**1. Introduction**

A mountain is an area with certain elevations, slopes, and relative heights that are a reflection and condensed point of the gradients in natural geographical and ecological features [1]. Due to relatively low human disturbance, mountains provide habitat and shelter for terrestrial biological species [2]. Mountains also represent the most abundant unit of biodiversity on Earth and are key areas for the conservation of biodiversity [3,4]. Mountain areas are extremely sensitive to climate change [5]. The response of mountain ecosystems to climate change is an increasing focus of global change research [6,7].

Biological groups form cluster patterns of vertical and horizontal gradients. Kattan et al., suggested that clustering patterns in dendrograms formed two major patterns of differentiation of the biological groups in Colombia: one horizonal and one elevational [8]. The distribution of vegetation has obvious patterns of horizontal and vertical zonality [9]. This implies that the composition of vegetation varies with altitude [10,11]. Multiple environmental factors drive the altitudinal zonality of vegetation [12]. Natural factors, such as

**Citation:** Liang, H.; Fu, T.; Gao, H.; Li, M.; Liu, J. Climatic and Non-Climatic Drivers of Plant Diversity along an Altitudinal Gradient in the Taihang Mountains of Northern China. *Diversity* **2023**, *15*, 66. https:// doi.org/10.3390/d15010066

Academic Editors: Lin Zhang and Jinniu Wang

Received: 28 October 2022 Revised: 18 December 2022 Accepted: 21 December 2022 Published: 5 January 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

climate, geomorphology, and hydrology, drive obvious changes in altitudinal gradients from the bottom to the top of mountains [13,14]. Generally, changes in a mountain landscape along elevation gradients are 1000 times higher than that in horizontal gradients [15], which can be understood to mean that the landscape change on the vertical scale condenses the change on the horizontal scale. Distribution models of plant species diversity in mountain areas are increasingly used in ecological community research. Such models lead to a better understanding of the mechanism of maintenance of biodiversity and altitudinal change in mountain vegetation [16,17].

Research is still not conclusive about the primary factors that determine the patterns of biodiversity on Earth [18,19]. The gradients of diversity patterns at a large scale have been explained mainly by temperature, productivity, water availability, and geographical area [20]. The study of the altitudinal patterns of species diversity in mountain plant communities is critical for the understanding of zonal variations in vegetation along elevation gradients [21]. Plant diversity varies with altitude gradients [22], showing a variety of altitudinal patterns [23,24]. In the study of altitudinal patterns of plant diversity in mountain terrains, the α-diversity (such as the species richness, Shannon–Weiner index and Simpson index), β-diversity (such as the Jaccard index, Cody index) and γ-diversity indices are widely used [23–25]. The β-diversity index best reflects a community composition and turnover under an environmental gradient [25,26], which is critical for biodiversity conservation [27,28].

The rapid development of quantitative ecological methods and computer technology are innovatively changing data processing and analysis of biodiversity [29]. Studies suggest that surveys in mountain regions provide valuable insights into biological conservation [30,31]. While the relationship between the patterns of biodiversity and elevation gradients depends on various environmental variables [32], altitudinal patterns of mountain plant communities in different geographical regions differ [33–35]. Studies on the process of the development of altitudinal patterns of plant diversity in mountain regions are also attracting much attention [6].

The Taihang Mountain range is a transitional zone from low-elevation plains to highelevation plateaus, and it is an important ecological barrier to the economic circle of Beijing– Tianjin–Hebei in northern China [36]. Because of the special geographical location of the Taihang Mountain range, it plays an important role as a windbreak, in sand-fixing, and in water conservation. Furthermore, it is a transition zone from economically developed regions to undeveloped regions, where the natural ecosystems are cross-distributed. The Taihang Mountain range is extremely ecologically sensitive. Serious damage to the natural environment and biodiversity has been caused by historical development, which led to serious soil and water loss, frequent droughts, and other significant environmental problems [37]. After decades of overexploitation, along the elevational gradient, the vegetation presents a different distribution pattern. As vegetation provides most of the ecosystem services, the elevational pattern is a reference for evaluating the effects of natural and anthropogenic factors on the vegetation in the Taihang Mountain range. Therefore, the objective of this study was to explore the impact of climate related factors and non-climatic factors on the vertical pattern of plant diversity in the Taihang Mountains.

### **2. Materials and Methods**

### *2.1. Study Area*

The Taihang Mountains (34◦36 –40◦47 N, 110◦42 –116◦34 E) are a highly heterogeneous geological setting, spreading from the northeast to southwest in northern China (Figure 1). The mountain range acts as a natural boundary between the North China Plain and the Loess Plateau. The altitude across the Taihang Mountains decreases from northwest to southeast, with the highest elevation of 2882 m in the north.

A temperate continental monsoon climate prevails in the study area. From 2009 to 2017, the annual mean temperature was 8.91 ◦C, and the annual mean precipitation was 529 mm. Both temperature and precipitation increase from the northwest to the southeast. The Taihang Mountains are also known for their expansive biodiversity in northern China. Warm temperate deciduous broad-leaved forest is the dominant vegetation type in the central Taihang Mountain region.

