Patterns and determinants of plant biodiversity in non-commercial forests of eastern China

Abstract

Non-commercial forests represent important habitats for the maintenance of biodiversity and ecosystem function in China, yet no studies have explored the patterns and determinants of plant biodiversity in these human dominated landscapes. Here we test the influence of (1) forest type (pine, mixed, and broad-leaved), (2) disturbance history, and (3) environmental factors, on tree species richness and composition in 600 study plots in eastern China. In total, we found 143 species in 53 families of woody plants, with a number of species rare and endemic in the study region. Species richness in mixed forest and broad-leaved forest was higher than that in pine forest, and was higher in forests with less disturbance. Species composition was influenced by environment factors in different ways in different forest types, with important variables including elevation, soil depth and aspect. Surprisingly, we found little effect of forest age after disturbance on species composition. Most non-commercial forests in this region are dominated by species poor pine forests and mixed young forests. As such, our results highlight the importance of broad-leaved forests for regional plant biodiversity conservation. To increase the representation of broad-leaved non-commercial forests, specific management practices such as thinning of pine trees could be undertaken.

About the Authors:

Chuping Wu

Roles Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Writing – original draft

Affiliations College of Life Sciences, Zhejiang University, Hangzhou, Zhejiang, China, Zhejiang Academy of Forestry, Hangzhou, Zhejiang, China

Mark Vellend

Roles Conceptualization, Formal analysis, Writing – review & editing

Affiliation: Departement de biologie, Université de Sherbrooke, Sherbrooke, QC, Canada

Weigao Yuan

Roles Conceptualization, Funding acquisition

Affiliation: Zhejiang Academy of Forestry, Hangzhou, Zhejiang, China

Bo Jiang

Roles Conceptualization, Resources

Affiliation: Zhejiang Academy of Forestry, Hangzhou, Zhejiang, China

Jiajia Liu

Roles Conceptualization, Formal analysis, Writing – review & editing

Affiliation: College of Life Sciences, Zhejiang University, Hangzhou, Zhejiang, China

Aihua Shen

Roles Conceptualization, Investigation

Affiliation: Zhejiang Academy of Forestry, Hangzhou, Zhejiang, China

Jinliang Liu

Roles Conceptualization, Formal analysis, Methodology

Affiliation: College of Life Sciences, Zhejiang University, Hangzhou, Zhejiang, China

Jinru Zhu

Roles Conceptualization, Investigation

Affiliation: Zhejiang Academy of Forestry, Hangzhou, Zhejiang, China

Mingjian Yu

Roles Conceptualization, Methodology, Writing – review & editing

* E-mail: [email protected]

Affiliation: College of Life Sciences, Zhejiang University, Hangzhou, Zhejiang, China

ORCID http://orcid.org/0000-0001-8060-8427

Abstract

Non-commercial forests represent important habitats for the maintenance of biodiversity and ecosystem function in China, yet no studies have explored the patterns and determinants of plant biodiversity in these human dominated landscapes. Here we test the influence of (1) forest type (pine, mixed, and broad-leaved), (2) disturbance history, and (3) environmental factors, on tree species richness and composition in 600 study plots in eastern China. In total, we found 143 species in 53 families of woody plants, with a number of species rare and endemic in the study region. Species richness in mixed forest and broad-leaved forest was higher than that in pine forest, and was higher in forests with less disturbance. Species composition was influenced by environment factors in different ways in different forest types, with important variables including elevation, soil depth and aspect. Surprisingly, we found little effect of forest age after disturbance on species composition. Most non-commercial forests in this region are dominated by species poor pine forests and mixed young forests. As such, our results highlight the importance of broad-leaved forests for regional plant biodiversity conservation. To increase the representation of broad-leaved non-commercial forests, specific management practices such as thinning of pine trees could be undertaken.

Figures

Fig 4

Table 4

Table 5

Fig 1

Table 1

Fig 2

Table 2

Fig 3

Table 3

Fig 4

Table 4

Table 5

Fig 1

Table 1

Fig 2

Citation: Wu C, Vellend M, Yuan W, Jiang B, Liu J, Shen A, et al. (2017) Patterns and determinants of plant biodiversity in non-commercial forests of eastern China. PLoS ONE12(11): e0188409. https://doi.org/10.1371/journal.pone.0188409

Editor: RunGuo Zang, Chinese Academy of Forestry, CHINA

Received: August 18, 2017; Accepted: November 6, 2017; Published: November 21, 2017

Copyright: © 2017 Wu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the paper and its Supporting Information files.

