change [6]. In the Altai Mountains, at the grassland source of two rivers [7], it was found that longitude and latitude exert significant effects on the biomass and species diversity of grassland plant communities. In the three-river headwater region [8], it was found that grassland plant diversity and biomass were the highest at medium elevation, and there were positive correlations between plant community biomass and species diversity and soil moisture, organic carbon and total nitrogen. In Ningxia Natural Grassland [9], it was found that soil factor was the most important environmental factor affecting biomass and that average annual precipitation was the most important environmental factor affecting species diversity.

The Burqin forest area is located in the southwest area of the Altai Mountains, with rich grassland resources and obvious vertical differentiation. At present, some investigations have been carried out in the Burqin forest area [10,11,12], but there have been fewer studies on the change characteristics of grassland plant communities and the relationships these have with environmental factors. Therefore, this study takes the Burqin forest area as a research object in order to analyze the change characteristics of grassland plant communities with elevation and their relationships with environmental factors. This study will help in understanding more deeply the influence of elevation on grassland plant communities, as well as the relationships between species diversity and biomass and environmental factors.

2. Materials and Methods

2.1. Overview of the Study Area

The Burqin Forest Region (86°25′~88°06′ E, 47°22′~49°11′ N) is located in Burqin County, Xinjiang Uygur Autonomous Region, China. With a width of 85.5 km from east to west and a length of 87 km from north to south, the region covers a total area of about 3770 km². Burqin Forestry District is located in the southwestern foothills of the Altai Mountains, on the southern slopes of the Altai Watershed, bordering Altai City to the east, the low hills in the central part of Burqin County to the south, Habahe County to the west, Kanas Nature Reserve to the northwest, and Mongolia to the north. The terrain of the forest area moves from high in the north to low in the south, tilting from northeast to southwest, with the highest elevation at 4330 m and the lowest at 450 m. The highest forest area is located in the north and the lowest in the south. There are 24 rivers within its total area of 932 km; it has an average annual precipitation of 300–600 mm and an average annual temperature of about −3.7 °C, which is a typical temperate alpine mountain climate [13]. Grassland in the Burqin forest area shows an obvious vertical distribution, and the distribution of grassland types, from bottom to top, in the range of 1000 m to 2300 m above sea level are: desert steppe in the mountains, mountain steppe, mountain meadow steppe and mountain meadow [12]. The main soil types are light brown calcareous soil, chestnut calcareous soil, black calcareous soil and alpine meadow soil. The Burqin forest area is one of the first important areas designated for the implementation of China’s natural forest protection project [13], containing a large number of grassland plant species, including Taraxacum mongolicum, Eleusine indica and Imperata cylindrica.

2.2. Data Sources

In this study, according to the actual situation in the forest area of Burqin and based on the principle of comprehensively reflecting the characteristics of grassland plant community changes with elevation, we selected the range of elevation from 1000 m to 2300 m as the study area. The study area was divided into a total of 13 elevation zones according to 100 m elevation intervals, denoted by I~XIII, respectively. The basic information for each elevation zone is shown in

Table 1. Overview of the elevation zones.

Elevation Zone	Elevation Range	Grassland Types	Coverage (%)	Rh (%)	Aap (mm)	Aat (°C)	Aae (mm)
I	1000~1100 m	Desert steppe in the mountains	82	67.4	238.8	1.8	717.4
II	1100~1200 m	Desert steppe in the mountains	80	68.3	241.6	1.9	712.2
III	1200~1300 m	Desert steppe in the mountains	82	64.3	229.0	−0.3	712.0
IV	1300~1400 m	Desert steppe in the mountains	86	67.7	234.0	−0.1	662.0
V	1400~1500 m	Mountain steppe	89	66.1	278.0	0.0	639.8
VI	1500~1600 m	Mountain steppe	95	67.8	276.0	−0.1	636.0
VII	1600~1700 m	Mountain meadow steppe	84	71.0	285.2	−0.3	639.6
VIII	1700~1800 m	Mountain meadow steppe	49	71.3	291.8	−0.6	627.6
IX	1800~1900 m	Mountain meadow	74	69.9	320.2	−2.4	600.2
X	1900~2000 m	Mountain meadow	72	70.4	300.8	−1.8	566.2
XI	2000~2100 m	Mountain meadow	72	71.3	318.8	−1.4	588.0
XII	2100~2200 m	Mountain meadow	59	70.7	328.0	−2.4	547.4
XIII	2200~2300 m	Mountain meadow	39	70.7	337.2	−2.4	551.8

Throughout the paper, “Aap” is the average annual precipitation; “Aat” is the average annual temperature; “Aae” refers to the average annual evaporation; “Rh” refers to the average relative humidity over the years.

