Next Article in Journal
Soil Microbial Community Responds to Elevation Gradient in an Arid Montane Ecosystem in Northwest China
Next Article in Special Issue
Vertical Variation in Leaf Traits and Crown Structure Promote the Coexistence of Forest Tree Species
Previous Article in Journal
Growing of the Containerized Seedlings of English Oak (Quercus robur L.) to Establish Sustainable Plantations in Forest-Steppe Ukraine
Previous Article in Special Issue
The Habitat Type and Scale Dependences of Interspecific Associations in a Subtropical Evergreen Broad-Leaved Forest
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Forests
Volume 13
Issue 9
10.3390/f13091358
forests-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Habitat Conditions and Tree Species Shape Liana Distribution in a Subtropical Forest

by
Buhang Li
1,
Yingming Zhang
2,
Fuchen Luan
2,
Zuoqiang Yuan
3,
Arshad Ali
4,
Chengjin Chu
1 and
Yue Bin
5,6,*
1
State Key Laboratory of Biocontrol and Department of Ecology, School of Life Sciences, Sun Yat-sen University, Guangzhou 510275, China
2
Guangdong Chebaling National Nature Reserve, Shaoguan 512500, China
3
School of Ecology and Environment, Northernwest Polytechnical University, Xi’an 710072, China
4
Forest Ecology Research Group, College of Life Sciences, Hebei University, Baoding 071002, China
5
Key Laboratory of Vegetation Restoration and Management of Degraded Ecosystems, South China Botanical Garden, Chinese Academy of Sciences, Guangzhou 510650, China
6
Southern Marine Science and Engineering Guangdong Laboratory (Guangzhou), Guangzhou 511458, China
*
Author to whom correspondence should be addressed.
Forests 2022, 13(9), 1358; https://doi.org/10.3390/f13091358
Submission received: 27 May 2022 / Revised: 29 July 2022 / Accepted: 9 August 2022 / Published: 26 August 2022
(This article belongs to the Special Issue Long-Term Monitoring of Forest Biodiversity and Dynamics in China)
Browse Figures
Review Reports Versions Notes

Abstract

:
Lianas are woody plants that require external support to reach the canopy. They are expanding in forests worldwide, possibly due to climate change and forest disturbance. Most studies on lianas have been conducted in tropical forests. Lianas are less explored in subtropical forests. We aimed to document the density and diversity of lianas, to test how habitat condition and the distribution of tree species affect the distributions of lianas based on data from a fully mapped 20 ha plot in subtropical China. We analyzed habitat association by fitting a generalized linear model with family-level liana abundance as response variable and family identity, and its interaction terms with topographic variables (slope, convexity, elevation, and sin(aspect)), as explanatory variables. We focused on the spatial associations of three liana species and 82 tree species with ≥100 individuals using the pair correlation function and redundancy analysis. We found a total of 1305 lianas, falling into 26 species, and 16 families, in the 20 ha plot. They accounted for 1.5% of individuals, 11.7% of species, and 0.4% of total basal area of woody plants in the plot. There were large variations in distributions of liana with respect to the four topographic variables among families, contrasting with former findings suggesting that lianas favor dry and hot habitats. The three most abundant liana species showed non-random associations with tree species, and they tended to positively associate with similar tree species but negatively associate with different tree species. The distribution of tree species explained 21.8% of variance in liana distribution. Our study suggested that both habitat conditions and tree composition intervene in determining liana distributions and that habitat heterogeneity may be a mechanism for liana diversity maintenance. Our study provides a basic understanding of liana diversity and distribution in this subtropical forest and contributes to future planning of liana studies and diversity conservation in subtropical forests under climate change.