Funding: This research was funded by Major Cooperation Project of Zhejiang Province and Chinese Academy of Forestry (No. 2014SY01), http://www.caf.ac.cn/; Science and Technology Project of Zhejiang Province (No. 2015C02016), http://www.zjinfo.gov.cn/; Zhejiang Provincial Natural Science Foundation (No. LY16C160003), http://www.zjinfo.gov.cn/. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

1. Introduction

Balancing economic development and biodiversity conservation is a major challenge for forest management, especially in developing countries [1, 2]. To address this issue, many countries have initiated forest management programs that distinguish different forest types, such as commercial forest and non-commercial forest [3, 4]. Non-commercial forest is protected from human intervention with a focus on the maintenance of forest ecosystem functions (e.g., carbon fixation, water filtration) and biodiversity, while commercial forest is the source of timber consumption. Non-commercial forests are under changing management regimes and might play an important role in biodiversity conservation in human dominated landscapes [5].

Non-commercial forest occupies a large area globally. For example, the area of non-commercial forest in China is roughly 85 million ha (about 41.6% China’s forest land) [6], consisting largely of secondary forests and plantations [2], such as Masson pine (Pinus massoniana) forests, which are widely distributed in China [7]. Many planted and natural forests are dominated by just one or a few tree species that are deemed to have relatively low conservation value compared to primary forests [8, 9]. In order to improve biodiversity and ecosystem functions of monocultures and young forests via acceleration of succession, management actions include thinning, planting of broad-leaved trees, and reducing competition around (i.e., singling) broad-leaved individuals [10, 11]. For example, Meng et al. [10] used thinning and enrichment planting to transform a degraded Masson pine plantation to a mixed forest with increased species diversity. However, no studies have compared biodiversity in different types of non-commercial forests in China, especially at broad scales.

Multiple factors can potentially affect biodiversity in non-commercial forests, which are products of natural environmental variation and past anthropogenic activities. Firstly, as found in most studies on primary forests, biodiversity patterns can be influenced by environmental factors [12–18], such as soil and topography. Secondly, dominant canopy trees (captured in forest type classifications) can be an important determinant of species diversity in the rest of the community [19], while also being related to environmental gradients. In eastern China, pine forest, mixed forest, and broad-leaved forest are the main types of non-commercial forest. Thirdly, disturbance history can have a strong influence on species richness, community structure and species composition [20, 21]. For example, as succession proceeds after disturbance, late successional tree species typically increase in abundance as early successional species decline, gradually converging on the species composition of mature forests [22, 23]. These shifts of biodiversity patterns in terms of plant functional traits, species composition and ecosystem function are often predicted by the age since abandonment [24]. Most non-commercial forests in China have regenerated since the 1960s, with variable disturbance histories, but how these disturbances influence biodiversity patterns are rarely studied.

Due to limited datasets combining information on all of these anthropogenic and environmental factors, we still lack a systematic understanding of biodiversity patterns and their determinants in non-commercial forests of China. In this paper, we present an analysis exploring the patterns and determinants of tree biodiversity in non-commercial forests in eastern China. We chose Zhejiang province as a case study as this region is mainly dominated by non-commercial forest, and local governments have made major investments in forest management for conserving biodiversity and ecosystem functioning. We aimed to test the influence of (1) forest type (pine, mixed, and broad-leaved), (2) forest age after disturbance and (3) environmental factors, on tree species diversity and composition among survey plots. The results can provide scientific guidance to the management of non-commercial forests in this region.

2. Material and methods

2.1 Study area

Our study was conducted in the subtropical forest region of eastern China (Fig 1). The region experiences a subtropical monsoon climate with mean annual precipitation of 980–2000 mm, and mean annual temperature of 15–18°C. July mean temperature maximum ranges from 33–43°C, and February mean minimum temperature ranges from -2.2 to -17.4°C (data from www.zj.gov.cn). Zonal mature vegetation in this region is mainly subtropical evergreen broad-leaved forest, in which evergreen broad-leaved trees dominate in the canopy layer [25]. Non-commercial forests of Zhejiang covered roughly 2.68 million ha in 2014, accounting for 25.7% of the total land area (10.41 million ha) and 40.6% of the forested land area (6.60 million ha). According to a 2010 report [26], pine-dominated forest accounted for 35.5% of the forested area, mixed broad-leaved-conifer forest 13.7%, broad-leaved forest 28.5%, and other forest types, such as Chinese fir, bamboo forest, and shrub forest, 22.3%.

[Figure omitted. See PDF.]

Fig 1. Geographic distribution of survey plots.

PF: pine forest; MF: mixed forest; BF: broad-leaved forest. China map is from http://www.tianditu.gov.cn/.