. Within each elevation band, one 20 m × 20 m sample plot was established if the elevation increased by 30 m (Figure 1). One 1 m × 1 m grass sample plot was set up at five locations, in the corners and center, in each sample plot (Figure 2). A total of three sample plots and 15 grass sample squares were set within each elevation band. Using relatively evenly distributed sample plots, we were able to cover more habitat types and environmental changes within the elevation range, allowing for a more comprehensive understanding of the diversity and species composition of grassland plant communities. A field survey of grassland plants was conducted in July–August 2022 to record data on vegetation characteristics such as height, cover and density for each sample plot. Meteorological data were collected from the World Climate Database (https://www.worldclim.org/) (accessed on 7 June 2023). After downloading the data, the “Sampling” tool in the ArcGIS 10.8 software was used to extract the climatic factors of monthly mean temperature, monthly mean relative humidity, monthly precipitation and monthly evapotranspiration at the sampling sites for the years 2000–2022, and multi-year averages were calculated.

2.3. Research Methodology

2.3.1. Vegetation Structure

(1) Community cover: The grassland plant community cover was determined using the square grid method [14]. Square frames with 200 small squares (1 cm × 1 cm) were made, and then the square frames were placed on top of the grass sample plots (the heights of the square frames were adjusted according to the height of the vegetation in the sample plots). The results were recorded as 1 when the vegetation was more than 1/2 of the area in the small square, and 0 when it was less than 1/2, and we took three consecutive measurements. Finally, the ratio of the number of squares with a result of “1” to the total number of squares in the square frame was calculated as the community cover.

(2) Community density: The counting method was used to determine the densities of different plant populations, and the sum of the densities of different plant populations was taken as the community density.

(3) Community height: Five individual plants of each species were randomly selected from the grass sample and measured vertically with a tape measure from the turf area to the highest point of the species in its natural upright state. Then, the average of the five heights was taken as the height of the plant [15].

(4) Community biomass (Agb): Using the mowing method, plants in each sample plot were mowed and the fresh weight of the plants was calculated and recorded as community biomass.

2.3.2. Species Importance Values and Dominance

Importance values and degrees of dominance are comprehensive measures of a species’ status and role in a community. In this study, the importance values (IV) and dominance degrees (SDR₄) of the grassland plant community species in the Burqin forest area were calculated on the basis of relative height, relative frequency, relative cover and relative density [16,17]. The formula is as follows:

$$\mathrm{IV}{\mathrm{i}}_{}=\left[\left(\frac{H_i}{\sum H}+\frac{F_i}{\sum F}+\frac{C_i}{\sum C}\right)/3\right]\times 100\%,$$

(1)

$$\mathrm{SDR}{4}_{\mathrm{i}}=\left[\left(\frac{H_i}{H_{max}}+\frac{F_i}{F_{max}}+\frac{C_i}{C_{max}}+\frac{D_i}{D_{max}}\right)/4\right]\times 100\%$$

(2)

In the formula, IV_i denotes the importance value of the ith species and SDR_4i denotes the combined dominance ratio of the ith species. H_i denotes the height of the ith species, F_i denotes the frequency of the ith species, C_i denotes the cover of the ith species, and D_i denotes the density of the ith species. H_max denotes the height of the species with the greatest height in the community, F_max denotes the frequency of the species with the greatest frequency in the community, C_max denotes the cover of the species with the greatest cover in the community, and D_max denotes the density of the species with the greatest density in the community.

2.3.3. Community Similarity Coefficient

The similarities between plant communities at different elevations were measured by means of the coefficient of community similarity, where the greater the coefficient of community similarity, the smaller the differences in vegetation composition between communities. The Sorensen similarity coefficient (SC) was used to calculate the similarity coefficients of grassland plant communities at all elevations in the Burqin forest area [18]. The formula is as follows:

$$\mathrm{SC}=\frac{2w}{a+b},$$

(3)

In the formula, SC denotes the plant community similarity coefficient, and w denotes the species shared by the two plant communities; a denotes the total number of species in the 1st plant community, while b denotes the total number of species in the 2nd plant community. Depending on the magnitude of the values of the community similarity coefficient SC, it can be divided into four intervals indicating different degrees of similarity [18]. It is extremely dissimilar when the SC is 0.00 to 0.25, moderately dissimilar when the SC is 0.25 to 0.50, moderately similar when the SC is 0.50 to 0.75 and extremely similar when the SC is 0.75 to 1.00.

