Effects of habitat factors on the plant diversity on naturally-restored wind farm slopes

Overview of the study area

The study area was located in four counties (or districts) of Hubei province China, namely Dongbao District, Dawu County, Tongshan County, and Sui County (Fig. 1). This area has a humid sub-tropical monsoon climate. The annual mean temperature, total sunshine hours, and total rainfall were 15.1 °C, 1950 h and 977 mm, respectively in Dongbao District; 15.3 °C, 2153 h and 1122 mm, respectively in Dawu County; 16.3 °C, 1845 h and 1500 mm, respectively in Tongshan County; and 15.4 °C, 2035 h and 968 mm, respectively in Sui County. The study area consists mostly of mountains and hills with a peak elevation of 684, 858, 954 and 1140 m in these four counties (or districts), respectively. The vegetation in the area is primarily evergreen broad-leaf forest and evergreen deciduous and broad-leaved mixed forest.

Selection of sample plots

Five mountainous wind farms were selected from four counties (or districts) of the Hubei province: Xianghe, Jiangjia, Damu, Fengming and Wuyue (Fig. 1). In order to obtain representative data, 21 naturally-restored slope plots were investigated. These plots were comprised mainly of rock excavation and soil filling. The habitat factors of all 21 slopes are illustrated in Fig. 1.

Selection of sample areas and investigation methods

This study focused on investigating the characteristics of side slope sample plots, including slope position, slope aspect, slope gradient, altitude, years since restoration and plant communities. Slope position can be divided into upper side slope (U), middle side slope (M) and lower side slope (L) from top to bottom, with each slope section’s height between 3 and 5 m. Slope aspect (with due north as 0°, degrees recorded clockwise) was divided into shady (337.6–67.5°), semi-shady (292.6–337.5°, 67.6–112.5°), semi-sunny (247.6–292.5°, 112.6–157.5°), and sunny (157.6–247.5°) slopes. The slope gradient and slope aspect were determined using hand-bearing a geological compass (ZN02; Jiesheng, Shenzen, China). The longitude, latitude, and altitude of the plots were determined using 3D GPS navigation (Xingyao X; Qianxun, Shanghai, China). The years since restoration was calculated starting from the completion of the restoration project in each area.

The sampling method for plant communities was as follows: in each sample plot, a 10 m × 10 m quadrat was set, and all the plants of tree species in the quadrant were investigated one by one. Each sample plot of tree species was then divided into four 5 m × 5 m quadrats, one of which was selected to investigate shrub species. Next, in each of the 5 m × 5 m quadrats, a 1 m × 1 m smaller quadrat was randomly chosen to investigate herbs species. The name, number of species and number of plants were recorded for each quadrat, and then the total number for the quadrant was calculate.

Data processing and analysis

Relative frequency reflects the distribution of a plant species in a certain region. It is the ratio of the number of a species in a quadrant to the total number in the quadrant. Relative abundance reflects the amount of a plant species in the overall plant communities. It is measured as the ratio of the number of a certain plant species to the number of plants of all species.

In this study, three indexes were used for assessing the plant community, the Shannon-Wiener diversity index (H′), Pielou’s species evenness index (J) and Margalef’s richness index (R), which were calculated, respectively, using the formulas as below:

${\displaystyle H^{\prime }=-\sum P_{i}ln{P}_i\left(i=1,2,\dots ,S\right)}$ ${\displaystyle J=H^{\prime }/lnS}$ ${\displaystyle R=\left(S-1\right)/lnN}$

where S represents the total number of plant species, N represents the total number of individual plants for all species, N_irepresents the numbers of each plant species, and P_i = N_i/N.

The data were analyzed using Microsoft Excel (version 2013) and a map was created by ArcGIS Desktop (Version 10.0). The figures were drawn by Origin (version 2021) and the stepwise regression analysis was performed using SPSS (version 20.0).

Species composition of plant communities on mountainous side slopes

The composition of the plant communities are as shown in Table 1. In the 21 sample plots, a total of 36 families, 54 genera,and 57 species were observed. The 14 families, 29 genera, and 30 species of herbs; 10 families, 12 genera, and 14 species of shrubs; and 12 families, 13 genera, and 13 species of trees identified. Of the 36 total families identified, Gramineae had 10 genera and 10 species, Compositae had five genera and six species, Rosaceae had four genera and six species, Liliaceae had two genera and two species, Pteridiaceae had two genera and two species, Juglandaceae had two genera and two species, and all other families had one genus and one species each. Gramineae was the largest family and accounted for 17.54% of the total species identified, followed by Compositae and Rosaceae, which each accounted for 10.53% of the total species.