https://doi.org/10.1371/journal.pone.0188409.g001

2.2 Study design

To monitor forest vegetation, the Forestry Department of Zhejiang province set up and surveyed 600 plots, each 20 × 20 m, in non- commercial forests from 2010 to 2015 (Fig 1). These plots were evenly distributed across the study region and reflected the main stand types characteristic of the local vegetation. All trees with diameter at breast height (DBH) ≥ 5 cm were identified and measured. Each plot was first classified as one of three types (pine forest, mixed forest, broad-leaved forest) using importance values (IV), calculated for a given species as IV = 100 × (relative density + relative frequency + relative basal area)/3. Relative frequency in every 20 × 20 m plot was calculated from 16 subplots (5 × 5 m). If the IV for pine was > 66.7%, the plot was classified as pine forest (n = 169 plots); if it was between 33.3% and 66.7%, the plot was classified as mixed forest (n = 170); if < 33.3% (i.e., the summed IV of broad-leaved species was > 66.7%), the plot was defined as broad-leaved forest (n = 261). Each plot was further classified according to forest age–stage 1 (pine forest: 30 plots, mixed forest: 32 plots, broad-leaved forest: 96 plots), stage 2 (46, 58, 125), and stage 3 (93, 80, 40)–based on the average age of the dominant trees in the plot (Table 1). Stage 1 stands are <20 years old, stage 2 stands 20–60 years old, and stage 3 stands >60 years old. Forest age after disturbance was estimated based on the records of cutting history in a given plot. In addition to tree composition data, for each plot we also estimated total canopy density, soil depth, humus depth, litter depth, elevation, slope, slope position, and aspect. Canopy density was estimated as the mean of visual estimates at 48 points per plot, with three points randomly set in each of the sixteen 5 × 5 m subplots. We also measured slope, aspect, elevation, longitude and latitude by compass and GPS.

[Figure omitted. See PDF.]

Table 1. Total number of plots, plant families, genera and species of trees in different forest types.

https://doi.org/10.1371/journal.pone.0188409.t001

2.3 Data analyses

In order to explore variation in species composition in different forest types, we first report importance values of the tree species recalculated by combining all plots of a given forest type. The relative frequency in a given forest type was calculated for all 20 × 20 m plots of that forest type.

We used one-way ANOVAs to test for differences in species richness among forest types. A generalized linear model was conducted to predict species richness as a function of environmental variables (i.e. forest age, canopy density, soil depth, humus depth, litter depth, elevation, slope, slope position, aspect). Before GLM analysis, we implemented a test for overdispersion using the function of dispersiontest in the R package AER [27]. Because our data showed significant overdispersion, we used a negative binomial error distribution with a log link function for species richness (count data). This analysis was performed using the “glm.nb” function from the MASS package in R Version 3.2.4 (R Core Team, 2016). To determine which of these explanatory variables were most important in predicting species richness, we used the “dredge” function in the MUMIN package, which is used to identify significant predictors in multiple models, to test all possible combinations of the environmental variables; we then used Akaike’s information criterion (AICc), corrected for small sample sizes, to select the best model [28]. We calculated a commonly used index, i.e. the Bray-Curtis index, as a metric of compositional dissimilarity between all pairs of plots [29, 30], as the input for non-metric multidimensional scaling (NMDS) ordinations (with 999 iterations), one for each forest type. These analyses were performed using the function “metaMDS” in the vegan package of R. In order to focus on the dominant components of the vegetation, rare species (<10 stems) were removed prior to the NMDS analyses, because their occurrences in the dataset can depend on chance and potentially distort the ordination [31, 32]. The significance of environmental variables in predicting species composition were tested using the function “envfit” in the vegan package [33].

3. Results

3.1 Species richness in different forest types

Overall, across all plots we found 143 species in 53 plant families. Broad-leaved forests had the greatest total diversity, with 128 species and 50 families, followed by mixed forests and pine forests (Table 1).

Species richness per plot was greater in mixed forest and broad-leaved forest than in pine forest (Fig 2). Species richness was qualitatively greater in broad-leaved forest than in mixed forest, but the differences were not significant. The best predictors of species richness were elevation and forest age in mixed forest and broad-leaved forest, and canopy density in pine forest (Table 2).

[Figure omitted. See PDF.]

Fig 2. Species richness per plot in three forest types.

PF: pine forest; MF: mixed forest; BF: broad-leaved forest.

https://doi.org/10.1371/journal.pone.0188409.g002

[Figure omitted. See PDF.]

Table 2. Results of a generalized linear model predicting species richness in each forest type as a function of environmental variables (i.e., forest age, canopy density, soil depth, humus depth, litter depth, elevation, slope, aspect, slope position).

https://doi.org/10.1371/journal.pone.0188409.t002

3.2 Species composition

The NMDS analysis revealed clear distinctions in tree species composition among the three forest types, as expected given that forest types are defined by their dominant trees (Fig 3). The average importance value (IV) of conifers such as Pinus massoniana was 70.1%, 40.8%, and 9.8% in pine forest, mixed forest and broad-leaved forest, respectively. The broad-leaved trees in pine forest were largely early successional species, including deciduous broad-leaved trees (e.g., Quercus fabri, Liquidambar formosana), and evergreen broad-leaved trees (e.g., Schima superba, which is early successional species and can grow in late stages). Deciduous broad-leaved trees and evergreen broad-leaved trees had similar dominance levels in mixed forest. Evergreen broad-leaved trees dominated in broad-leaved forest (Table 3).

[Figure omitted. See PDF.]

Fig 3. Two dimensional Non-Metric Multidimensional Scaling (NMDS) ordination diagram of all forest types together.