2.3.4. Species Diversity Calculations

Species diversity indices can reflect community species’ richness and evenness, among other things. In this study, Margalef’s species richness index (S), Simpson’s dominance index (D), Shannon–Wiener’s diversity index (H), and Alatalo’s evenness index of the α-diversity indices (Ea) were used to analyze the species diversity of grassland plant communities at different elevations [19,20]. The formulae are as follows:

$$\mathrm{S}=(\mathrm{S}-1)/\ln N,$$

(4)

$$\mathrm{D}=1-\sum P_i^2$$

(5)

$$\mathrm{H}=-\sum P_i\ln P_i$$

(6)

$$\mathrm{Ea}=\left[1/\left(\sum P_i^2\right)-1\right]/\left[exp\left(-P_i\mathit{ln}{P}_i\right)-1\right]$$

(7)

$${\mathrm{P}}_{\mathrm{i}}=\frac{N_i}{N}$$

(8)

In the above formulae, S is the total number of species in the sample. N is the total number of individuals in the sample. N_i is the number of individuals of the ith plant. P_i is the ratio of the number of individuals of the ith plant to the total number of individuals.

2.3.5. Redundancy Analysis (RDA)

Redundancy analysis, which is a method based on constraint analysis, is primarily used to analyze the relationships between explanatory data and environmental variables. Environmental variables with greater explanatory power are selected from a set of environmental variables to reveal the reasons for changes in the data [21]. De-trending correspondence analysis (DCA) was performed using the Canoco 5.0 software, and redundancy analysis was performed if the analysis showed that the maximum value of the gradient was less than 3 in all 4 axes. If it was greater than 4, canonical correspondence analysis (CCA) was selected. If it was between 3 and 4, both methods of analysis could be used [22]. After de-trending the correspondence analysis, the results showed that the gradient lengths of the four sorting axes were less than 3 (

Table A1. DCA analysis of biomass, species diversity and environmental factors of grassland plant communities in the Burqin forest area.

Statistic	Axis 1	Axis 2	Axis 3	Axis 4
Eigenvalues	0.0045	0.0011	0.0002	0.0001
Explained variation (cumulative)	67.45	83.66	86.45	88.55
Gradient length	0.20	0.12	0.13	0.13
Pseudo-canonical correlation (suppl.)	0.9251	0.6469	0.6843	0.6637

). Therefore, the redundancy analysis method based on linear modeling was chosen in this study. Thus, the relationships between the biomass, species diversity and environmental factors of grassland plant communities in the Burqin forest area were analyzed.

2.3.6. Grey Correlation Analysis

Grey correlation analysis quantitatively describes and compares a system to determine its correlations by analyzing the shape of the reference and comparison series. (The data series reflecting the characteristics of the system is called the reference series, while the series consisting of factors that affect the characteristics of the system is called the comparison series.) We selected significant environmental factors with a greater degree of correlation with grassland plant community biomass and species diversity through redundancy analysis, and then the correlation degrees between the significant environmental factors and grassland plant community biomass and species diversity were calculated. In this study, the correlations between the biomass and species diversity of grassland plant communities and significant environmental factors in the Burqin forest area were calculated using the grey correlation method [23,24].

(1) First, we determined the reference and comparison series. Community biomass, species diversity and each significant environmental factor were set as a whole, community biomass (species diversity) was set as the reference series Y₀, and each significant environmental factor was included in the comparison series Y₁, Y₂, Y₃….

(2) Second, we performed data standardization. Because the measures of community biomass, species diversity, and significant environmental factors varied in the raw data, the raw data required standardization using the max–min normalization method.

(3) Third, we calculated the degree of association. Processing results based on max–min normalization of the original data, the absolute difference between community biomass (species diversity) Y₀ and the sum of the significant environmental factors Y_i (ΔG_i) was found. Then, the correlation coefficient (δG_i) was found, and the degrees of correlation (rG_i) were found and ranked by the correlation coefficient. The formulae are as follows:

$$\mathsf{\Delta }{\mathrm{G}}_{\mathrm{i}}\left(\mathrm{k}\right)=\left|Y_0\left(k\right)-Y_i\left(k\right)\right|,$$

(9)

$$\mathsf{\delta }{\mathrm{G}}_{\mathrm{i}}\left(\mathrm{k}\right)=\frac{\Delta min+\rho \Delta max}{\Delta G_i\left(k\right)+\rho \Delta max}$$

(10)

$${\mathrm{rG}}_{\mathrm{i}}=\frac{1}{n}{\sum }_{k=1}^n\delta G_i\left(k\right).$$

(11)

In the formulae, Δmin and Δmax denote the minimum and maximum values of the absolute differences of all comparison factors, respectively; ρ is the resolution factor, which usually takes a value from 0 to 1, and is 0.5 here; rG_i is the correlation between the comparison series and the reference series; n is the number of data in the comparison series. The criteria for evaluating the strength of association are shown in

Table 2. Correlation strength standards.

Correlation Degree	0 < rGi ≤ 0.35	0.35 < rGi ≤ 0.65	0.65 < rGi ≤ 0.85	0.85 < rGi ≤ 1.00
Relevance evaluation	Low	Middle	Strong	Extremely strong

.

