Composition of Species and Spatial Patterns of Phyllostachys edulis Mixed Forests in a Succession after Enclosure

Abstract

This paper investigates the changes in species diversity and the spatial pattern of the population of a moso bamboo (Phyllostachys edulis) forest community in the Wuyi Mountain Nature Reserve in Fujian Province, China. Using the method of space–time substitution, the composition of species and changes in the spatial patterns of different communities were analyzed by taking Phyllostachys edulis mixed forests after 0, 3, 5, 10, 15, and 40 years of enclosure as the survey object. The results showed that as the number of years of enclosure extended, the important values of the Japanese bay tree (Machilus thunbergii Sieb. & Zucc.), chinquapin (Castanopsis faberi Hance), and Masson’s pine (Pinus massoniana Lamb.) increased continuously. These trees became the dominant species of mature forest after 40 years of enclosure. The species diversity of mature and young trees in the community generally increased in parallel with the years of enclosure, and the species diversity of the shrubs generally increased first and then decreased as the number of years of enclosure increased. With the extension of enclosure years, the average diameter at breast height of the mature trees tended to increase, indicating that there were increasing advantages of mature growth as the time of enclosure increased. With the extension of enclosure years, the pattern of spatial distribution of the mature trees in the forest community was generally an aggregated distribution. The Phyllostachys edulis forest community under the prolonged enclosure conditions responded in a positive direction and ultimately formed a more stable Phyllostachys edulis mixed forest community.

1. Introduction

Community succession is a natural phenomenon that results from the continuous replacement of community composition types and changes in the habitat caused by the transformation of communities in the environment [1,2,3,4]. Community succession in enclosed forests is called closed succession. The results of this study showed that the vegetation cover of degraded forest land can be improved through enclosure. Numerous studies have shown that enclosure can increase the diversity of vegetation and biomass of the community [5,6] and can also improve the soil moisture [7]. Studies on species diversity and spatial patterns provide important information to understand the mechanisms used by species to coexist [8,9]. The study of species diversity and spatial patterns has been favored by researchers in China and throughout the world [10,11,12,13]. Changes in the diversity of community species are a central objective of community succession [14,15]. Such changes reflect the diversity and spatial heterogeneity of the community structure [16,17]. The pattern of spatial distribution can be an important indicator of community health and stability [18]. Many researchers in China and throughout the world have conducted extensive studies on the composition of species and the spatial pattern of communities at different phases of succession. Shan [19] and other researchers have studied the species diversity of grassland communities at different stages of succession. Isermann [20] studied the changes in species diversity during the succession of coastal dune communities, while Liu et al. [21] studied the spatial pattern of the distribution of Artemisia ordosica Krasch. populations in response to the cordoning off of their site. In addition, Meiners [22] studied the population distribution dynamics of plant species during different succession processes. However, little research has been reported on the changes in species diversity and spatial patterns of the plant populations of Phyllostachys edulis forest communities at different stages of succession.

The quality, production value, and economic value of Phyllostachys edulis in the Wuyi Mountain Nature Reserve, Fujian Province, China, are among the best in the country. In recent years, Wuyi Mountain City has adopted a policy of enclosing the forest, which encourages the Phyllostachys edulis forest community to use its ability for self-renewal to restore and improve the forest environment. In this study, we selected mixed Phyllostachys edulis forests on Wuyi Mountain with different years of enclosure as the research object and comprehensively investigated and studied the mixed Phyllostachys edulis forest communities with different years of enclosure using established standard plots. In this study, Phyllostachys edulis mixed forests in the Wuyi Mountain Nature Reserve in Fujian Province, China were used the experimental object, their enclosure times were 3, 5, 10, 15, 40 years and were compared with the conventional operation of the Phyllostachys edulis forest. The results of this study provide a reference for the successional progress of enclosed mangosteen forests and their management.

2. Materials and Methods

2.1. Study Site

Wuyi Mountain National Park is located in the northern part of Fujian Province (117°24′13″–117°59′19″, 27°31′20″–27°55′49″), which is situated in the humid zone of the central subtropical monsoon, with an annual average temperature of 17.7 °C, precipitation of 1800 mm, an annual average duration of sunshine of 1910.2 h, and an average of 79% relative humidity. As the altitude of these mountains increases, there are four types of vegetation zones with different community characteristics, namely, evergreen broad-leaved forests, mixed coniferous and broad-leaved forests, dwarf forests, and mesic meadows, with clear demarcation lines and obvious plant vertical zones from the foot of the mountain upwards [23].

2.2. Study Design

After a comprehensive survey of the Wuyishan National Forest Park in 2021, mixed Phyllostachys edulis forests with 3, 5, 10, 15, and 40 years of enclosure in the Aotou, Tongmu, and Da’an villages in Wuyishan City were selected as the survey object, and conventionally operated Phyllostachys edulis forests were used as the control (0 years). The six communities were sequentially numbered as Type A (control), Type B (3 years), Type C (5 years), Type D (10 years), Type E (15 years), and Type F (40 years). The basic conditions of the sample plots and stands in the six different treatments are shown in

Table 1. The characteristics of Phyllostachys edulis forests.

Type	Longitude	Latitude	Altitude (m)	Slope (°)	Slope Position	Soil Type
A	117°51′09″ E	27°52′26″ N	692 ± 11	23.3 ± 2.9	uphill position	red soil
B	117°38′24″ E	27°39′44″ N	790 ± 26	23.7 ± 5.1	mid-level	yellowish red soil
C	117°38′59″ E	27°42′08″ N	867 ± 15	16.0 ± 1.7	mid-level	yellowish red soil
D	117°42′00″ E	27°43′55″ N	749 ± 16	21.7 ± 2.9	uphill position	yellowish red soil
E	117°50′38″ E	27°52′44″ N	673 ± 15	20.0 ± 5.0	uphill position	red soil
F	117°42′50″ E	27°45′40″ N	874 ± 26	17.3 ± 1.2	mid-level	yellowish red soil

and

Table 2. Basic information of the Phyllostachys edulis forest stands.

Type	Plant Height (m)		DBH (cm)		Density (Plant·hm−2)		Bamboo Tree Mixed Ratio
	Phyllostachys edulis	Tree	Phyllostachys edulis	Tree	Phyllostachys edulis	Tree
A	15.3 ± 0.1	-	11.6 ± 0.2	-	2764 ± 450	-	1:0
B	16.2 ± 1.6	15.3 ± 5.2	11.8 ± 0.5	31.6 ± 8.7	2123 ± 221	45 ± 25	47:1
C	13.9 ± 1.0	4.7 ± 1.8	11.7 ± 0.2	5.6 ± 3.1	4086 ± 147	45 ± 42	91:1
D	17.5 ± 1.9	11.0 ± 2.3	12.2 ± 0.2	19.5 ± 3.3	3365 ± 340	95 ± 50	35:1
E	13.7 ± 0.6	10.9 ± 0.9	12.0 ± 0.2	14.9 ± 0.6	5909 ± 1480	1172 ± 137	5:1
F	19.7 ± 0.8	10.6 ± 2.4	11.8 ± 0.5	16.0 ± 4.0	2965 ± 1212	756 ± 282	4:1

Note: Tree refers to other trees in the community except Phyllostachys edulis.

. Three sample plots were established in each of the communities that had been enclosed in different years. This resulted in 18 plots in total that were 25.8 m × 25.8 m. The enclosed mangosteen (Garcinia mangostana Linn) forests were not disturbed by anthropogenic activities, such as digging for bamboo shoots and cleaning old bamboo, and were in a completely enclosed state. Regularly operated Phyllostachys edulis forests are normally utilized for a variety of business activities.