PF: pine forest; MF: mixed forest; BF: broad-leaved forest.

https://doi.org/10.1371/journal.pone.0188409.g003

[Figure omitted. See PDF.]

Table 3. Importance values in each forest type of the 10 most common tree species.

https://doi.org/10.1371/journal.pone.0188409.t003

3.3 Influence of environmental factors on species composition

We analyzed the influence of environmental factors on species composition in each forest type separately (Fig 4, Table 4). In broad-leaved forests, soil depth (R2 = 0.074, P = 0.001) was the best predictor of species composition followed by elevation (R2 = 0.061, P = 0.001), canopy density (R2 = 0.056, P = 0.002), slope (R2 = 0.050, P = 0.003), litter depth (R2 = 0.026, P = 0.025), and aspect (R2 = 0.024, P = 0.044). Species composition of mixed forests was significantly affected by elevation (R2 = 0.050, P = 0.023), and aspect (R2 = 0.044, P = 0.031). Species composition in pine forests was explained by elevation (R2 = 0.064, P = 0.002) and canopy density (R2 = 0.046, P = 0.028).

[Figure omitted. See PDF.]

Fig 4. NMDS ordinations of species composition and environmental factors in three forest types.

A: pine-dominated forest; B: mixed broad-leaved-conifer forest; C: broad-leaved forest. SL: slope; SP: slope position; AS: aspect; EL: elevation; SD: soil depth; HD: humus depth; LD: litter depth; CD: canopy density; AG: age.

https://doi.org/10.1371/journal.pone.0188409.g004

[Figure omitted. See PDF.]

Table 4. Significant correlations of environmental variables with NMDS axes.

https://doi.org/10.1371/journal.pone.0188409.t004

4. Discussion

4.1 Biodiversity in non-commercial forests

To improve biodiversity conservation in China’s forests, non-commercial forests have been protected from anthropogenic disturbances and currently occupy 41.6% of China’s forested land [6]. Our study, to our knowledge, is the first to investigate biodiversity patterns in non-commercial forests at the regional level with a large dataset. In total, we found 143 species across 600 plots. These forests harbor some rare species, e.g. Emmenopterys henryi and Cercidiphyllum japonicum (both are Grade II of National Key Protected Wild Plants) and some commercially valuable species, e.g. Phoebe sheareri and Zelkova serrata. In addition, these non-commercial forests provide habitat for a wide variety of animals such as insects and vertebrates [34–36]. Therefore, we suggest that these non-commercial forests play an important role in protecting regional biodiversity. However, biodiversity varies significantly from site to site. For example, tree species richness is clearly lower in pine forest than in mixed and broad-leaved forest, the latter of which have more late successional tree species (Fig 2). This is consistent with other studies that have found higher functional diversity, more late successional species, and high ecosystem functions in broad-leaved forests compared to pine forests [19, 37]. The conservation value of non-commercial forests clearly depends on local environmental conditions and forest type.

4.2 Determinants of plant biodiversity

Biodiversity patterns are driven by various factors such as landscape history, environmental variables, and anthropogenic activities [38–40]. Here, in non-commercial forests, we found that plant species richness and composition are mainly driven by forest type and environmental factors, rather than forest age (Figs 2–4; Table 2; Table 4).

Species richness and composition varied significantly among forest types, with more late successional tree species in the broad-leaved forests (Fig 3, Table 3). In eastern China, pine forest, mixed forest, and broad-leaved forest are the main forest types. Pinus massoniana is a fast-growing, dominant species during early succession, with a high average importance value (60.9%). The broad-leaved trees in pine forests were largely early successional species, such as Quercus fabri and Liquidambar formosana. In some nutrient-poor sites (e.g. ridge tops), pine forest will persist for a long time [41]. In some environments (e.g. valleys), shade-tolerant trees (mostly evergreen broad-leaved trees) can increase in abundance and richness while pines die out as succession proceeds. Therefore, evergreen broad-leaved trees such as Cyclobalanopsis glauca, Schima superba, and Castanopsis sclerophylla, dominated broad-leaved forests. Similarly, a study by Wang et al. [42] showed that mixed and broad-leaved forests contain both early and late successional species, together contributing to high species richness. Yuan et al. [43] also found more late successional tree species in mature forests, and a greater representation of fast growing shrub species in pine forests. These patterns pertain not only to taxonomic diversity: functional diversity and ecosystem functions are also reported to be higher in broad-leaved forests than in the other two forest types [19]. Shifting from pine forest to broad-leaved forest is a long-term process, lasting many decades [21], although disturbance such as thinning pine or pine wilt disease should accelerate succession [44]. Hence, our study confirmed the importance of broad-leaved forests for biodiversity conservation in this region.