2.4. Data Processing and Analysis

One-way ANOVA, regression analysis and correlation analysis were performed using the IBMSPSS statistics 25.0 software. Grey correlation analysis was performed via the DPS data processing system. Redundancy analysis (RDA) was performed using the Canoco 5.0 software. ArcGIS 10.8 and Origin 2018 software were utilized for mapping.

3. Results

3.1. Characteristics of Grassland Plant Community Structure at Different Elevations

3.1.1. Species Composition of Grassland Plant Communities at Different Elevations

A survey of grassland plant communities at all elevations in the Burqin forest area revealed that the combined communities contained 72 species of grassland plants from 34 families and 68 genera, dominated by Poaceae, Asteraceae, Ranunculaceae, Rosaceae and Fabaceae. The composition of grassland plants varied significantly at different elevations. As the elevation increased, the number of grassland plant species showed an overall trend of an “M” pattern (

Table 3. Species composition of grassland plant communities at different elevations.

Elevation Zone	Number of Families		Number of Genera		Number of Special
	Quantity	Percentage in Total Families (%)	Quantity	Percentage in Total Genus (%)	Quantity	Percentage in Total Special (%)
I	7	20.59	8	11.76	8	11.11
II	10	29.41	14	20.59	14	19.44
III	13	38.24	16	23.53	16	22.22
IV	15	44.12	26	38.24	26	36.11
V	14	41.18	16	23.53	24	33.33
VI	11	32.35	15	22.06	21	29.17
VII	11	32.35	17	25.00	19	26.39
VIII	8	23.53	9	13.24	9	12.50
IX	7	20.59	7	10.29	14	19.44
X	14	41.18	18	26.47	18	25.00
XI	8	23.53	10	14.71	13	18.06
XII	11	32.35	13	19.12	13	18.06
XIII	7	20.59	10	14.71	10	13.89

). The lowest number of grassland plant species was found at elevation zone I. There were eight species, accounting for 11.11% of the total number of species, Imperata cylindrica, Polygonum aviculare and Fragaria vesca being the most important. Grassland plant species were the most abundant at elevation zone IV, with 26 species, accounting for 36.11% of the total number of species, mainly Hydrocotyle nepalensis, Potentilla fruticosa and Geranium wilfordii.

By calculating the species importance values and dominance degrees of grassland plant communities at different elevations, we were able to select the top three plants when ranked in terms of species importance values and dominance degrees at each elevation (

Table 4. Species importance and dominance of grassland plant communities at different elevations.

Elevation Zone	Species Name	Life Form	Importance Value (Dominance)/%
I	Imperata cylindrica	P	52.2 (84.6)
	Fragaria ananassa	P	50.4 (67.2)
	Polygonum aviculare	A	46.7 (71.5)
II	Polygonum aviculare	A	43.3 (77.1)
	Medicago sativa	P	42.4 (66.3)
	Eleusine indica	A	40.4 (69.6)
III	Eleusine indica	A	53.5 (76.4)
	Gunnera perpensa	P	44.1 (42.6)
	Plantago asiatica	P	43.2 (80.3)
IV	Imperata cylindrica	P	54.2 (90.6)
	Polygonum aviculare	A	54.1 (81.5)
	Eleusine indica	A	54.2 (87.6)
V	Eleusine indica	A	52.1 (98.5)
	Rumex acetosa	P	45.5 (77.3)
	Geranium wilfordii	P	45.2 (56.1)
VI	Imperata cylindrica	P	53.2 (92.1)
	Eleusine indica	A	50.2 (83.3)
	Alchemilla japonica	P	48.2 (82.4)
VII	Equisetum arvense	P	47.5 (91.1)
	Ranunculus japonicus	P	46.3 (91.5)
	Leibnitzia anandria	A	44.6 (79.2)
VIII	Imperata cylindrica	P	45.2 (96.7)
	Eleusine indica	A	49.6 (93.5)
	Niphotrichum canescens	P	43.2 (50.7)
IX	Imperata cylindrica	P	47.9 (79.8)
	Polygonum aviculare	A	45.6 (85.4)
	Convallaria keiskei	P	45.2 (80.3)
X	Imperata cylindrica	P	43.1 (75.8)
	Eleusine indica	A	45.7 (79.8)
	Adoxa moschatellina	P	47.1 (81.6)
XI	Imperata cylindrica	P	60.5 (99.1)
	Gagea nakaiana	P	49.2 (96.4)
	Hydrocotyle nepalensis	P	48.4 (81.2)
XII	Gagea nakaiana	P	51.3 (87.6)
	Hydrocotyle nepalensis	P	43.6 (71.5)
	Achillea millefolium	P	46.1 (75.6)
XIII	Geranium wilfordii	P	42.6 (80.8)
	Ranunculus japonicus	P	48.7 (91.2)
	Gagea nakaiana	P	45.6 (66.3)

“P” represents perennial plants; “A” represents an annual plant.