1. Introduction

Lianas are woody plants that rely on other plants for support to reach the canopy. In recent decades, one of the most prominent changes in tropical forests is the expansion of lianas [1,2]. Prolonged dry seasons under climate change are considered to favor the growth and expansion of lianas, since lianas grow much faster than trees during the dry season [3]. Though lianas constitute part of species diversity [4,5,6] and provide shelters and food sources for animals in forests [7,8], the expansion of lianas, however, has detrimental effects on tree growth, survival, and reproduction, (reviewed by Estrada-Villegas and Schnitzer [9]) and can potentially alter forest composition, structure, and functions [8,9,10,11]. Given that the causes of climate change and the risks of liana expansion may probably last [1,2], an understanding of diversity, distribution, and species association of lianas in forests is essential so that proper conservation measures can be implemented when necessary.
Liana distributions in tropical forests are influenced by habitat conditions, but no consensus has been reached currently. In a Panamanian forest, lianas grew seven times faster than trees during the dry season [3], suggesting lianas have comparative advantage over trees in drier habitats, and may be more abundant under such habitat conditions in forests [3,10]. On the other hand, evidence showed that lianas were more abundant in concave sites with high moisture supply [12]. Though it was demonstrated that liana species generally lack ecophysiological structures for releasing potentially toxic compounds due to anaerobiosis [13], in a savanna–forest transition zone in central Brazil, flooded forests had high liana abundance dominated by Combretum laxum [14]. Similarly, in southern Amazonia, lianas were more abundant in habitats with impeded drainage [15]. These results suggest that some liana species may prefer high moisture supply. Further studies are needed to understand how liana distributions are affected by habitat conditions.
Besides habitat condition, tree species also affect the distribution of lianas [11,16,17,18]. Since lianas rely on tree species for support, liana species may show preference for certain tree species [18]. Though lianas do not establish species-specific associations with trees, some tree species are more likely to have lianas on them, such as White oaks in northern temperate forests in the USA [18], Liquidambar styraciflua (sweet gum) in floodplain forests in the southeastern United States [19,20], and Allophylus edulis and Cordia ecalyculata in a subtropical Atlantic forest [21]. As lianas are fast growing and maintained by disturbance [8], their distributions should tend to avoid shade-tolerant species. However, a field study found that shade-tolerant tree species held a high level of liana prevalence [11]. The first step to understand why some tree species attract more lianas is to document the tree–liana spatial associations.
In subtropical forests, lianas are less prominent in communities and hence have received much less attention than in tropical forests (however, see Cai and Song, 2000; Kusumoto et al., 2008) [5,12]. Under the regulation of a subtropical monsoon climate, precipitation in the Chinese subtropical zone shows strong seasonality, with a dry season lasting about half a year [22]. Winter is relatively mild, and with the lowest temperature around −5 °C, subtropical forests are also suitable habitats for lianas. Recent studies showed that the rainfall pattern in subtropical areas in China has shifted towards more heavy rains in the wet seasons, and more no-rain days and fewer light-rain days in the dry season since around 1980, and vegetation may be affected because soil moisture declined significantly from 1950 to 2009 [23]. Such change in precipitation pattern suggests the risk of increasing liana expansion for subtropical forests in China. However, in subtropical forests, lianas are often ignored during forest censuses. As a result, poorly known are the liana species diversity and what shapes liana distributions in subtropical forests.
In this study, we included lianas in our census conducted in a 20 ha plot in a subtropical forest in China. Based on these data, we aim to provide primary knowledge of species diversity of lianas, to test whether liana distribution is related to habitat conditions reflecting gradients of drought, and whether lianas are specifically associated with certain tree species in a subtropical forest. In doing so, we tried to answer the following questions: (1) How much do lianas contribute to forest diversity? (2) How do habitat conditions shape the distribution of lianas? (3) How do tree species influence the distribution of lianas? Our primary goal was to provide primary knowledge of liana diversity and distribution in this forest but the results of this study were useful for the conservation of species diversity in subtropical forests and could also set the basis for future studies on the temporal dynamics of lianas in the context of ongoing climate change.

2. Materials and Methods

2.1. Study Site and Forest Plot

This study was conducted in the Chebaling National Nature Reserve, which was established in 1981 and is located in the southeast part of Shixing, Shaoguan, Guangdong, China (114°09′04″–114°16′46″ E, 24°40′29″–24°46′21″ N). The altitude of this nature reserve ranges from 330 to 1256 m a.s.l. The climate regime belongs to the subtropical monsoon climate, with an annual average temperature of 19.6 °C and an annual precipitation of 2126 mm. Precipitation is relatively centered on a rainy season from April to October. The dry season lasts from November to March for the next year. The nature reserve is covered by subtropical evergreen broadleaved forests. Broadleaved, evergreen tree species from Fagaceae, Lauraceae, Thestarae, Hamamelidaceae, and Magno-Listarae dominate the forest communities in this reserve.
In September 2017, a 20 ha forest dynamic plot (the CBL plot) was established in the nature reserve. The plot was first divided into 500 quadrats with projected areas of 20 m × 20 m. The elevations at the corners of all 20 m × 20 m quadrats were recorded. Elevation, slope, convexity, and aspect for all quadrats were calculated based on the recorded elevations [24]). Every woody stem, including trees and lianas, with diameter at breast (DBH) height ≥1 cm was mapped, tagged, measured for DBH, and identified to species [25]. All rooted stems of lianas were measured and they were considered as individuals following Schnitzer et al. [26].