Environmental factors affect the distribution of plant species and create spatial patterns of biodiversity [45–47]. Our results were similar to some studies in subtropical regions where environmental conditions explained about 20% or less of the variance in species composition [2, 48–50]. In terms of specific environmental variables, we found that elevation and soil depth are especially important in driving species composition patterns (Table 4), highlighting the importance of topography and soil in shaping community structure [51, 52]. The relatively low explanatory power of these models suggest that species composition of non-commercial forests is determined to a considerable degree by other unmeasured factors, such as landscape context, land use history, or climate variables, all of which require further research [39, 53, 54]. Interestingly, species composition among broad-leaved forests was more strongly predicted by environmental factors. This suggests that stochastic colonization and establishment might be more important determinants of species composition in pine and mixed forests [55]. In broad-leaved forests, species might be more strongly filtered by environmental variables [54, 56]. Given the strong compositional differences among forest types, our study also suggests that high habitat heterogeneity supports greater species diversity [57].

Species richness is positively correlated with forest age after disturbance in mixed and broad-leaved forests (Table 2), indicating that species richness can increase rapidly after disturbance. However, species richness in pine forests was not affected, which might due to the long-term persistence of the dominant Pinus massoniana tree, which can persist for more than 150 years, potentially preventing colonization by new species [41]. Surprisingly, we didn’t find any effects of forest age on species composition (Fig 4). This result contradicts many previous studies that found directional trends in species composition during succession [58, 59], with more late successional tree species increasing over time [60]. This result might have several causes. First, landscape context can play an important role in driving biodiversity patterns. For example, most of the non-commercial forests in Zhejiang province are fragmented, but we did not evaluate this possible effect [61], which can potentially favor early successional species and therefore counter the expected effect of forest age [62, 63]. Second, the recovery of species richness and composition in secondary forest is generally slow process [64], which might take even more than a hundred years [21]. It is possible that the age differences among our forests (decades) were not sufficient to reveal differences that will emerge over longer periods of time (e.g., centuries). These possibilities are in need of future study.

4.3 Implications of future forest management

Anthropogenic activities play an increasingly dominant role in determining forest structure and composition, making forest management an increasingly important global issue [65]. Forest management practices have traditionally paid little attention to biodiversity [66]. Here we focused on biodiversity in non-commercial forests, and our findings provide several implications for forest management.

First, we found that most of this region’s non-commercial forests (i.e., pine dominated) are of relatively low conservation value. Hence, an increase in forest area does not necessarily translate into increasing biodiversity. Biodiversity in newly planted pine forests are much lower than that in old growth broad-leaved forests [7, 8]. Conifer forests dominated by early successional species have expanded considerably in coverage, mainly due to large scale tree plantations establishment in the 1970s [67]. Mature broad-leaved forests are especially important for biodiversity conservation, representing a restoration target, but account for only a small portion of non-commercial forests. Therefore, conservation efforts aimed at biodiversity conservation should focus on promoting attributes found in mature broad-leaved forests [9, 68].

Secondary forests, created by logging in recent decades, are common in eastern China, and to the extent that pine forests can be converted to mixed and broad-leaved forests [22, 43], local biodiversity would be enhanced. However, natural succession is a long-term process, affected by many different factors [21, 69]. Forest management activities can potentially accelerate forest succession. For example, mixed broad-leaved-conifer forest dominated by Pinus massoniana in Qiandao Lake National Forest Park of Zhejiang Province, developed rapidly toward succession of evergreen broad-leaved forest after six years of selective cutting of pines [70]. Thinning pine and planting broad-leaved trees is also an effective method of pine forest transformation to broad-leaved forest [10]. Planting can begin with native early successional broad-leaved trees such as Liquidambar formosana, which can grow in pine forests (Table 3).

Finally, management activities can be tailored to different forest types (Table 5). For example, while introducing broad-leaved species to pine forests might be necessary, in mixed forests many target species are already present [43], so their growth can simply be enhanced (e.g., by crop tree release and fertilization). For the relatively rare mature broad-leaved forests, strict protection from anthropogenic disturbances can help promote biodiversity conservation.

[Figure omitted. See PDF.]

Table 5. Current status and management activities in different forest types in this region.

https://doi.org/10.1371/journal.pone.0188409.t005

Supporting information

[Figure omitted. See PDF.]

S1 Table. Environmental data of the 600 plots.

https://doi.org/10.1371/journal.pone.0188409.s001

(XLSX)

S2 Table. Species list of the 600 plots.

https://doi.org/10.1371/journal.pone.0188409.s002

(XLSX)

Acknowledgments

We are grateful to Jun Zhang, Yuejie Huang, Tingting Li, Jiejie Jiao and many students and volunteers for their assistance in the field and lab. We thank the administrations of the all study sites and local communities for permitting us to access the study sites and for their collaboration during fieldwork. We also thank Hong Qian, Ping Ding, Yi Jin for their constructive comments.

Citation: Wu C, Vellend M, Yuan W, Jiang B, Liu J, Shen A, et al. (2017) Patterns and determinants of plant biodiversity in non-commercial forests of eastern China. PLoS ONE12(11): e0188409. https://doi.org/10.1371/journal.pone.0188409

References

1. Cubbage F, Harou P, Sills E. Policy instruments to enhance multi-functional forest management. Forest Policy and Economics. 2007;9:833–851.