Sample plots selected during the survey were fully representative, did not cross streams, roads, or logged survey lines, and were far away from the forest edge. The sample plots were established in places where the tree species and stand densities were evenly distributed. Three different replicate sample plots were established in the same stand and did not cross the stand. The different replicate sample plots were spaced not less than 25 m and not more than 50 m apart (Figure 1).

2.3. Community Investigation Methods

For the adult trees, 25.8 m × 25.8 m sample plots were considered to be the basic unit. A per-tree survey for all plants with a diameter at breast height (DBH) > 3 cm in the sample plot was conducted. The species name, DBH, tree height (H), and crown width of each individual were recorded. Simultaneously, the lower-left corner of each sample plot was considered to be the origin when the coordinates of each individual (x- and y-axis) were recorded. For young trees, 25.8 m × 25.8 m was considered to be the basic unit, and per-tree surveys for all plants >1 m high and which had a DBH < 3 cm in the sample area were recorded along with the species name, DBH, H, and crown width of each individual plant. For the shrub and herbaceous layers, 5 m × 5 m was the basic unit for shrubs and 1 m × 1 m was the basic unit for the herbaceous plants. The species name, number of individuals, coverage, and average height of the shrubs and herbaceous plants in each plot were recorded. The environmental characteristics of each site, including latitude, longitude, altitude, geomorphology, soil type, slope, and slope position, among others, were also recorded.

2.4. Calculation of the Species Diversity Index (SDI)

The relative density, relative frequency, and relative significance (relative cover) of each species were calculated based on the information of the sample plots and the content of the survey, and their important value (IV) was calculated.

The significant values for mature and young trees were shown as follows:

IV = (RA + RF + RP)/3

(1)

The importance values of the shrub and herb layers were as follows:

IV = (RA + RF + RC)/3

(2)

α-diversity, also known as the diversity within habitats, focuses on the number of species in a homogeneous environment within a local area. Based on the principle of multi-index selection, richness, diversity, and evenness indices were chosen for calculation, which can reflect the changes in α-diversity in a more comprehensive way.

(1)

Richness index, described as the Margalef index:

Ma = (S − 1)/ln(B)

(3)

(2)

Diversity indices:

Simpson’s index:

$$\mathrm{H}=1-{\sum }_{\mathrm{i}=1}^{\mathrm{s}}{\mathrm{p}}_{\mathrm{i}}^2$$

(4)

Shannon–Wiener index:

$${\mathrm{H}}^{\prime }=-{\sum }_{\mathrm{i}=1}^{\mathrm{s}}{\mathrm{p}}_{\mathrm{i}}\mathrm{l}\mathrm{n}\left({\mathrm{p}}_{\mathrm{i}}\right)$$

(5)

(3)

Uniformity index:

Pielou index:

J_sw = H′/InS

(6)

Ma is the Margalef’s index; H is the Simpson’s index; H′ is the Shannon–Wiener’s index; J_SW is the Pielou’s index; p_i is the proportion of the number of individuals of species i; n_i, is the total n of individuals of all the species, i.e., p_i = n_i/n; i = 1, 2, 3 S; S is the average number of species in the sample plot sample; and B is the total biomass of the community [24].

2.5. Point Pattern Analysis

The point pattern analysis can be divided into univariate pattern analysis, bivariate pattern analysis, and multivariate pattern analysis, and the distribution patterns of communities and populations, the relationships between and among populations, and the reflection of their ecological processes at different scales can be explored using different null hypothesis models. The L (r) function was derived from Ripley’s K function and can explain the actual spatial pattern more directly and simply than the K (r) function. Thus, the L (r) function was used to describe the spatial patterns of individuals after different periods of enclosure at various spatial scales in the Phyllostachys edulis forests. In analyzing the spatial associations among different species, the K₁₂ (r) function was extrapolated on the basis of Ripley’s K function. Similarly, the L₁₂ (r) function was obtained based on the variance correction and linearization of the K₁₂ (r) function. The principle and calculation procedure are as described by Wiegand and Moloney [25,26].

2.6. Zero Model Selection

In this study, the univariate L (r) function was chosen to be the complete spatial randomness (CSR) model [27,28], and 100 stochastic simulations were conducted using the Monte Carlo method to test the deviation of the null model from the observed values and to construct the upper and lower two packet trajectory lines by applying the simulated maxima and minima, referred to as T. The intervals of the distributions with a confidence level of 95% were then intercepted. If the actual distributions of the L (r) values and L₁₂ (r) fell within the envelope, the plants were randomly distributed or the two enclosure years were independent of each other and lacked a spatial correlation. If they were above the envelope, they were clustered or there was a spatial positive correlation between the two years of enclosure. If they were below the envelope, they were uniformly distributed or there was a spatial negative correlation between the two enclosures.

2.7. Statistical Analysis

Microsoft Excel 2013 (Redmond, WA, USA) was used to integrate data, and SPSS 19.0 (IBM, Inc., Armonk, NY, USA) was used for analysis of variance (ANOVA). The Kruskal–Wallis test was used for the indicators within the moso bamboo forest communities under different periods of enclosure, while significant difference was found using the Dunn test. A point pattern analysis was performed using Programita (2014 version) software. The pairwise correlation function g (r) was chosen; the spatial scale of the sample plots was 0–40 m; the ring width was established as 3 m; the analytical step size was 0.2 m; and the 95% confidence intervals were determined through 199 simulations using the Monte Carlo method.

3. Results

3.1. Plant Species and Importance Values in Communities following Different Periods of Enclosure

The primary adult trees (importance value ≥ 1) were members of 30 genera in 23 families, which resulted in a total of 41 species in the six different treatment samples (

Table 3. The importance values of mature trees after different periods of natural enclosure.

Plant Type	Plant Important Values
	Type A	Type B	Type C	Type D	Type E	Type F
Castanopsis faberi Hance	-	12.417	-	12.934	13.310	14.431
Symplocos congesta Benth.	-	41.772	-	7.503	-	3.745
Manglietia fordiana Oliv.	-	27.079	-	-	-	-
Castanopsis eyrei (Champ. ex Benth.) Tutch.	-	18.732	-	-	3.187	1.967
Photinia bodinieri Lévl.	-	-	41.999	3.741	-	-
Machilus thunbergii Sieb. et Zucc.	-	-	9.881	12.965	13.192	13.324
Liquidambar formosana Hance	-	-	21.127	-	-	-
Cunninghamia lanceolata (Lamb.) Hook.	-	-	16.373	6.348	18.869	-
Trachycarpus fortunei (Hook.) H. Wendl.	-	-	10.620	-	-	-
Engelhardia roxburghiana Wall.	-	-	-	15.565	-	1.520
Schima superba Gardn. et Champ.	-	-	-	13.055	2.714	2.792
Castanopsis sclerophylla (Lindl. et Paxton) Schottky	-	-	-	8.346	-	1.041
Quercus myrsinifolia Blume	-	-	-	5.863	-	-
Elaeocarpus decipiens Hemsl.	-	-	-	6.181	-	-
Eurya nitida Korthals	-	-	-	3.749	2.061	3.842
Adinandra millettii (Hook. et Arn.) Benth. et Hook. f. ex Hance	-	-	-	3.749	-	2.380
Pinus massoniana Lamb.	-	-	-	-	9.266	13.511
Fagus longipetiolata Seem.	-	-	-	-	6.065	-
Rhododendron latoucheae Franch.	-	-	-	-	5.196	-
Dendropanax dentiger (Harms) Merr.	-	-	-	-	4.896	2.020
Adinandra glischroloma var. macrosepala (Metcalf) Kobuski	-	-	-	-	3.690	-
Eurya muricata Dunn	-	-	-	-	3.658	-
Quercus glauca Thunb.	-	-	-	-	2.825	1.415
Lithocarpus iteaphyllus (Hance) Rehd.	-	-	-	-	2.688	-
Vaccinium mandarinorum Diels	-	-	-	-	2.258	-
Litsea rotundifolia var. oblongifolia (Nees) Allen	-	-	-	-	1.506	-
Castanopsis chinensis (Sprengel) Hance	-	-	-	-	1.342	-
Taxus wallichiana var. chinensis (Pilger) Florin	-	-	-	-	1.228	-
Itea omeiensis C. K. Schneider	-	-	-	-	1.025	-
Symplocos sumuntia Buch.-Ham. ex D. Don	-	-	-	-	1.023	-
Alniphyllum fortunei (Hemsl.) Makino	-	-	-	-	-	4.993
Castanopsis fargesii Franch.	-	-	-	-	-	2.928
Loropetalum chinense (R. Br.) Oliver	-	-	-	-	-	2.833
Choerospondias axillaris (Roxb.) B. L. Burtt & A. W. Hill	-	-	-	-	-	2.455
Lithocarpus hancei (Bentham) Rehd.	-	-	-	-	-	2.087
Daphniphyllum oldhamii (Hemsl.) Rosenthal	-	-	-	-	-	1.974
Ternstroemia gymnanthera (Wight et Arn.) Beddome	-	-	-	-	-	1.650
Symplocos theophrastifolia Siebold et Zucc.	-	-	-	-	-	1.391
Diospyros morrisiana Hance	-	-	-	-	-	1.270
Syzygium buxifolium Hook. et Arn.	-	-	-	-	-	1.260
Symplocos anomala Brand	-	-	-	-	-	1.111