). It can be seen through Table 3 and Table 4 that at lower elevations, plant species are more abundant and both perennials and annuals grow, with higher importance values and dominance of Eleusine indica, Polygonum aviculare and Imperata cylindrica. The plant species showed a decreasing trend with increasing elevation, and the grassland plants were primarily dominated by perennials, with Imperata cylindrica and Gagea nakaiana having the highest importance and dominance values.

3.1.2. Characterization of Grassland Vegetation Structure at Different Elevations

Vegetation biomass, density and cover showed an “M” pattern as elevation rose (Figure 3). Community biomass was highest at elevation zone VI and lowest at elevation zone VIII, with values of 201.1 g and 94.5 g, respectively. The differences between elevation zone VIII and elevation zones I, III, VII, X, and XII were significant (p < 0.05). Community cover was greatest at elevation zone VI and least at elevation zone XIII, with values of 95% and 39.3%, respectively. Elevation zone VI and elevation zones VIII to XIII were found to be significantly different upon comparison (p < 0.05). Community density was highest at elevation zone IV and lowest at elevation zone XIII. Elevation zone VIII and elevation zones II to VII differed significantly (p < 0.05). The height of grassland plant communities underwent irregular change with increasing elevation, with community heights being greatest at elevation zone IV and least at elevation zone IX. Elevation zone I and elevation zones II and IX differed significantly (p < 0.05).

3.2. Similarity Coefficients of Grassland Plant Communities at Different Elevations

By analyzing the similarities of grassland plant communities at different elevations in the Burqin forest area, it was found (

Table 5. Similarity coefficients of grassland plant communities at different elevations.

Elevation Zone	I	II	III	IV	V	VI	VII	VIII	IX	X	XI	XII	XIII
I	—	4	3	5	3	3	4	1	3	2	1	2	1
II	0.36	—	9	10	5	6	5	5	4	4	3	3	2
III	0.28	0.60	—	8	5	5	5	5	5	3	2	3	2
IV	0.29	0.50	0.38	—	9	9	8	7	8	8	4	6	4
V	0.35	0.33	0.31	0.48	—	8	7	6	5	8	3	2	3
VI	0.25	0.40	0.31	0.48	0.48	—	9	6	7	6	4	3	2
VII	0.32	0.32	0.30	0.37	0.38	0.55	—	8	7	8	4	3	3
VIII	0.10	0.37	0.34	0.46	0.36	0.41	0.53	—	6	4	5	3	5
IX	0.29	0.24	0.34	0.41	0.34	0.48	0.47	0.62	—	7	6	7	4
X	0.15	0.25	0.17	0.36	0.46	0.34	0.44	0.50	0.58	—	7	5	4
XI	0.06	0.25	0.15	0.22	0.23	0.31	0.30	0.52	0.52	0.48	—	5	4
XII	0.19	0.18	0.21	0.31	0.24	0.25	0.27	0.46	0.54	0.44	0.43	—	7
XIII	0.11	0.15	0.18	0.22	0.18	0.21	0.24	0.52	0.43	0.31	0.40	0.61	—

The lower left of the diagonal in the table is the community similarity coefficient (SC); the upper right is the number of shared species.

) that the similarity coefficients of each grassland plant community ranged from 0.06 to 0.62, i.e., from very dissimilar to moderately similar. The analysis revealed that the coefficient of similarity between communities in elevation zones IX and VIII was the largest, at 0.62, representing a moderate level of similarity. Elevation zones XI and I had the smallest coefficient of similarity, of 0.06, which represents a very dissimilar level. As can be seen from Table 5, the similarity index of grassland plant communities was higher for closer distances between elevation zones and, vice versa, the similarity index of grassland plant communities was lower. This result suggests that the dispersal and settlement of grassland plant communities are affected by elevation, and that there are changes in the composition of grassland plant communities when there is a change in elevation, resulting in the formation of different grassland plant communities.