The eastern slopes (sunny slopes) of the Taihang Mountains are steeper than the western slopes (shady slopes); therefore, the vegetation has a more obvious vertical change on the sunny slopes. The survey plots in this study were mainly on the sunny slopes in the central Taihang Mountain region, where the highest peak (Tuoliang) is 2282 m. The central Taihang Mountain region is divided into three ecological zones: a hilly zone (<500 m), mid-mountain zone (500–1500 m), and sub-alpine zone (>1500 m) [38]. Each zone is characterized by a different set of biodiversity and ecosystem services.

**Figure 1.** A map depicting the location of Taihang Mountains in northern China (left panel) and an expanded map depicting elevation and sampling sites in the study area.

### *2.2. Field Survey*

This field survey was carried out during the growing season from May 2017 to October 2019 in Tuoliang National Nature Reserve, which is located in the central part of the Taihang Mountain range. A total of 480 survey plots (including tree, shrub, and herb plots) were established at 16 elevations (100 m, 200 m, 300 m, 400 m, 500 m, 600 m, 700 m, 900 m, 1100 m, 1300 m, 1500 m, 1700 m, 1800 m, 1900 m, 2100 m, and 2200 m). At each elevation, five tree plots (10 m × 10 m for each plot), 10 shrub plots (5 m × 5 m), and 15 herb plots (1 m × 1 m) were established, and the species and individual numbers of trees, shrubs and herbs were recorded in each plot. For soil water content (VWC%) and soil pH analyses, three soil samples at a 0–20 cm depth were collected by a cylindrical soil sampler of 5 cm diameter at the same time as the field survey.

### *2.3. Environmental Data Source*

The climatic factors (temperature and precipitation) were collected for the period of 2008–2017 from 101 automatic weather stations installed across the Taihang Mountain range. The temperature and precipitation of each survey plot were derived by kriging interpolation. Non-climatic factors in this study, including elevation, slope, aspect, human footprint index (Hfp), human influence index (Hii), net primary productivity (NPP), human population density, soil pH, and soil water content (VWC%) were derived by synchronizing with the field survey region. The elevation of each plot was measured by GPS, and slope and aspect were detected by a gradiometer. Net primary productivity (NPP) was derived from MOD17A3 data released by the University of Montana, USA (http://ipdaac.usgs.gov, accessed on 28 October 2022). Hfp and Hii were derived from the Socioeconomic Data and Applications Center of NASA (SEDAC, https://sedac.ciesin.columbia.edu/, accessed on 28 October 2022). Human population density is often used as an indicator of vegetation disturbance [39]. The population density was obtained from the Geographical Information Monitoring Cloud Platform maintained by China. The datasets were interpolated for each sampling plot using ordinary kriging. Descriptive statistics of the climatic and non-climatic factors along the altitudinal gradient in the Taihang Mountain study area are given in Table 1.

**Table 1.** Descriptive statistics of climatic and non-climatic factors along the vertical gradient of Taihang Mountain study area in northern China.


Notes: Climatic factors: temperature and precipitation; non-climatic factors: factors in addition to temperature and precipitation; temperature: annual mean temperature from 2008 to 2017; precipitation: annual precipitation from 2008 to 2017; Hfp: human footprint index; Hii: human influence index; p in p/km<sup>2</sup> denotes persons; pH: soil pH; VWC%: soil water content to 20 cm depth.

### *2.4. α-Diversity and β-Diversity Indices*

The α-diversity represents species richness within a community. The diversity indices at different altitudes were calculated in terms of the plot survey. The richness index represents the number of species in the sampling plots. The formulas for the Shannon– Weiner index (H) and the Simpson index (D) are as follows:

Shannon-Weiner index:

$$\mathbf{H} = -\sum\_{i=1}^{s} P\_i \ln P\_i \tag{1}$$

Simpson index:

$$D = 1 - \sum\_{i=1}^{s} \left(\frac{n\_i}{N}\right)^2 \tag{2}$$

where *Pi* is the proportion of the *i*th individual to the total number of individuals, *n*<sup>i</sup> represents the number of individuals of the *i*th species, and *N* represents the number of individuals of all species in the community.

The β-diversity is often expressed as the ratio of regional (γ-diversity) to α-diversity, and it is often measured as species turnover between different communities. In this study, we used the Jaccard index and Cody index to explore the traits of plant community succession along the altitudinal gradient. While the Jaccard index represents the similarity of different communities and quadrats, the Cody index represents the turnover rate of species along an environmental gradient. The indices are calculated as follows:

Jaccard index:

$$\text{CJ} = \frac{c}{a+b-c} \tag{3}$$

Cody index:

$$\beta\_c = \frac{\lg(H) + l(H)}{2} = \frac{a + b - 2c}{2} \tag{4}$$

where *a* and *b* denote the number of species in two communities, *c* denotes the number of species shared by the two communities, *g*(*H*) is the number of species increasing along the gradient (*H*), and *l*(*H*) is the number of species lost along the gradient.