). Among them, plants in the beech family (Fagaceae) had the highest number of species, which totaled ten, followed by members of the Pentaphylacaceae family and Symplocaceae, with five and four species, respectively. In addition, there were 18 families with only one species, which comprised 78.26% of the total number of families. In Type B, the ranking of plant IVs in order was: Symplocos congesta Benth., Manglietia fordiana Oliv., Castanopsis eyrei (Champ. ex Benth.) Tutch., and Castanopsis faberi Hance. Symplocos congesta Benth. (41.772%) and Manglietia fordiana Oliv. (27.079%) had significantly higher IVs than the other adult trees. In Type C, the dominant plant IVs, in order, were Photinia bodinieri Lévl., Liquidambar formosana Hance, Cunninghamia lanceolata (Lamb.) Hook., Trachycarpus fortunei (Hook.) H. Wendl., and Machilus thunbergii Sieb. et Zucc., with Photinia bodinieri Lévl. (41.999%) and Liquidambar formosana Hance (21.127%) serving as the predominant adult tree species. In Type D, the top five adult trees in terms of plant species IVs were Engelhardia roxburghiana Wall., Schima superba Gardn. et Champ., Machilus thunbergii Sieb. et Zucc., Castanopsis faberi Hance, and Castanopsis sclerophylla (Lindl. et Paxton) Schottky (in order). Engelhardia roxburghiana Wall. (15.565%) and Schima superba Gardn. et Champ. (13.055%) were the predominant adult trees. In Type E, the top five adult trees in terms of the plant species IVs were Cunninghamia lanceolata (Lamb.) Hook., Castanopsis faberi Hance, Machilus thunbergii Sieb. et Zucc., Pinus massoniana Lamb., and Fagus longipetiolata Seem. Cunninghamia lanceolata (Lamb.) Hook. (18.869%) and Castanopsis faberi Hance (13.310%) were the most predominant of the adult trees. In Type F, the top five adult trees ranked according to the IVs of the plant species were Castanopsis faberi Hance, Pinus massoniana Lamb., Machilus thunbergii Sieb. et Zucc., Alniphyllum fortunei (Hemsl.) Makino, and Eurya nitida Korthals in descending order. Castanopsis faberi Hance (14.431%) and Pinus massoniana Lamb. (13.511%) were significantly more important than the other adult tree plants. The dominant species of adult trees after different periods of enclosure varied. With the extension of the enclosure of the forests, the IVs of Machilus thunbergii Sieb. et Zucc., Castanopsis faberi Hance, and Pinus massoniana Lamb. increased continuously; they gradually became the dominant species of adult trees. The IVs of Cunninghamia lanceolata (Lamb.) Hook. showed a tendency to decrease, increase, and then decrease. It became the dominant species of adult tree in the community following 15 years of enclosure.

In the six different treatment plots, the predominant young trees (importance value ≥ 1) were members of 32 genera in 24 families, which totaled 47 species (

Table 4. The importance values of young trees after different periods of natural enclosure.

Plant Type	Plant Importance Values
	Type A	Type B	Type C	Type D	Type E	Type F
Liquidambar formosana Hance	-	39.069	-	-	-	-
Camellia oleifera Abel.	-	4.809	-	-	-	-
Photinia bodinieri Lévl.	-	13.736	-	10.996	-	-
Machilus thunbergii Sieb. et Zucc.	-	15.851	8.854	3.645	5.489	2.802
Trachycarpus fortunei (Hook.) H. Wendl.	-	1.720	-	-	-	-
Macropanax rosthornii (Harms) C. Y. Wu ex Hoo	-	4.095	-	-	-	-
Cunninghamia lanceolata (Lamb.) Hook.	-	1.715	6.017	5.350	7.792	-
Machilus chrysotricha H. W. Li	-	3.476	-	11.367	-	10.533
Symplocos stellaris Brand	-	1.797	-	1.359	1.422	4.695
Lindera aggregata (Sims) Kosterm.	-	1.795	-	-	-	-
Schima superba Gardn. et Champ.	-	4.348	11.888	13.449	5.687	4.089
Ilex triflora Bl.	-	3.607	-	-	-	-
Castanopsis faberi Hance	-	1.924	2.192	5.289	13.640	3.360
Manglietia fordiana Oliv.	-	2.056	7.950	-	-	1.547
Photinia prunifolia (Hook. et Arn.) Lindl.	-	-	15.491	2.589	2.171	3.038
Castanopsis eyrei (Champ. ex Benth.) Tutch.	-	-	17.419	-	3.437	-
Itea omeiensis C. K. Schneider	-	-	4.724	3.207	-	2.905
Symplocos pseudobarberina Gontsch.	-	-	20.767	-	-	-
Loropetalum chinense (R. Br.) Oliver	-	-	2.788	1.996	-	3.501
Adinandra glischroloma var. macrosepala (Metcalf) Kobuski	-	-	1.910	-	16.948	3.256
Castanopsis fargesii Franch.	-	-	-	1.109	-	2.069
Symplocos congesta Benth.	-	-	-	5.156	-	4.076
Adinandra millettii (Hook. et Arn.) Benth. et Hook. f. ex Hance	-	-	-	5.937	-	4.598
Eurya nitida Korthals	-	-	-	5.891	2.660	4.088
Castanopsis sclerophylla (Lindl. et Paxton) Schottky	-	-	-	5.760	-	1.312
Diospyros morrisiana Hance	-	-	-	2.064	-	4.582
Meliosma rigida var. pannosa (Hand.-Mazz.) Law	-	-	-	3.078	-	1.496
Photinia hirsuta Hand.-Mazz.	-	-	-	1.360	-	-
Rhododendron latoucheae Franch.	-	-	-	3.314	7.219	3.815
Castanopsis fissa (Champion ex Bentham) Rehder et E. H. Wilson	-	-	-	2.431	-	-
Engelhardia roxburghiana Wall.	-	-	-	3.676	-	4.286
Quercus glauca Thunb.	-	-	-	-	11.068	2.952
Cinnamomum austrosinense H. T. Chang	-	-	-	-	2.057	-
Eurya muricata Dunn	-	-	-	-	8.502	-
Ternstroemia gymnanthera (Wight et Arn.) Beddome	-	-	-	-	1.430	-
Oligostachyum oedogonatum (Z. P. Wang et G. H. Ye) Q. F. Zhang et K. F. Huan	-	-	-	-	6.581	-
Fagus longipetiolata Seem.	-	-	-	-	3.898	-
Elaeocarpus chinensis (Gardn. et Chanp.) Hook. f. ex Benth.	-	-	-	-	-	2.988
Symplocos anomala Brand	-	-	-	-	-	2.331
Ilex pubescens Hook. et Arn.	-	-	-	-	-	2.003
Lasianthus chinensis (Champ.) Benth.	-	-	-	-	-	1.102
Syzygium buxifolium Hook. et Arn.	-	-	-	-	-	1.988
Camellia chekiangoleosa Hu	-	-	-	-	-	2.311
Daphniphyllum oldhamii (Hemsl.) Rosenthal	-	-	-	-	-	1.781
Ilex ficoidea Hemsl.	-	-	-	-	-	1.054
Helicia cochinchinensis Lour.	-	-	-	-	-	1.162
Parakmeria lotungensis (Chun et C. Tsoong) Law	-	-	-	-	-	2.349