3.3. Species Diversity Characteristics of Grassland Plant Communities at Different Elevations

We performed alpha-diversity index analysis of grassland plant communities at different elevations. The results showed (Figure 4) that the Margalef species richness index, Shannon–Wiener diversity index, Simpson dominance index, and Alatalo evenness index varied consistently, with all showing an “M” pattern with the increase in elevation. Margalef’s species richness index, Shannon–Wiener’s diversity index, Simpson’s dominance index and Alatalo’s evenness index were all minimized at elevation zone I, with values of 0.37, 1.42, 0.54 and 0.64, respectively. The Margalef species richness index was greatest at elevation zone VII, with a value of 0.98. The Shannon–Wiener diversity index was greatest at elevation zone VI, with a value of 2.28. Simpson’s index of dominance was greatest at elevation zone V, with a value of 1.43. The Alatalo uniformity index was greatest at elevation zone VI at 0.86. Regression analysis of the α-diversity index and each elevation zone separately revealed that, as opposed to the Alatalo evenness index, the Margalef species richness index, the Shannon–Wiener diversity index, and the Simpson dominance index fitted with each elevation zone had an R² greater than 0.5, which constituted a good fitting relationship. Neither the Margalef species richness index nor the Alatalo evenness index was significantly correlated (p > 0.05) with the different elevation zones. Both the Shannon–Wiener diversity index and Simpson dominance index were significantly correlated (p > 0.05) with differences in elevation zone.

3.4. RDA Analysis of Biomass and Species Diversity of Grassland Plant Communities with Environmental Factors

In RDA ranking diagrams, environmental factors are generally represented by hollow arrows. The length of the arrow connecting the lines represents the magnitude of the correlation between an environmental factor and the analyzed indicator. An angle > 90° between the arrows indicates a negative correlation between the two factors, while an angle < 90° indicates a positive correlation between the two. The results of the redundancy analysis of the biomass and species diversity of grassland plant communities using environmental factors showed that the first and second ordering axes explained 63.55% and 17.81%, respectively. That is, the two ordination axes combined explained a total of 81.36% of the relationships between grassland plant community biomass, species diversity and environmental factors, indicating that the ordination was effective. As can be seen in Figure 5, the Margalef species richness index, Shannon–Wiener diversity index, Simpson dominance index, and Alatalo evenness index were positively correlated with community cover, elevation, latitude, and multi-year mean relative humidity, and negatively correlated with multi-year mean precipitation, longitude, multi-year mean evapotranspiration, and multi-year mean air temperature, respectively. Community biomass was positively correlated with mean multi-year evapotranspiration, mean multi-year air temperature, mean multi-year precipitation, and longitude and latitude. Conversely, it was negatively correlated with mean multi-year relative humidity, elevation, and community cover. Elevation, community cover, average multi-year precipitation, average multi-year evapotranspiration, and average multi-year air temperature had longer connecting lines than did other parameters. This indicates that these five environmental factors are correlated with the biomass and species diversity of grassland plant communities to a greater extent than the others and can thus better explain the relationship between the biomass and species diversity of grassland plant communities and environmental factors.

3.5. Analysis of Biomass and Species Diversity of Grassland Plant Communities in Relation to Environmental Factors

Based on the results of RDA analysis, five environmental factors, namely, elevation, community cover, average multi-year precipitation, average multi-year evapotranspiration and average multi-year air temperature, were selected as the comparison series, respectively, for use in this study. Community biomass, Margalef’s species richness index, Shannon–Wiener’s diversity index, Simpson’s dominance index, and Alatalo’s evenness index were used as the reference series, respectively. We performed grey correlation analysis to obtain and rank the relationships between the biomass and species diversity of grassland plant communities and the environmental factors, respectively (

Table 6. Correlation between biomass, species diversity, and significant environmental factors of grassland plant communities.

Environmental Factors	Agb	S	H	D	Ea
	Correlation Degree	Correlation Degree	Correlation Degree	Correlation Degree	Correlation Degree
Asl	0.496 (5)	0.591 (5)	0.569 (4)	0.595 (5)	0.585 (4)
Cov	0.507 (4)	0.638 (3)	0.547 (5)	0.620 (4)	0.626 (3)
Aap	0.701 (2)	0.721 (1)	0.753 (1)	0.736 (2)	0.794 (1)
Aat	0.816 (1)	0.664 (2)	0.716 (2)	0.737 (1)	0.782 (2)
Aae	0.688 (3)	0.638 (4)	0.699 (3)	0.688 (3)	0.584 (5)

Numbers in parentheses indicate the order of correlation of environmental factors.

). The results of the correlation analysis showed that the correlation between different environmental factors and the biomass and species diversity of grassland plant communities varied. As can be seen from Table 6, the correlations between each environmental factor and the biomass and species diversity of grassland plant communities ranged from 0.496 to 0.816, values that were above the medium correlation strength. The correlation between average multi-year precipitation, average multi-year air temperature and the biomass and species diversity of grassland plant communities ranged from 0.65 to 0.85. This was a strong correlation, playing a key factor in plant growth and distribution. Overall, all environmental factors influenced the biomass and species diversity of grassland plant communities, but the effects of air temperature and precipitation were far more obvious. This indicates that hygro-thermal conditions limit the size of the biomass and the species diversity of grassland plant communities.