2.2. Data Analyses

We summarized information on abundance (i.e., the total number of individuals) and diversity at both the species and family levels for the lianas in the plot. These data for lianas were compared with the corresponding data for tree species in the plot.
For analyzing liana distribution with respect to topographic variables, we calculated elevation, slope, aspect, and convexity for each 20 m × 20 m quadrat, following Harms et al. [24]. We set zero aspect for a quadrat facing right towards the east. The value of sin(aspect) for a quadrat is associated with the amount of sunshine captured by it. A quadrat facing the south (π/2) received the largest amount of sunshine (sin (π/2) = 1), while a quadrat facing the north received the least (sin (3π/2) = −1). Thus, the sin(aspect) axis suggested increasing insolation and drier habitat condition. If lianas prefer drier habitat conditions in this forest, liana abundance would increase along the sin(aspect) axis. We expected that the light captured should be more informative for liana distribution than the angle itself. Therefore, we used sin(aspect) instead of the value of aspect in subsequent analysis.
We grouped data of lianas to the family level, and combined families with <30 individuals together to form a group (Others). We then fit a generalized linear model with the number of lianas of each group in each 20 m × 20 m quadrat as the response variable, while Family and its interactions with the linear and quadratic terms of elevation, slope, convexity, and sin(aspect) as explanatory variables, assuming a Poisson distribution for the response with a “log” link function [12]. The model was then subjected to a stepwise procedure to exclude non-significant variables. The significance of an interaction term of a topographic variable suggested significant variation among families in coefficients for that topographic variable.
We analyzed the distributions of liana species with ≥ 100 individuals with respect to tree species with ≥100 individuals using the bivariate pair correlation function, G12(r). A total of 3 liana species and 82 tree species were involved in this analysis. G12(r) is defined as the expected number of individuals of species 2 in the rings with radius r of an individual of species 1, divided by the density of species 2 in these rings [27]. Values greater than 1 for g(r) indicate positive associations, while smaller than 1 suggests negative associations. G(r) = 1 when the spatial association is random. We examined whether the deviation from one was significant by comparing the observed G12(r) value with the confidence envelope obtained from 99 random toroidal shifts. Both G12(r) and the confidence envelopes were estimated using the spatstat package on R 2.4.1 [28,29].
Only a few liana–tree associations were positive at some scales but negative at other scales. Excluding these liana–tree associations, we classified the remaining liana–tree associations into three types: “positive” for those positive at ≥one scale and random at other scales; “negative” for those negative at ≥one scale and random at other scales; “random” for those random throughout the range of scales studied. For simplicity, a tree species with positive and negative associations with a liana species was defined as the liana species’ attractor and detractor, respectively.
In order to obtain an integrative result for liana–tree association, we also analyzed their spatial distribution with redundancy analysis (RDA) using the vegan package implemented on the R platform [30]. As rare species often have an unduly large influence on the analysis, we included only the three liana species and tree species with ≥100 individuals in this analysis. We constructed two matrices, X and Y. Xij, Yij were the abundance of liana species j and tree species j in quadrat i, respectively. We first fit a null model with only the intercept term and a full model with all 82 tree species included. Then, variable selection was conducted using the step function based on Akaike Information Criterion (AIC, [31]. The significance of the final model was tested using 999 permutation tests.

3. Results

3.1. Diversity and Distribution of Lianas

There were a total of 1305 lianas, falling into 26 species and 16 families in the 20 ha CBL plot. They accounted for 1.5% of individuals, 11.7% of species, and 0.4% of the basal area in the plot. Among these stems, 387 (22 species) were ≥5 cm and 1249 (from 25 species) were ≥2 cm in diameter. Within the 20 m × 20 m quadrats, the number of liana stems ranged from 0 to 23 and the number of liana species ranged from 0 to 5. The ten most abundant quadrats collectively had 144 liana individuals from 12 species, corresponding to a density of 360 ha−1. Fissistigma oldhamii (Fisold) was the most abundant species, followed by Ampelopsis cantoniensis (Ampcan) and Callerya dielsiana (Caldie) (Table 1). Other species had fewer than 100 individuals in the plot. At the family level, Annonaceae had the largest number of lianas individuals but only Ampcan was in this family, followed by Fabaceae (257 individuals from 4 species) and then Vitaceae (183 individuals from 2 species) (Figure 1). The remaining families had ≤100 individuals and ≤2 species in the plot. The ranking based on basal area was the same as that based on number of individuals for the first three species and families.