). The highest number of species was found in the family Fagaceae, with seven species, followed by five and four species in the families Pentaphylacaceae, Symplocaceae, and Lauraceae, and 16 families with only one species, which comprised 66.67% of the total number of families. Juvenile plants that occurred within all of the communities of the different treatment samples were Machilus thunbergii Sieb. et Zucc., Schima superba Gardn.et Champ., and Castanopsis faberi Hance. In Type B, the top five juvenile tree plants in terms of the plant species IV were Liquidambar formosana Hance, Machilus thunbergii Sieb. et Zucc., Photinia bodinieri Lévl., Camellia oleifera Abel., and Schima superba Gardn.et Champ. in order, with Liquidambar formosana Hance (39.069%) and Machilus thunbergii Sieb. et Zucc. (15.851%) comprising the predominant juvenile trees. In Type C, Symplocos pseudobarberina Gontsch., Castanopsis eyrei (Champ. ex Benth.) Tutch., Photinia prunifolia (Hook. et Arn.) Lindl., Schima superba Gardn. et Champ., and Machilus thunbergii Sieb. et Zucc. were the predominant species (in order). Symplocos pseudobarberina Gontsch. (20.767%) and Castanopsis eyrei (Champ. ex Benth.) Tutch. (17.419%) were significantly more important than the other juvenile trees. In Type D, the top five juvenile trees in terms of the plant species IVs were Schima superba Gardn.et Champ., Machilus chrysotricha H. W. Li, Photinia bodinieri Lévl., Adinandra millettii (Hook. et Arn.) Benth. et Hook. f. ex Hance, and Eurya nitida Korthals, with populations of Schima superba Gardn.et Champ. (13.449%) and Machilus chrysotricha H. W. Li (11.367%) that were significantly higher than those of the other juvenile trees. In Type E, the plants ranked according to their IVs were Adinandra glischroloma var. macrosepala (Metcalf) Kobuski, Castanopsis faberi Hance, Cyclobalanopsis glauca Thunb., Eurya muricata Dunn, and Cunninghamia lanceolata (Lamb.) Hook., with the IVs of Adinandra glischroloma var. macrosepala (Metcalf) Kobuski (16.948%) and Castanopsis faberi Hance (13.640%) being significantly higher than those of the other juvenile trees. In Type F, the top five juvenile trees in terms of the IVs of the plant species were Machilus chrysotricha H. W. Li, Symplocos stellaris Brand, Adinandra millettii (Hook. et Arn.) Benth. et Hook. f. ex Hance, Diospyros morrisiana Hance, and Engelhardia roxburghiana Wall. (in order). The dominance of Machilus chrysotricha H. W. Li (10.533%) and Symplocos stellaris Brand (4.695%) was readily apparent among the juvenile trees. With the increase in the length of time that the forests were enclosed, among the young trees, the IVs of Machilus thunbergii Sieb. et Zucc. decreased, and its dominance gradually weakened, while the IVs of Machilus chrysotricha H. W. Li gradually increased, and it became the dominant plant among the young trees in the community when the length of enclosure was 40 years.

Plants in the main shrub layer (importance value ≥ 1) were members of 32 genera in 23 families and totaled 45 species in the six different treatment plots (

Table 5. The importance values of shrubs in different years of enclosure.

Plant Type	Plant Important Values
	Type A	Type B	Type C	Type D	Type E	Type F
Eurya nitida Korthals	4.192	-	1.862	-	-	-
Eurya rubiginosa var. attenuata H.T.Chang	3.647	12.955	11.811	43.329	40.892	17.807
Ardisia crenata Sims	5.612	-	-	-	-	-
Maesa japonica (Thunb.) Moritzi. ex Zoll.	12.500	17.985	-	22.546	15.037	-
Kadsura longipedunculata Finet et Gagnep.	-	-	-	-	5.159	-
Smilax china L.	5.344	-	5.162	-	9.744	-
Tashiroea sinensis Diels	12.590	2.292	-	11.122	-	-
Camellia oleifera Abel.	2.719	-	-	-	5.730	-
Itea omeiensis C. K. Schneider	5.036	1.150	6.382	7.693	-	3.761
Diplospora dubia (Lindl.) Masam.	3.389	1.150	-	-	-	-
Rhododendron latoucheae Franch.	3.129	1.146	1.570	-	-	3.923
Castanopsis faberi Hance	3.411	-	-	-	-	-
Loropetalum chinense (R. Br.) Oliver	3.196	5.133	10.900	-	-	6.211
Vaccinium trichocladum Merr. et Metc.	2.802	-	1.197	-	3.937	-
Helicia cochinchinensis Lour.	1.547	1.144	-	-	4.562	-
Rubus tsangiorum Handel-Mazzetti	2.523	-	-	-	-	-
Camellia cuspidata (Kochs) Wright ex Gard.	6.607	2.334	12.415	5.841	-	-
Symplocos congesta Benth.	3.730	6.654	1.570	-	-	6.730
Coptosapelta diffusa (Champ. ex Benth.) Van Steenis	7.125	-	-	-	-	-
Rhododendron ovatum (Lindl.) Planch.	2.238	5.811	2.081	-	-	31.715
Zanthoxylum scandens Bl.	1.009	-	-	-	-	-
Ternstroemia gymnanthera (Wight et Arn.) Beddome	1.009	1.146	-	4.530	-	-
Engelhardia roxburghiana Wall.	4.127	1.172	-	-	-	-
Syzygium buxifolium Hook. et Arn.	-	2.584	2.088	-	-	-
Eurya japonica Thunb.	-	4.908	-	-	-	-
Photinia serratifolia (Desf.) Kalkman	-	1.348	-	-	-	-
Mahonia bealei (Fort.) Carr.	-	3.020	5.965	-	-	17.364
Ilex pubescens Hook. et Arn.	-	3.696	-	-	-	-
Ligustrum sinense Lour.	-	1.144	-	-	-	-
Smilax glabra Roxb.	-	2.130	-	-	-	-
Machilus leptophylla Hand.-Mazz.	-	1.148	-	-	-	-
Symplocos anomala Brand	-	1.148	-	-	-	-
Dendropanax dentiger (Harms) Merr.	-	1.145	-	-	-	-
Melastoma malabathricum Linnaeus	-	15.272	-	-	-	-
Mucuna sempervirens Hemsl.	-	1.235	-	-	-	-
Rosa laevigata Michx.	-	1.150	-	-	-	-
Lindera aggregata (Sims) Kosterm.	-	-	9.137	-	-	-
Camellia sinensis (L.) O. Ktze.	-	-	18.751	-	3.274	-
Stauntonia chinensis DC.	-	-	1.959	-	4.133	2.922
Rubus amphidasys Focke ex Diels	-	-	5.715	-	4.960	-
Lasianthus japonicus Miq.	-	-	1.435	-	-	-
Symplocos stellaris Brand	-	-	-	4.940	-	-
Ardisia japonica (Thunberg) Blume	-	-	-	-	2.573	-
Castanopsis eyrei (Champ. ex Benth.) Tutch.	-	-	-	-	-	4.291
Ternstroemia kwangtungensis Merr.	-	-	-	-	-	5.277