2. Liu SR, Dai LM, Wen YG, Wang H. A review on forest ecosystem management towards ecosystem services: status, challenges, and future perspectives. Acta Ecologica Sinica. 2015;35:1–9. (in Chinese, with English abstract).

3. Ticktin T, Nantel P, Ramirez F, Johns T. Effects of variation on harvest limits for nontimber forest species in Mexico. Conservation Biology. 2002;16:691–705.

4. Jones ET, Lynch KA. Nontimber forest products and biodiversity management in the Pacific Northwest. Forest Ecology and Management. 2007;246:29–37.

5. Sodhi NS, Koh LP, Clements R, Wanger TC, Hill JK, Hamer KC, et al. Conserving Southeast Asian forest biodiversity in human-modified landscapes. Biological Conservation. 2010;143:2375–2384.

6. FAO. China-Global Forest Resources Assessment 2015- Country Report. UN Food and Agriculture Organization, Rome. 2015:Available at: http://www.fao.org/documents/card/en/c/a6afe3a7-0fb6-4920-bb18-ae3465c2f0a9/.

7. Xu JC. China’s new forests aren’t as green as they seem. Nature. 2011;477:371. pmid:21938029

8. Hua F, Wang X, Zheng X, Fisher B, Wang L, Zhu J, et al. Opportunities for biodiversity gains under the world's largest reforestation programme. Nature Communications. 2016. pmid:27598524

9. Gibson L, Lee TM, Koh LP, Brook BW, Gardner TA, Barlow J, et al. Primary forests are irreplaceable for sustaining tropical biodiversity. Nature. 2011;478:378–381. pmid:21918513

10. Meng JH, Lu YC, Zeng J. Transformation of a degraded Pinus massoniana plantation into a mixed-species irregular forest: impacts on stand structure and growth in southern China. Forests. 2014;5:3199–3221.

11. Zeng DH, Jiang FQ, Guo H, Fan ZP, Du XJ. Transformation and assessment of Chinese pine pure plantations for soil and water conservation in western Liaoning Province, P.R. China. Journal of Forestry Research. 2004;15:127–134.

12. Marcelo T, Jose MC, Claude G. Forest fragmentation, synergisms and the impove shment of Neotropieal forests. Biodiversity and Conservation. 2004;13:1419–1425.

13. Davies RG, Orme CD, Webster AJ, Jones KE, Blackburn TM, Gaston KJ. Environmental predictors of global parrot (Aves: Psittaciformes) species richness and phylogenetic diversity. Global Ecology and Biogeography. 2007;16:220–233.

14. Qian H. Environmental determinants of woody plant diversity at a regional scale in china. PLoS ONE. 2013.

15. Gilbert B, Lechowicz MJ. Neutrality, niches, and dispersal in a temperate forest understory. Proceedings of the National Academy of Science of the United Sataes of America. 2004;101:7651–7656.

16. Legendre P, Mi XC, Ren HB, Ma KP, Yu MJ, Sun IF, et al. Partitioning beta diversity in a subtropical broad-leaved forest of China. Ecology. 2009;90:663–674. pmid:19341137

17. Lu P, Jin Y, Chen JH, Li MH, Yu MJ. Influences of geographical distance and topographic difference on β diversity of two large-scale forest dynamics plots. Biodiversity Science. 2013;21:554–563. (in Chines, with English abstract).

18. Shen GC, He FL, Waagepetersen R, Sun IF, Hao ZQ, Chen ZS, et al. Quantifying effects of habitat heterogeneity and other clustering processes on spatial distributions of tree species. Ecology. 2013;94:2436–2443. pmid:24400495

19. Maeshiro R, Kusumoto B, Fujii S, Shiono T, Kubota Y. Using tree functional diversity to evaluate management impacts in a subtropical forest. Ecosphere. 2013.

20. Osazuwa-Peters OL, Jimenez I, Oberle B, Chapman CA, Zanne AE. Selective logging: do rates of forest turnover in stems, species composition and functional traits decrease with time since disturbance?—A 45 year perspective. Forest Ecology and Management. 2015;357:10–21. pmid:26339115

21. Liebsch D, Marques MC, Goldenberg R. How long does the Atlantic Rain Forest take to recover after a disturbance? Changes in species composition and ecological features during secondary succession. Biological Conservation. 2008;141:1717–1725.

22. Shen Q, Zhang J, Zhu JR, Jiang B, Ge Y, Liu QX, et al. Changes of species composition and diversity in the restoration processes of ecological public-welfare forests in Zhejiang, East China. Acta Ecologica Sinica. 2005;25:2131–2138. (in Chinese, with English abstract).

23. Hu ZH, Yu MJ. Study on successions sequence of evergreen broad-leaved forest in Gutian Mountain of Zhejiang, Eastern China: species diversity. Frontiers of Biology in China. 2008;3:45–49.