4. Discussion

4.1. Effects of Different Elevations on Grassland Plant Community Structure

In nature, the structural characteristics of grassland plant communities are visual representations of the ecological statuses of grasslands in different habitat conditions [25] and play an important role in reflecting the quality of grassland ecosystems. The number of species (number of families, number of genus and number of species), community biomass, community cover and community densities of grassland plant communities showed an “M” pattern of change with increasing elevation, all gradually reaching the highest point and then rapidly decreasing to the medium elevation level. The trend of the vertical distribution of grassland vegetation at middle elevations is consistent with the “mid-domain effect” hypothesis [26,27]. That is, at low elevations, there is sufficient heat for grass plants to grow, but water is in short supply. At high elevations, where moisture is abundant but temperatures are low, heat becomes a limiting factor for growth [28]. At medium elevations, although the absolute amounts of water and heat are not the highest, conditions such as temperature and precipitation are more favorable, the water–heat mix is optimal, and the availability of resources is at its optimum, promoting the growth of more species. Therefore, the number of grassland plant community species and the quantity of community biomass, as well as the community and community density, all gradually reached their highest points at a medium level of elevation, which is consistent with the results of the study conducted on the east slope of Taihang Mountain [29]. However, these results differed from the conclusion of the study on the Northern Tibetan Plateau that found a linear decrease in grassland plant community cover and biomass with elevation [30]. This effect may be due to the fact that the elevation of the present study area is around 1000–2300 m, while the selected study area in the northern Tibetan plateau had an elevation of 4000 m. Differences in the selected elevational scales resulted in different environments within the habitat, which led to different conclusions in this study from those of previous studies. However, the number of grassland plant community species, community biomass, community cover, and community density peaked at moderate elevations and then rapidly declined, possibly due to anthropogenic grazing in elevation zone VIII. Lower elevations are characterized by higher temperatures, arid climates, and more anthropogenic impacts that are not conducive to the growth of pasture grasses. The lower temperatures, harsh climate, and potential for severe rainfall and snowfall at high elevations make pasture survival more difficult, rendering it challenging for livestock to find edible plants [31]. Medium-elevation areas have a suitable and stable climate and are less influenced by human activities, as well as being endowed with more fertile soils. These conditions are favorable for the growth and development of pasture grasses [32], and so summer pastures were set up at an elevation of 1700–1800 m (within the elevational zone VIII) in the Burqin forest area [33]. However, during the grazing process, the number of species, community biomass, community cover and community densities of grassland plant communities may gradually show a decreasing trend due to the preference nature of livestock, continuous trampling [34], competition among plants [35], or overgrazing [36].

The importance and dominance of species at different elevations varied significantly, which also held true for the same species at different degrees of elevation. At lower elevations, plant species were more diverse and dominated by perennials and annuals, with Eleusine indica, Polygonum aviculare and Imperata cylindrica prominent. As the elevation increased, the temperature gradually decreased, whereas the precipitation and the degree of soil erosion gradually rose [37]. This resulted in many annuals not being able to adapt to the environmental conditions at this time, and limited renewal [38]. Thus, there was a gradual decrease in annuals and a gradual increase in hardy and well-adapted perennials, which is the same as the conclusion of a study at Jiajin Mountain [39]. Overall, when the elevation changes, the temperature, humidity and other environmental factors also change [40]. At higher elevations, hygro-thermal conditions are also harsher, and many species are less able to adapt and compete [41]. Therefore, there are obvious differences in species importance values and dominance degrees of grassland plant communities among different elevations.

4.2. Effects of Different Elevations on the Similarity of Grassland Plant Communities

The similarity coefficient of grassland plant communities is an important indicator of heterogeneity among communities and can also reflect the degrees of similarity of different grassland plant communities’ habitats [42]. Sorensen’s similarity coefficient analysis showed that elevation zone IX and community VIII had the largest similarity coefficient (0.62), while elevation zone XI and community I had the smallest similarity coefficient (0.06). This suggests that population dispersal and settlement are influenced by elevation, resulting in large habitat differences between elevations [43]. This contrasts with studies in the Maiji Mountains that found small differences in plant communities between elevations [44]. This distinction occurred due to differences in the number and span of elevational belts selected in the present study area and in the Maiji Mountains study area, which led the study to draw different conclusions. From Table 5, it can be seen that the closer the distance between elevation zones, the higher is the grassland plant community similarity index. The farther the distance between elevation zones, the lower the grassland plant community similarity index. This conclusion is the same as that seen in a study of the Xiaoxing’an Mountains, which found higher similarity coefficients of Acanthopanax senticosus communities at closer elevations [45].