). Among them, plants of the Rosaceae, Theaceae, and Pentaphylacaceae families had the highest number of species, four each in total, while the Primulaceae, Ericaceae, Cyperaceae, and Symplocaceae families had the second highest number of species, each with three. A total of 13 families had only one species, which accounted for 56.52% of the total number of families. Shrub plants that occurred in all the communities in the different treatment plots were Eurya rubiginosa var. Attenuata H.T.Chang. In Type A, the top five shrub-layer plants in terms of the plant species IVs were Tashiroea sinensis Diels, Maesa japonica (Thunb.) Moritzi. ex Zoll., Coptosapelta diffusa (Champ. ex Benth.) Van Steenis, Camellia cuspidata (Kochs) Wright ex Gard., and Ardisia crenata Sims, where the degree of dominance of Bredia sinensis Diels (12.590%) and Maesa japonica (Thunb.) Moritzi. ex Zoll. (12.50%) in the shrub layer was significant. In Type B, the plants in order were Maesa japonica (Thunb.) Moritzi. ex Zoll., Melastoma malabathricum Linnaeus, Eurya rubiginosa var. Attenuata H.T.Chang, Symplocos congesta Benth., and Rhododendron ovatum (Lindl.) Planch., with Maesa japonica (Thunb.) Moritzi. ex Zoll. (17.985%) and Melastoma malabathricum Linnaeus (15.272%) being significantly more important than the other plants in the shrub layer. In Type C, the top five shrub-layer plants in terms of plant species IVs were Camellia sinensis (L.) O. Ktze., Camellia cuspidata (Kochs) Wright ex Gard., Eurya rubiginosa var. Attenuata H.T.Chang, Loropetalum chinense (R. Br.) Oliver, and Lindera aggregata (Sims) Kosterm. in order of the IVs. Camellia sinensis (L.) O. Ktze. (18.751%) and Camellia cuspidata (Kochs) Wright ex Gard. (12.415%) were the predominant shrub layer species. In Type D, Eurya rubiginosa var. Attenuata H.T.Chang, Maesa japonica (Thunb.) Moritzi. ex Zoll., Tashiroea sinensis Diels, Itea omeiensis C. K. Schneider, and Camellia cuspidata (Kochs) Wright ex Gard. were the predominant species of shrubs, in that order. Eurya rubiginosa var. Attenuata H.T.Chang (43.329%) and Maesa japonica (Thunb.) Moritzi. ex Zoll. (22.546%) were dominant in the shrub layer. In Type E, the top five plants in the shrub layer ranked in terms of the plant species IVs were Eurya rubiginosa var. Attenuata H.T.Chang, Maesa japonica (Thunb.) Moritzi. ex Zoll., Smilax china L., Camellia oleifera Abel., and Kadsura longipedunculata Finet et Gagnep. (in order). Eurya rubiginosa var. Attenuata H.T.Chang (40.892%) and Maesa japonica (Thunb.) Moritzi. ex Zoll. (15.037%) were significantly more important than the other plants in the shrub layer. In Type F, Rhododendron ovatum (Lindl.) Planch., Eurya rubiginosa var. Attenuata H.T.Chang, Mahonia bealei (Fort.) Carr., Symplocos congesta Benth., and Loropetalum chinense (R. Br.) Oliver were the most important plants. Rhododendron ovatum (Lindl.) Planch. (31.715%) and Eurya rubiginosa var. Attenuata H.T.Chang (17.807%) were the predominant species in the shrub layer. As the number of years of enclosure increased, the IVs of Eurya rubiginosa var. Attenuata H.T.Chang in the shrub layer generally tended to increase and then decrease, and it became the dominant species in the understory shrub layer after 10 and 15 years after enclosure.

Plants in the primary herbaceous layer (IV ≥ 1) were members of 27 families and 38 genera and totaled 46 species in the six different treatment samples (

Table 6. The importance values of herbaceous plants after different periods of enclosure.

Plant Type	Plants Important Value
	Type A	Type B	Type C	Type D	Type E	Type F
Carex cruciata Wahlenb.	9.073	9.124	10.070	28.048	8.387	53.395
Rubus buergeri Miq.	2.618	4.190	-	-	-	-
Cyclosorus interruptus (Willd.) H. Ito	1.047	-	-	-	-	-
Woodwardia japonica (L. F.) Sm.	16.372	9.964	4.485	16.599	18.558	9.331
Parathelypteris glanduligera (Kze.) Ching	-	-	-	-	-	-
Rubus amphidasys Focke ex Diels	1.260	-	-	-	-	-
Tainia dunnii Rolfe	1.369	-	-	-	-	-
Ainsliaea fragrans Champ.	1.365	2.907	-	-	2.021	-
Liparis nervosa (Thunb. ex A. Murray) Lindl.	1.061	-	1.362	-	1.955	-
Liriope graminifolia (L.) Baker	3.295	-	4.247	3.869	3.273	-
Polystichum balansae Christ	4.652	-	-	1.540	-	-
Smilax glabra Roxb.	1.872	-	-	-	-	-
Trachelospermum jasminoides (Lindl.) Lem.	20.765	-	1.497	-	30.140	-
Dicranopteris pedata (Houttuyn) Nakaike	1.116	-	-	3.502	3.972	-
Smilax china L.	6.973	-	-	-	-	-
Dryopteris fuscipes C. Chr.	2.340	3.580	4.716	2.578	3.235	-
Selaginella tamariscina (P. Beauv.) Spring	1.582	-	-	-	-	-
Diplopterygium glaucum (Thunberg ex Houttuyn) Nakai	1.949	-	-	8.516	17.211	22.166
Viola diffusa Ging.	-	12.197	-	-	-	4.156
Sarcandra glabra (Thunb.) Nakai	5.323	-	1.474	2.511	4.266	-
Hylodesmum podocarpum (Candolle) H. Ohashi & R. R. Mill	1.228	-	-	3.576	1.445	-
Tashiroea sinensis Diels	2.358	-	-	-	-	-
Alpinia japonica (Thunb.) Miq.	2.434	-	-	6.287	2.121	-
Selaginella doederleinii Hieron.	2.035	-	-	4.607	-	-
Anoectochilus roxburghii (Wall.) Lindl.	1.347	-	1.422	2.315	-	-
Agrimonia pilosa Ldb.	-	2.278	-	-	-	-
Carex maubertiana Boott	-	3.302	-	5.184	-	-
Lygodium japonicum (Thunb.) Sw.	-	1.441	-	-	-	-
Carex scaposa C. B. Clarke	-	13.537	-	2.155	-	-
Dianella ensifolia (L.) Redouté	-	21.136	-	-	-	5.476
Selaginella involvens (Sw.) Spring	-	4.848	-	-	-	-
Goodyera schlechtendaliana Rchb. F.	-	1.345	-	-	-	-
Ardisia japonica (Thunberg) Blume	-	8.723	51.810	-	-	-
Cymbidium ensifolium (L.) Sw.	-	1.427	1.371	4.954	-	-
Huperzia serrata (Thunb. ex Murray) Trevis.	-	-	5.348	-	-	-
Lysimachia christinae Hance	-	-	2.881	-	-	-
Liriope spicata (Thunb.) Lour.	-	-	7.663	-	-	-
Osmunda japonica Thunb.	-	-	1.655	-	-	5.476
Alkekengi officinarum Moench	-	-	-	1.928	-	-
Calanthe graciliflora Hayata	-	-	-	1.830	-	-
Rubus tsangiorum Handel-Mazzetti	-	-	-	-	3.417	-