24. Garnier E, Cortez J, Billes G, Navas ML, Roumet C, Debussche M, et al. Plant functional markers capture ecosystem properties during secondary succession. Ecology. 2004;85:2630–2637.

25. ZY. W. Vegetation of China. Science Press, Beijing. (in Chinese). 1995.

26. Forestry Department of Zhejiang Province. Zhejiang Ecological Non-commercial Forests Report in 2010. (in Chinese). 2011.

27. Cameron AC, Trivedi PK. Regression-based tests for overdispersion in the poisson model. Journal of Econometrics. 1990;46:347–364.

28. Burnham KP, Anderson DR, Huyvaert KP. AIC model selection and multimodel inference in behavioral ecology: some background, observations, and comparisons. Behavioral Ecology and Sociobiology. 2011;65:23–35.

29. Condit R, Pitman N, Leigh JE, Chave J, Terborgh J, Foster RB, et al. Beta-Diversity in Tropical Forest Trees. Science. 2002;295:666–669. pmid:11809969

30. Barwell LJ, Isaac NJ, Kunin WE. Measuring β-diversity with species abundance data. Journal of Animal Ecology. 2015;84:1112–1122. pmid:25732937

31. Legendre P, Gallagher ED. Ecologically meaningful transformations for ordination of species data. Oecologia. 2001;129:271–280. pmid:28547606

32. Katovai E, Sirikolo M, Srinivasan U, Edwards W, Laurance W. Factors influencing tree diversity and compositional change across logged forests in the Solomon Islands. Forest Ecology and Management. 2016;372:53–63.

33. Oksanen J, Blanchet FG, Kindt R, Legendre P, Minchin PR, O’hara RB, et al. Package ‘vegan’. Community ecology package, version, 2(9). 2013.

34. Schuldt A, Baruffol M, Bohnke M, Bruelheide H, Hardtle W, Lang AC, et al. Tree diversity promotes insect herbivory in subtropical forests of south-east China. Journal of Ecology. 2010;98:917–926. pmid:20852667

35. Yu MJ, Hu G, Feeley KJ, Wu JG, Ding P. Richness and composition of plants and birds on land-bridge islands: effects of island attributes and differential responses of species groups. Journal of Biogeography. 2012;39:1124–1133.

36. Warren-Thomas E, Zou Y, Dong LJ, Yao XN, Yang MJ, Zhang XL, et al. Ground beetle assemblages in Beijing’s new mountain forests. Forest Ecology and Management. 2014;334:369–376.

37. Minden V, Scherber C, Cebrian Piqueras MA, Trinogga J, Trenkamp A, Mantilla-Contreras J, et al. Consistent drivers of plant biodiversity across managed ecosystems. Philosophical Transactions B. 2016. pmid:27114585

38. Kreft H, Jetz W. Global patterns and determinants of vascular plant diversity. Ecology. 2007;104:5925–5930.

39. Slik JW, Bernard CS, Van Beek M, Breman FC, Eichhorn KA. Tree diversity, composition, forest structure and aboveground biomass dynamics after single and repeated fire in a Bornean rain forest. Oecologia. 2008;158:579–588. pmid:18839212

40. Feng G, Mao LF, Benito BM, Swenson NG, Svenning JC. Historical anthropogenic footprints in the distribution of threatened plants in China. Biological Conservation. 2016.

41. Lin D, Lai J, Muller-Landau HC, Mi X, Ma K. Topographic variation in aboveground biomass in a subtropical evergreen broad-leaved forest in China. PLoS ONE. 2012. pmid:23118961

42. Wang DP, Ji SY, Chen FP, Xing FW, Peng SL. Diversity and relationship with succession of naturally regenerated southern subtropical forests in Shenzhen, China and its comparison with the zonal climax of Hong Kong. Forest Ecology and Management. 2006;222:384–390.

43. Yuan JF, Hu RY, Shen JH, Zhang L, Zhang XY, Yu MJ. Comparison of species composition and diversity of four successional forest communities in Zhejiang Province, East China. Bulletin of Botanical Research. 2011;31:61–66. (in Chinese, with English abstract).

44. Swinfield T, Afriandi R, Antoni F, Harrison RD. Accelerating tropical forest restoration through the selective removal of pioneer species. Forest Ecology and Management. 2016;381:209–216.

45. Diniz-Filho JA, Rangel TF, Hawkins BA. A test of multiple hypotheses for the species richness gradient of South American owls. Oecologia. 2004;140:633–638. pmid:15248060

46. Kaboli M, Guillaumet A, Prodon R. Avifaunal gradients in two arid zones of central Iran in relation to vegetation, climate, and topography. Journal of Biogeography. 2006;33:133–144.

47. Qian H, Wang SL, Li YL, Wang XH. Breeding bird diversity in relation to environmental gradients in China. Acta Oecologica. 2009;35:819–823.