### *2.5. Data Analysis*

Based on the sample data and remote sensing image data for the Taihang Mountains, climatic and non-climatic factors were processed and analyzed by SPSS 23 and mapped in ArcGIS 12.2. Ordination analysis is often used to explain variations in data in relation to species and area [40]. To investigate the distribution patterns of plant diversity and lifeform groups in the survey plots, principal components analysis (PCA) was used to analyze the altitudinal gradient. The influences of climatic and non-climatic factors on species altitudinal distribution were evaluated using canonical correspondence analysis (CCA) in R 3.4.5, and contributions of climatic and non-climatic factors to the plant altitudinal pattern were estimated with partial CCA in R 3.4.5. To verify the significance of environmental factors and the plant species altitudinal distribution, a Monte Carlo permutation test was performed in CANOCO 4.5.

### **3. Results**

### *3.1. Altitudinal Distribution of A-Diversity in Plants*

Based on the field survey, 54 vascular plant species were recorded in the hilly zone, belonging to 32 families and 49 genera; 103 species in the mid-mountain zone, belonging to 47 families and 88 genera; and 58 species in the sub-alpine zone, belonging to 21 families and 48 genera. As the overall elevation was not very high (with the lowest altitude of 0 m and the highest peak of 2282 m in the central Taihang Mountain region), there was not an obvious altitudinal spectrum of vegetation in this region.

Generally, the vascular plants (including trees, shrubs, and herbs) had the same altitudinal pattern, in which the number of plant species increased with increasing elevation. At elevation ranges of 600–900 m and 1500–1900 m, the plant richness (Figure 2d), Shannon-Wiener index (Figure 2a) and Simpson index (Figure 2b) had two peak intervals, and then began to decrease with increasing elevation. The distribution of plant diversity in the central Taihang Mountain region was relatively complicated in terms of the elevation gradient.

**Figure 2.** Indices of plant diversity along the altitudinal gradient in the central Taihang Mountain region, northern China.

Quantitative measurement is needed to show differences in communities [41]. PCA analysis (Figure 3) indicated that plant species richness was mainly concentrated in the mid-elevation zone (sites 5–14). Individual plants, species of trees and herbs, and the richness and Shannon–Weiner indices were highest in the area covering sites 5–14, which meant that plant richness was highest in the mid-elevation zone. Shrubs mainly occurred in the low elevation zone, in the area covering sites 1–4.

**Figure 3.** PCA ordination showing the characteristics of altitudinal distribution of plants in the central Taihang Mountain region.

### *3.2. B-Diversity of Plants along the Altitudinal Gradient*

Figure 2 shows that the Shannon–Weiner index, Simpson index and Cody index had coincident altitudinal patterns of plant richness, implying a relatively high plant diversity concentrated in the mid-elevation zone. The patterns of the Jaccard index indicated that the sampling plots in the mid-elevation range had a high similarity in plant community structure.

Low β-diversity can lead to low species turnover rates [42]. The Cody index represents the rate of species turnover between communities, with the highest species turnover always taking place in pioneer and mountain species [43]. In this study, the Jaccard index showed the opposite altitudinal patterns to the richness, Shannon–Weiner index and Simpson index, while the Cody index showed a consistent distribution trend, implying that the altitudinal gradient with lower community similarity had a higher species turnover rate of plant communities.

### *3.3. Relationships between Plant Diversity and Environmental Factors along the Altitudinal Gradient*

In predicting species distribution, canonical correspondence analysis (CCA) is widely used [44]. CCA ordination (Figure 4) showed that the driving factors with positive effects on tree, shrub, and herb richness were soil water content (VWC%), precipitation, and pH. The factors with negative effects on plant richness were temperature, Hii, Hfp, and population density. Slope and pH had the smallest effects on altitudinal distribution of plants in the central Taihang Mountain area.

**Figure 4.** CCA ordination showing the relationship between environmental factors and altitudinal plant patterns in the Taihang Mountain study area, northern China.

### **4. Discussion**

### *4.1. Plant Diversity Pattern along the Altitudinal Gradient*

Based on the field survey and species identification, there were 54 species of vascular plants in the hilly zone, belonging to 32 families and 49 genera; 103 species in the midmountain zone, belonging to 47 families and 88 genera; and 58 species in the sub-alpine zone, belonging to 21 families and 48 genera. This conformed with the richness distribution theory on "middle height expansion," consistent with studies of vine plants in other mountain regions [45–48].

In this study, trees were mainly found in the hilly and mid-mountain zones, with few trees in the sub-alpine zone. Shrubs and herbs were widely distributed from the low to the high elevation zones in the central Taihang Mountain region. The richness of herbs was higher than that of shrubs and trees. PCA ordination showed that plant groups were mainly concentrated in the mid-elevation range in the central Taihang Mountain region.

In the sub-alpine zone, temperate herbal plants were the main vegetation type. There was a higher proportion of annual herbaceous plants with more endemic species in this zone, indicating that biodiversity endured to a certain degree even in an area with intense human disturbance. While species richness decreased significantly at elevations above 2000 m, the community similarity index increased sharply. This implied that more common species of vascular plants were concentrated in the sub-alpine areas; therefore, there was relatively low species diversity.