). The Orchidaceae family had the highest number of species, totaling six, followed by the Rosaceae and Cyperaceae families, with four and three species, respectively. A total of 13 families had only one species, which comprised 48.15% of the total number of families. The herbaceous plants that occurred within all the communities of the different treatment sample plots were Carex cruciata Wahlenb. and Woodwardia japonica (L. F.) Sm. In Type A, the top five herbaceous-layer plants ranked by the IVs of the plant species were Trachelospermum jasminoides (Lindl.) Lem., Woodwardia japonica (L. F.) Sm., Carex cruciata Wahlenb., Smilax china L., and Sarcandra glabra (Thunb.) Nakai (in order). Trachelospermum jasminoides Lem. (20.765%) and Woodwardia japonica (L. F.) Sm. (16.372%) in the herbaceous layer were significantly dominant. In Type B, the dominant plants, in order, were Dianella ensifolia Redouté, Carex scaposa C. B. Clarke, Viola diffusa Ging., Woodwardia japonica (L. F.) Sm., and Carex cruciata Wahlenb., among which Dianella ensifolia Redouté (21.136%) and Carex scaposa C. B. Clarke (13.537%) had significantly higher IVs than the other plants in the herbaceous layer. In Type C, the top five herbaceous plants in terms of the importance of plant species were Ardisia japonica (Thunberg) Blume, Carex cruciata Wahlenb., Liriope spicata (Thunb.) Lour., Huperzia serrata (Thunb. ex Murray) Trevis., and Dryopteris fuscipes C. Chr. (in order). Ardisia japonica Blume (51.810%) and Carex cruciata Wahlenb. (10.070%) had significantly higher IVs than the other plants in the herbaceous layer. In Type D, the most important plants were Carex cruciata Wahlenb., Woodwardia japonica (L. F.) Sm., Diplopterygium glaucum (Thunberg ex Houttuyn) Nakai, Alpinia japonica (Thunb.) Miq., and Carex maubertiana Boott (in order). Carex cruciata Wahlenb. (28.048%) and Woodwardia japonica (L. F.) Sm. (16.599%) were the dominant herbaceous plants. In Type E, the top five plants in the herbaceous layer ranked in the order of importance of the plant species were Trachelospermum jasminoides (Lindl.) Lem., Woodwardia japonica (L. F.) Sm., Diplopterygium glaucum (Thunberg ex Houttuyn) Nakai, Carex cruciata Wahlenb., and Sarcandra glabra (Thunb.) Nakai. Trachelospermum jasminoides (Lindl.) Lem. (30.140%) and Woodwardia japonica (L. F.) Sm. (18.558%) were significantly more important than the other herbaceous-layer plants. In Type F, the most dominant plants were Carex cruciata Wahlenb., Diplopterygium glaucum (Thunberg ex Houttuyn) Nakai, Woodwardia japonica (L. F.) Sm., Osmunda japonica Thunb., and Dianella ensifolia (L.) Redouté. Carex cruciata Wahlenb. (53.395%) and Diplopterygium glaucum (Thunberg ex Houttuyn) Nakai (22.166%) were significantly predominant in the herbaceous layer. The IVs of Carex cruciata Wahlenb. generally increased with the increase in the number of years of enclosure, and it became the dominant species in the herbaceous layer of the understory after 40 years of enclosure.

3.2. Species Diversity Indices within the Communities following Different Periods of Enclosure

In the six treatment plots, the Margalef, Simpson, and Shannon–Wiener indices of the adult trees gradually increased with the number of years that the forest had been enclosed (

Table 7. Kruskal–Wallis test of mature arbors after different periods of enclosure.

Analysis Items	Margalef Index	Simpson Index	Shannon–Wiener Index	Pielou Index
p	0.013 **	0.085 *	0.013 **	0.028 **
Cohen’s f value	0.657	0.531	0.661	0.546

Note: **, and * represent 5%, and 10% significance levels, respectively.

and

Table 8. Diversity indices of mature arbors after different periods of enclosure.

Type	Margalef Index	Simpson Index	Shannon–Wiener Index	Pielou Index
A	-	-	-	-
B	0.927 ± 0.179 c	0.673 ± 0.035 c	0.873 ± 0.216 c	0.973 ± 0.025 ab
C	1.187 ± 0.603 c	0.767 ± 0.208 bc	0.893 ± 0.070 c	0.897 ± 0.038 bc
D	2.553 ± 0.681 b	0.893 ± 0.029 ab	1.643 ± 0.439 b	0.993 ± 0.012 a
E	3.263 ± 0.463 b	0.907 ± 0.045 ab	2.283 ± 0.155 a	0.850 ± 0.066 c
F	4.913 ± 0.765 a	0.987 ± 0.023 a	2.637 ± 0.320 a	0.883 ± 0.060 c

Note: Different lowercase letters indicate significant differences at p < 0.05.

). The Margalef index of Type B was significantly different from those of both Types D and F (p < 0.05), and the Margalef index of Type B was 64% and 81% higher than those of Types D and F, respectively. The Simpson index of Type B was significantly different from that of Type F (p < 0.05), while the Simpson index of Type B was 32% higher than that of Type F. The Shannon–Wiener index of Type F was significantly different (p < 0.05) from those of Types D, C, and B. The Shannon–Wiener index of Type F was 61%, 195%, and 202% higher than those of Types D, C, and B, respectively. There was no significant change in the Pielou index of the six sample plots over different periods with the increase in the number of years of enclosure. The Pielou index of Type F was significantly different from that of Type D (p < 0.05), and the Pielou index of Type F was 11% higher than that of Type D. This indicated that most of the diversity indices of the adult arboreal plants increased with the extension of the enclosure period.

The Margalef and Shannon–Wiener indices of the juvenile trees increased with the extension of the periods of enclosure in the six treatment samples (

Table 9. Kruskal–Wallis test of the young arbors after different periods of enclosure.

Analysis Items	Margalef Index	Simpson Index	Shannon–Wiener Index	Pielou Index
p	0.020 **	0.038 **	0.023 **	0.297
Cohen’s f value	0.659	0.579	0.639	0.360

Note: ** represent 5% significance levels, respectively.

and

Table 10. Diversity indices of the young arbors after different periods of enclosure.

Type	Margalef Index	Simpson Index	Shannon–Wiener Index	Pielou Index
A	-	-	-	-
B	2.003 ± 0.165 cd	0.813 ± 0.023 c	1.753 ± 0.170 cd	0.827 ± 0.040 a
C	2.313 ± 0.189 cd	0.820 ± 0.050 bc	1.850 ± 0.285 c	0.867 ± 0.091 a
D	2.560 ± 0.728 c	0.907 ± 0.035 ab	1.987 ± 0.220 c	0.893 ± 0.040 a
E	3.547 ± 0.478 b	0.953 ± 0.015 a	2.437 ± 0.200 b	0.930 ± 0.026 a
F	5.523 ± 0.650 a	0.847 ± 0.051 bc	2.943 ± 0.140 a	0.873 ± 0.064 a

Note: Different lowercase letters indicate significant differences at p < 0.05.