The calculations of the relative frequency and the abundance of species found that, Cynodon dactylon had the highest relative frequency (42.86%) and relative abundance (38.82%),followed by Setaria viridis, Allium macrostemon, Ophiopogon bodinieri, Erigeron canadensis, Artemisia argyi, and Rumex acetosa. Among shrubs, Rubus lambertianus had the highest frequency (14.29%) and relative abundance (63.09%), followed by Lonicera japonica and Pyracantha fortuneana. Among trees, Lindera glauca was predominant in terms of relative frequency (80.95%), followed by Pinus massoniana (33.33%) and Platycarya strobilacea (14.29%).

Species richness under different habitat conditions

As shown in Fig. 2, under different slope positions, the total number of plant species decreased from the upper slope to the lower side slopes, with the lower side slopes having 4.00 times as many plants species as the middle side slopes and 4.67 times as many species as the upper side slopes. The species number of herbs, shrubs, and trees were all the most on lower side slopes with 33, 11, and 12 species identified on these slopes, and were all the least on upper side slopes with seven, two and three species identified, respectively. The total number of plant species also followed a descending order among semi-sunny, sunny, semi-shady, and shady side slopes. The total number of plant species identified on semi-sunny side slopes was 19.23%, 24.00%, and 93.75% higher than on sunny, semi-shady, and shady side slopes, respectively. The total number of herbs, shrubs, and tree species identified were all the most numerous on semi-sunny side slopes with the 18, seven, and six species identified, respectively. Species of herbs and shrubs were the least numerous on shady side slopes with seven and two species identified, respectively. The total number of plant species was the highest at the slope gradient between 30° and 50°, with this slope gradient having 16.7% more species than other slope gradients. Herbs were the most numerous on side slopes with slope gradients less than or equal to 30° with at 30 species of herbs identified on these slopes. Species of herbs and shrubs were the least numerous on side slopes with slope gradients less than 50° with 17 species of herbs and three species of shrubs identified on these slopes. Altitude also impacted the total number of plant species, with the largest number of species identified on slopes in this altitude range was 50.00% higher than on slopes between 500 and 700 m and 116.67% higher than on side slopes above 700 m in altitude. Species of herbs, shrubs, and trees were all the most numerous on side slopes less than or equal to 500 m in altitude with 21 species of herbs, 12 species of shrubs, and 6 tree species identified on these slopes. Species of herbs and shrubs were the least numerous on side slopes above 700 m in altitude with nine herb species and two shrub species identified at this altitude. The recovery period also impacted plant diversity with the total number of plant species increasing with the years since restoration. Species of herbs, shrubs, and trees were all the least numerous when there had been less than one year since restoration, with six herb species, two shrub species, and two tree species identified. Species of herbs and shrubs were the most numerous when there had been 5 or 6 years since restoration.

Vegetation diversity under different habitat conditions

As shown in Fig. 3, Shannon-Wiener diversity index (H′), Pielou’s species evenness index (J), and Margalef’s richness index (R) decreased with side slope position (P < 0.05), with the variation ranges at 0.84−1.51 on lower slide slopes, 0.60−0.75 on middle side slopes, and 0.82−1.64 on upper slide slopes. All the indexes peaked on lower side slopes with values 79.84%, 24.73%, and 100.36% higher than upper side slopes, respectively, and 30.95%, 2.32%, and 59.22% higher than middle side slopes, respectively. The Shannon-Wiener diversity index (H′) and Margalef’s richness index (R) were more influenced by slope aspect, with both indexes H′ and R reaching their minimums on sunny side slopes (1.20 and 1.26) and their maximums on semi-shady side slopes (1.60 and 1.84). Both indexes were significantly higher on the semi-shady slopes than the sunny side slopes (P < 0.05). Pielou’s species evenness index (J) reached its maximum (0.84) on shady side slopes, which was 25.37% higher than on sunny side slopes. Generally, H′ and R steadily increased with increased years since restoration. H′, J, and R all reached their maximums after six years had passed since restoration, with index level 22.39%, 17.81%, and 25.17% higher than after only one year since restoration, respectively. In this study, there were no obvious variation trends in the plant diversity indexes with changes in the side slope gradient and altitude.

The stepwise regression analysis model and its parameters are shown in Table 2. The stepwise regression of the five habitat factors found that the main influencing factors of Shannon-Wiener diversity index (H′) and Pielou’s species evenness index (J) were side slope position and aspect, while the main influencing factor on Margalef’s richness index (R) was side slope aspect. The regression coefficients of the years since restoration, slope gradient, and altitude were so small that they were excluded. Therefore, side slope positions and aspects were the key factors influencing plant diversity, while other factors had no significant effect on plant diversity.