48. De Cáceres M, Legendre P, Valencia R, Cao M, Chang LW, Chuyong GB, et al. The variation of tree beta diversity across a global network of forest plots. Global Ecology and Biogeography. 2012;21:1191–1202.

49. Wang QG, Xu YZ, Lu ZJ, Bao DC, Guo Yl, Lu JM, et al. Disentangling the effects of topography and space on the distributions of dominant species in a subtropical forest. Chinese Science Bulletin. 2014;59:5113–5122.

50. Liu JL, Qian H, Jin Y, Wu CP, Chen JH, Yu SQ, et al. Disentangling the drivers of taxonomic and phylogenetic beta diversities in disturbed and undisturbed subtropical forests. Scientific Reports. 2016. pmid:27775021

51. Baldeck CA, Harms KE, Yavitt JB, John R, Turner BL, Valencia R, et al. Soil resources and topography shape local tree community structure in tropical forests. Proceedings of The Royal Society B. 2013. pmid:23256196

52. Liu JJ, Tan YH, Slik JWF. Topography related habitat associations of tree species traits, composition and diversity in a Chinese tropical forest. Forest Ecology and Management. 2014;330:75–81.

53. Echeverría C, Newton AC, Lara A, Benayas JM, Coomes DA. Impacts of forest fragmentation on species composition and forest structure in the temperate landscape of southern Chile. Global Ecology and Biogeography. 2007;16:426–439.

54. Wang SP, Tang ZY, Qiao XJ, Shen ZH, Wang XP, Zheng CY, et al. The influence of species pools and local processes on the community structure: a test case with woody plant communities in China's mountains. Ecography. 2012;35:1168–1175.

55. Xu CG, Gertner GZ, Scheller RM. Importance of colonization and competition in forest landscape response to global climatic change. Climatic Change. 2012;110:53–83.

56. Liu YN, Tang ZY, Fang JY. Contribution of enviroinmental filtering and dispersal limitation to species turnover of temperate deciduous broad-leaved forests in China. Applied Vegetation Science. 2015;18:34–42.

57. Cramer MJ, Willig MR. Habitat heterogeneity, habitat associations, and rodent species diverstiy in a sand-shinnery-oak landscape. Journal of mammalogy. 2002;83:743–753.

58. Feeley KJ, Davies SJ, Perez R, Hubbell SP, Foster RB. Directional changes in the species composition of a tropical forest. Ecology. 2011;92:871–882. pmid:21661550

59. Chai SL, Tanner EV. 150-year legacy of land use on tree species composition in old-secondary forests of Jamaica. Journal of Ecology. 2011;99:113–121.

60. Matthews JW, Spyreas G. Convergence and divergence in plant community trajectories as a framework for monitoring wetland restoration progress. Journal of Applied Ecology. 2010;47:1128–1136.

61. Li MS, Mao LJ, Lu CL, Wang Y. Temporal change in fragmentation in China's primary forest ecoregions. Materials Science and Information Technology. 2012;433–440:1286–1296.

62. Goosem M, Paz C, Fensham R, Preece N, Goosem S, Laurance SG. Forest age and isolation affect the rate of recovery of plant species diversity and community composition in secondary rain forests in tropical Australia. Journal of Vegetation Science. 2016;27:504–514.

63. Laurance WF, Camargo JL, Luizão RC, Laurance SG, Pimm SL, Bruna EM, et al. The fate of Amazonian forest fragments: A 32-year investigation. Biological Conservation. 2011;144:56–67.

64. Martin PA, Newton AC, Bullock JM. Carbon pools recover more quickly than plant biodiversity in tropical secondary forests. Proceedings of The Royal Society B 2017. pmid:24197410

65. Lewis SL, Edwards DP, Galbraith D. Increasing human dominance of tropical forests. Science. 2015;349:827–832. pmid:26293955

66. Mo XX, Zhu H, Zhang YJ, Ferry Slik JW, Liu JX. Traditional forest management has limited impact on plant diversity and composition in a tropical seasonal rainforest in SW China. Biological Conservation. 2011;144:1832–1840.

67. Song C, Zhang Y. Forest cover in China from 1949 to 2006. Reforesting Landscapes: Lingking Pattern and Process. 2009;10.

68. Zhang Z, Swaisgood RR, Zhang S, Nordstrom LA, Wang H, Gu X, et al. Old-growth forest is what giant pandas really need. Biology Letters. 2011;7:403–406. pmid:21227979

69. Arroyo-Rodriguez V, Melo FP, Martinez-Ramos M, Bongers F, Chazdon RL, Meave JA, et al. Multiple successional pathways in human-modified tropical landscapes: new insights from forest succession, forest fragmentation and landscape ecology research. Biological Reviews. 2015. pmid:26537849

70. Zhou ZC, Xu GF, Jin GQ, He JP. Growth of broad-leaved species and community restoration of secondary masson pine forest after selective cutting. Forest Research. 2004;17:420–426 (in Chinese, with English abstract).