). The Margalef index of Type B was significantly different from those of both Types E and F (p < 0.05), and the Margalef index of Type B was 44% and 64% higher than those of Types E and F, respectively. The Shannon–Wiener index of Type B was significantly different from those of Types E and F (p < 0.05), and the Shannon–Wiener index was 28% and 40% higher than those of Types E and F, respectively. The Simpson index of Type B was significantly different from that of Type F (p < 0.05), and there was no significant difference between the Pielou indices of the six different treatment samples. With the increase in the number of years of enclosure, the Simpson and Pielou indices of the six different treatment samples did not change significantly, indicating that the changes in the diversity index of juvenile arborvitae were less affected by the number of years of enclosure.

In the six treatment plots, the Margalef, Shannon–Wiener, and Pielou indices of the shrub layer increased and then decreased with the extension of the enclosure years (

Table 11. Kruskal–Wallis test of shrubs after different periods of enclosure.

Analysis Items	Margalef Index	Simpson Index	Shannon–Wiener Index	Pielou Index
p	0.027 **	0.114	0.037 **	0.424
Cohen’s f value	0.572	0.420	0.580	0.306

Note: ** represent 5% significance levels, respectively.

and

Table 12. Diversity index of shrubs after different periods of enclosure.

Type	Margalef Index	Simpson Index	Shannon–Wiener Index	Pielou Index
A	0.870 ± 0.164 c	0.863 ± 0.038 a	1.040 ± 0.193 c	0.803 ± 0.091 a
B	1.557 ± 0.155 bc	0.720 ± 0.101 a	1.403 ± 0.227 bc	0.807 ± 0.090 a
C	1.700 ± 0.177 abc	0.870 ± 0.036 a	1.450 ± 0.092 bc	0.817 ± 0.042 a
D	2.453 ± 0.301 ab	0.713 ± 0.059 a	1.913 ± 0.665 ab	0.823 ± 0.076 a
E	3.830 ± 0.300 a	0.800 ± 0.148 a	2.360 ± 0.154 a	0.890 ± 0.044 a
F	2.927 ± 1.650 ab	0.797 ± 0.057 a	2.127 ± 0.206 a	0.883 ± 0.085 a

Note: Different lowercase letters indicate significant differences at p < 0.05.

). The Margalef index of Type E was significantly different from that of Type A (p < 0.05), and the Margalef index of Type E was 340% higher than that of Type A. The Shannon–Wiener indices of Types F and E were significantly different from that of Type A (p < 0.05), and the Shannon–Wiener indices of Types F and E were 105% and 127% higher than that of Type A, respectively. The Simpson index did not change significantly with the increase in the years of enclosure, and there was no significant difference between the Simpson and Pielou indices in the six sample plots with different treatments (p > 0.05). This indicated that, in the plant diversity index of the shrub layer, the change in enclosure years had a greater effect on the Margalef and Shannon–Wiener indices. The Margalef index and Shannon–Wiener index were significantly affected by the change in enclosure years, but the Simpson index and Pielou index were less affected.

In the six treatment plots, the Margalef, Simpson, and Shannon–Wiener indices of the herbaceous layer gradually decreased with the extension of the enclosure years (

Table 13. Kruskal–Wallis test of the herbaceous plants after different periods of enclosure.

Analysis Items	Margalef Index	Simpson Index	Shannon–Wiener Index	Pielou Index
p	0.024 **	0.305	0.030 **	0.294
Cohen’s f value	0.490	0.309	0.564	0.391

Note: ** represent 5% significance levels, respectively.

and

Table 14. Diversity index of the herbaceous plants after different periods of enclosure.

Type	Margalef Index	Simpson Index	Shannon–Wiener Index	Pielou Index
A	3.377 ± 1.960 a	0.830 ± 0.046 a	1.937 ± 0.733 a	0.733 ± 0.045 a
B	2.083 ± 0.701 ab	0.800 ± 0.053 a	1.890 ± 0.280 a	0.813 ± 0.012 a
C	1.587 ± 0.420 b	0.793 ± 0.042 a	1.770 ± 0.246 a	0.737 ± 0.096 a
D	1.520 ± 0.364 b	0.760 ± 0.060 a	1.550 ± 0.178 ab	0.870 ± 0.056 a
E	1.203 ± 0.356 b	0.737 ± 0.151 a	0.990 ± 0.295 bc	0.757 ± 0.095 a
F	0.583 ± 0.278 b	0.720 ± 0.061 a	0.613 ± 0.191 c	0.790 ± 0.089 a

Note: Different lowercase letters indicate significant differences at p < 0.05.

). There was a significant difference (p < 0.05) between the Margalef index of Types F and A, and the Margalef index of Type F was 83% higher than that of Type A. There were significant differences (p < 0.05) in the Shannon–Wiener index values of Types F and C and Types B and A. In addition, the Shannon–Wiener index of Type F was higher than those of Types C, B, and A by 65%, 68%, and 68%, respectively. The Pielou index did not change significantly with the increase in years of enclosure, and there was no significant difference between the Simpson and Pielou indices of the six sample plots with different treatments (p > 0.05), which indicated that, in the plant diversity indices of the herbaceous layer, the change in enclosure years had no significant effect on the Margalef and Shannon–Wiener indices and less on the Simpson and Pielou indices.

3.3. Size of the Bamboo Forest Community Structure and Spatial Distribution of Individuals in Each Period of Enclosure

A total of 421 individual tree-layer plants with DBH ≥ 1 cm were recorded in the study sample (Figure 2). The largest DBH was 69.0 cm, and the average DBH of the plant community was 15.7 cm. A total of 208, 97, 50, 29, 24, 12, and one plants had DBHs of 1–10 cm, 10–20 cm, 20–30 cm, 30–40 cm, 40–50 cm, 50–60 cm, and 60–70 cm, respectively, which accounted for 49.4%, 23.0%, 11.9%, 6.9%, 5.7%, 2.9%, and 0.2% of the total number of individuals in the tree layer, respectively. The number of individuals in the tree layer with a DBH of 1–10 cm comprised the largest percentage of the total number of individuals, and the number of individuals with a diameter in the range of 1–70 cm decreased with the increase in the diameter level. In terms of the distribution of plant size, there were fewer old trees in the tree layer in the study site, and there was a higher proportion of small plants. The distribution of tree-layer plants showed an inverted “J” shape, and the population belonged to the growth type. This indicated that the tree-layer plant population in the Phyllostachys edulis forest community had a high reserve of forest renewal, and the rate of renewal was rapid and very effective. This indicates that the plant population in the tree layer of the Phyllostachys edulis forest community has a high reserve of forest trees, the forest trees renewed quickly, the state of renewal is very good, and the community is in a relatively stable state. The years of enclosure had some effect on the spatial distribution of the six different individuals in the same site (Figure 3). When the duration of enclosure was 3 years, the tree layer was densely distributed in the upper right part of the sample site and sparsely distributed in the middle and lower parts of the sample site. When the period of enclosure was 15 years, the tree layer was more densely distributed in the lower left part of the sample plot. When the enclosure period was 40 years, the tree-layer plants were densely distributed in the upper left, lower left, and upper right of the sample plots, and they were sparsely distributed in the lower middle and lower right of the sample plots. Overall, the layers of trees within the Phyllostachys edulis forest community varied in the sparseness of their individual distribution within the sample plot.