4.3. Effects of Different Elevations on Species Diversity of Grassland Plant Communities

Through the investigation and analysis of the species diversity of grassland plant communities at different elevations, we found that the Margalef species richness index, the Shannon–Wiener diversity index, the Simpson dominance index, and the Alatalo evenness index showed an “M” pattern of change as the elevation increased, and that all of them showed a decreasing trend at elevation zone VIII. Further, all revealed a decreasing trend at elevation zone VIII. This was different from the findings of a study at Helan Mountain [46], which found that species diversity showed a “single-peak” pattern with increasing elevation, and a study in the Baotianman National Nature Reserve [47], which found that species diversity showed a decreasing trend with increasing elevation. This trend in species diversity may be due to the fact that at lower elevations, where environmental conditions are relatively favorable, species diversity is lower due to the loss of certain species as a result of excessive disturbance from human activities [48]. At high elevations, the temperature gradually decreases, precipitation gradually increases, and the environmental conditions become harsher, resulting in many species not being able to grow and fewer types of plants surviving, and therefore lower species diversity [49]. The impact of human interference is smaller at the medium elevation than at the low elevation, and environmental factors such as precipitation, temperature and light are more suitable in such a setting than at high elevation, which is favorable for growing vegetation and a wide variety of grassland plants. Thus, Margalef’s species richness index, Shannon–Wiener’s diversity index, Simpson’s dominance index, and Alatalo’s evenness index gradually increased at a medium level of elevation [50]. However, elevation zone VIII also fell within the medium elevation range. The grazing of summer pastures in elevation zone VIII resulted in a gradual increase in the Margalef species richness index, the Shannon–Wiener diversity index, the Simpson dominance index, and the Alatalo evenness index at moderate elevations, followed by a rapid decline. The species diversity of plant communities varied at different grazing intensities. Under conditions of overgrazing, the intensity of plant feeding by livestock increases [51], and problems such as grassland soil erosion [52] and declining soil carbon content [53] are highlighted. Such damage reduces the regeneration capacity of grassland plants, thus leading to a gradual decrease in community species diversity [54]. Moderate grazing not only benefits plant growth, but also promotes increase in community species diversity, which is conducive to maintaining the stability and development of grassland ecosystems [55]. As can be seen from Figure 4, the species diversity of grassland plant communities showed a decreasing trend at 1700–1800 m above sea level, so the authors hypothesize that there might be overgrazing in the summer pastures at 1700–1800 m in the Burqin forest area.

4.4. Biomass and Species Diversity of Grassland Plant Communities in Relation to Environmental Factors

Changes in the biomass and species diversity of grassland plant communities are the result of the interaction of many factors, and elevation, slope, soil and anthropogenic disturbances all affect the growth and development of grassland plants [56,57,58]. Elevation better integrates changes in different environmental conditions such as temperature, humidity and precipitation relative to other factors [59]. In montane forests, the role of elevation in plant community biomass and species diversity is very important [60]. When the elevation increases, the temperature shows a gradual decreasing trend, the rainfall shows a gradual increasing trend, and the water–heat combination shows different changes with elevation [61]. Taken together, hygro-thermal conditions were the main limiting factor for the growth and development of grassland plants in the Burqin forest area and had significant effects on the magnitude of biomass and species diversity of the plant community, a conclusion that is the same as that reached in a study in the Qilian Mountains [62]. However, it is different from the conclusion that precipitation is the main influence on the species diversity and biomass of grassland vegetation communities, as argued in a study on the Alxa Plateau [63]. It is possible that precipitation is the main limiting factor in the Alxa Plateau because of its long sunshine hours and high temperatures and evaporation.

5. Conclusions

By analyzing the change characteristics of grassland plant communities in the Burqin forest area on the basis of elevation and their relationships with environmental factors, the following conclusions were drawn: (1) With the increase in elevation, the number of species, community biomass, community cover and community densities of grassland plant communities showed the trend of an “M” pattern. There were also significant changes in the importance and dominance of grassland plant communities at the same elevation; as the elevation increased, the grassland plants were dominated by cold-tolerant and well-adapted perennials. (2) The similarity coefficients of grassland plant communities at different elevations in the Burqin forest area ranged from 0.06 to 0.62, moving from extremely dissimilar to moderately similar, with higher similarity indices for grassland plant communities closer to each other relative to elevation zones, and lower similarity indices for grassland plant communities farther away from each other relative to elevation zones. Elevation had a significant effect on the spread and settlement of plant populations. (3) The species diversity of grassland plant communities at different elevations differed significantly, and the Margalef species richness index, the Shannon–Wiener diversity index, the Simpson dominance index, and the Alatalo evenness index showed an “M” pattern of change as the elevation increased. (4) Environmental factors such as elevation, community cover, mean multi-year evapotranspiration, mean multi-year precipitation and mean multi-year air temperature all affected grassland plant community biomass and species diversity. Temperature and precipitation were the most important environmental factors affecting the biomass and species diversity of grassland plant communities in the Burqin forest area.