3.4. Spatial Distribution Pattern of Moso Bamboo Forest Communities with Different Enclosure Periods

The year of enclosure had an effect on the pattern of spatial distribution of the populations in the six different sites (Figure 4). When the period of enclosure was 3 years, the estimates of arborvitae plants at the 0–3.5 cm scale were above the confidence interval, and the plants of the community showed a clustered distribution. When the period of enclosure was 5 years, the estimates of the arborvitae at the 0–4 cm scale were above the confidence interval, and the plants in the community showed a clustered distribution. When the enclosure period was 10 years, the estimates of the arborvitae at the 0–7 cm scale were below the confidence interval, and the plants in the community were uniformly distributed. When the enclosure period was 15 years, the estimates of the arborvitae at the 0–7 cm scale were above the confidence interval, and the plants in the community showed a clustered distribution. At 40 years of enclosure, the estimates of the arborvitae at the 0–7 cm scale were above the confidence interval, and the plants in the community showed a clustered distribution. This indicates that the pattern of spatial distribution of the arborvitae in the Phyllostachys edulis forest community was generally clustered with the extension of the enclosure period.

4. Discussion

The relative importance of plants within a community can be expressed by the IVs of plants, and the IV is also one of the important indicators to assess the amount of plant diversity within a population. In this study, with the extension of the period of enclosure, the IVs of Machilus thunbergii Sieb. et Zucc., Castanopsis faberi Hance, and Pinus massoniana Lamb. increased continuously, and they gradually became the dominant species of adult trees. The IVs of Cunninghamia lanceolata (Lamb.) Hook.tended to decrease, increase, and then decrease again, and it became the dominant plant of adult trees within the community when the forests had been enclosed for 15 years. This study showed that Cunninghamia lanceolata (Lamb.) Hook. and Pinus massoniana are the establishment species of subtropical evergreen coniferous forests in China, with a strong adaptive ability, and the vegetation gradually transitioned to mixed coniferous and broadleaf forests during the process of sequestration; they finally developed into a plant community dominated by evergreen broadleaf forests [29]. In the process, Machilus thunbergii Sieb. et Zucc. and Castanopsis faberi Hance were the dominant zonal plants, and they were the top species in the succession [30]. In this study, the IVs of Eurya rubiginosa var. attenuata H.T.Chang in the shrub layer showed a general trend of increasing and then decreasing, and it became the dominant species in the shrub layer under the Phyllostachys edulis forest after 10 and 15 years after enclosure, while the IVs of Carex cruciata in the herbaceous layer generally increased. It became the dominant species in the herbaceous layer under the Phyllostachys edulis forest after 40 years of enclosure. The results showed that Eurya rubiginosa var. attenuata H.T.Chang has shallow roots and usually grows in the understory and barren slope thickets on mountains and hilltops. It is a common understory shrub-layer plant in subtropical broad-leaved evergreen forests in China [31,32], and thus, it can become the dominant species of shrub layer vegetation after 10 and 15 years after enclosure. Carex cruciata Wahlenb. has a well-developed root system and a strong ability to adapt to the environment, and it can grow tenaciously in different environments; this makes it an important constituent of the herbaceous layer [33]. It gradually eliminated the other species and became the absolute dominant plant in the herbaceous layer as the number of years of enclosure increased.

The restoration of species diversity and community structure within forests is an important task for modern researchers [34]. In this study, the shrub layer dominance Simpson index was lower in the community after 10 years of enclosure than in the other communities, primarily owing to the presence of the absolutely dominant shrub plant, Eurya rubiginosa var. attenuata H.T.Chang, in the community after 10 years of enclosure. The Shannon–Wiener index of plant diversity in the shrub layer tended to increase and then decrease with the increasing period of enclosure. With the further increase in the period of enclosure, the diversity of shrub layer vegetation tended to decrease, primarily owing to the increase in the density of Phyllostachys edulis and trees. Consequently, there is an increasingly large degree of canopy enclosure; the penetration of sunlight is weakened, and the competition between plants is more intense, which inhibits the growth of shrubs [35,36]. The shrub layer uniformity Pielou index tended to increase and then decrease with the period of enclosure, and the shrub layer uniformity Pielou index was lower in the community that had not been subjected to a full year of enclosure. This may be owing to the fact that the species richness, S, of the shrub layer was higher in the zero-year community, which increased the denominator of Jsw = H′/InS and led to a decrease in the Pielou index. The vegetation evenness index and the Pielou index were positively correlated with the diversity index and the Shannon–Wiener index and were negatively correlated with species richness, S. The results of other studies were similar to those of this study.

In this study, the Margalef richness index of the herbaceous layer tended to decrease as the period of enclosure increased, and the Simpson dominance index and the Shannon–Wiener diversity index were negatively correlated with the age of enclosure. With the increase in the enclosure time, the stand enclosure gradually increased, which led to a significant decrease in the light available to the herbaceous plants. Simultaneously, the increase in stand density led to a more intensive competitive environment for the herbaceous plants, so that only the better adapted plant species could survive and become the absolute dominant species in the herbaceous layer. This pattern resulted in a gradual decrease in the diversity of the plant species and the trend of the gradual decline of the herbaceous-layer plants [37].

The distribution of plants according to the DBH or the percentage of different DBH sizes in a community is called the diameter structure of the community, and the current status of the population structure and renewal strategy can be reflected by the diameter structure, which can also be used as an effective method to predict the direction of the succession of a population or a community [38,39]. In this study, the structure of the adult trees was determined by the size of the tree. For example, the average diameter class of the adult trees tended to increase in parallel with the increase in the period of enclosure, which made them part of the growing population. They showed that the number of young trees in these communities increased with the duration of enclosure; the vegetation regenerated effectively, and the number of plants in the communities remained stable or continued to increase [40]. The results of this study showed that the number of young trees in these communities is increasing with the time of enclosure.

The spatial pattern of distribution of the plant populations can be represented by a quantitative expression of the distribution of populations in two dimensions [41], which is closely related to scale and changes with scale [42]. In this study, the spatial pattern of distribution of adult arborvitae plants within the community was broadly shown as an aggregated distribution with the extension of the years of enclosure. This is related to the scale of the study, where the spatial distribution pattern of plant populations is primarily related to the biological characteristics of the plants within the populations [43]. The pattern of spatial distribution of the plant populations at large scales is affected by different factors, and this pattern of plant populations at different scales is not the same. The research sample sites were selected in a small-scale environment, which resulted in the spatial distribution pattern of adult tree plants being mostly aggregated. With the increase in the time of enclosure, the need for nutrients, light, and other materials of plants in the community gradually increased, and the competition between different individuals was also strengthened. Thus, many individuals of different species of plants died along with those within the same species owing to the intense competition, the phenomenon of self-thinning, and other thinning. This resulted in a substantial reduction in population density, and the intensity of aggregation within the population was also weakened [41]. However, this phenomenon differs from the pattern derived from the results of this study. This may be owing to the fact that there were still many Phyllostachys edulis plants in the study area. Thus, the competition within the same species and between different species did not lead to a gradual decrease in the aggregation strength of the adult tree plant populations with the development of closed succession. In contrast, with the development of closed succession, the spatial distribution pattern of adult tree plant populations was mostly aggregated, which led to the conclusion that is inconsistent with the phenomenon of self-thinning and other thinning.

5. Conclusions

Comprehensive analyses showed that, with the increase in the time of enclosure, the dominance of adult trees in the community increased and gradually formed a symbiotic pattern with Phyllostachys edulis. The spatial pattern of distribution of the populations was generally in a state of aggregation, and it was apparent that the Phyllostachys edulis forest community under prolonged enclosure conditions responded in a positive direction and ultimately formed a more stable Phyllostachys edulis mixed forest community. In order to better understand the effects of Phyllostachys edulis succession on ecosystem functions, it is necessary to study the growth and environmental changes of Phyllostachys edulis.