3.2. Liana Distribution in Relation to Topographic Variables

The final model kept Family, and its interactions with the linear forms of all four topographic variables, and the quadratic forms of sin(aspect) and elevation (Table 2, Figure 2). The coefficients of slope for all families were positive. However, discernable increases within the ranges of slope in the plot were observed only for Annonaceae and Rubiaceae (Figure 2a). The coefficients of convexity were positive for all families except Annonaceae and Rubiaceae, indicating that most liana families had more individuals in convex habitats (Figure 2c). Though Annonaceae, Gnetaceae, and Others had positive coefficients for the quadratic term of sin(aspect), only Annonaceae exhibited a concave curve with an inflection point at sin(aspect) = 0.146 (Figure 2b).
Fabaceae, Primulaceae, Gnetaceae, and Others exhibited decreasing number of lianas within the main, if not the whole, range of sin(aspect) (Figure 2b), and all families except Annonaceae showed decreasing number of lianas when sin(aspect) > 0, suggesting that most liana families did not distribute more individuals in habitats facing the south (Figure 1b). Rubiaceae and Vitaceae had inflection points relatively close to zero (Rubiaceae, −0.047; Vitaceae, −0.478), forming convex curves with mild peaks (Figure 2b).
All families except Vitaceae and Others had negative coefficients for the quadratic term of elevation and their inflection points were within the range of elevation for the plot, showing convex curves for the expected number of lianas with respect to elevation (Figure 2d). In most cases, the curves were mild. The inflection point for Annonaceae was close to the lower limit, resulting in a decreasing trend for the main range of elevation (Figure 2d). Vitaceae and Others showed decreasing number of lianas with increasing elevation (Figure 2d).

3.3. Lianas’ Spatial Association with Tree Species

In a total of 246 liana–tree associations, only 20 were positive at some scales while negative at others (Figure 3a,b). Ten such associations were observed for Fisold, six for Ampcan, and four for Caldie. All three liana species showed more positive than negative associations with tree species (Caldie 53 vs. 22; Ampcan 38 vs. 19; Fisold 53 vs. 22).
Though in general, the majority (144/246, 58.5%) of liana–tree associations were positive, the G12(r) curves usually did not deviate from the upper confidence envelopes much (Figure 3). Among these associations, a considerable proportion (61/144) of G12(r) curves deviated from random toroidal shift at fewer than five scales examined (Figure 3c,e). This proportion was even larger for negative associations. Almost all (49/53, 92.5%) G12(r) curves for negative associations deviated from random toroidal shift at more than five scales examined. The four exceptions were Ilex dasyphylla, Elaeocarpus chinensis, Machilus grijsii with Fisold, and M. grijsii with Ampcan (Figure 3f as an example).
Most tree species (74/82) showed positive associations with at least one liana species. Among them, 64.9% (48) were positively associated with ≥2 liana species and 22 with all three liana species (Figure 4a). These 22 species made up 41.5% of the attractors of Caldie and Fisold, and 57.9% of those of Ampcan.
Only 37 tree species showed negative associations with liana species and 22 of them were negatively associated with only one species (Figure 4b). Compared with attractors, liana species tended to have different sets of tree detractors. An exception was that the majority (63.2%) of detractors of Ampcan were also detractors of Fisold, and vice versa (54.4%). Caldie shared only two detractors with Ampcan and three detractors with Fisold.
The final model of the RDA analysis retained only 20 tree species (Figure 5), explaining 21.8% of the total variance (total Inertia = 5.989, constrained Inertial = 1.302) in distributions of the three liana species. RDA axes 1 and 2 contributed to 68.3% and 19.1% of the variance explained by the model, respectively. RDA axis 3 explained the remaining 12.5%. The permutation test showed that the model was significant (p < 0.001).

4. Discussion

Our study is one of the first studies that examined liana diversity and distribution based on large-scale permanent forest dynamic plots in subtropical China. Liana density and diversity in this plot were substantially lower than those reported in tropical areas, such as Brazil, Malaysia, and Ecuador [14,32,33,34]. Though liana density and diversity varied greatly in tropical forests (reviewed by Schnitzer and Bongers [8]), they were all much greater than those observed in the CBL plot, the subtropical forest plot we studied.
Differences in sampling protocol were the most important factor for such larger differences in liana density and diversity observed in CBL versus other forests [12,14,21,32,33]. Similar to de Oliveira et al. [14] and Kusumoto et al. (2008)’s studies [12], our study was conducted in a single 20 ha plot, whereas some liana surveys were conducted by sampling separated quadrats [14,21,32,33,35]. The latter approach could cover much larger environmental gradients and therefore recorded higher species diversity for a given area. Additionally, as lianas were the focus of these studies, sites with high density might have been chosen so as to better document the diversity or effects of lianas in these areas [14,21,32,33]. Large-scale sampling covered continuous large areas, such as 20 ha in our study and 16 ha in Kusumoto et al. (2008)’s study [12], and included many quadrats without lianas. If sampled over the top ten most abundant quadrats, the density was 360 ha−1 in CBL, much higher than the density for the whole plot and much closer to the densities (about 1000 ha−1) for the Argentinian subtropical forest and other tropical forests [14,21,32,33]. As such, the difference in sampling methods was an important reason for the difference in recorded liana density and diversity between our and Kusumoto et al.’s studies (2008) [12] and others [14,21,32,33].
The difference in latitude or air temperature among these studies was another reason for the observed variation in liana density and diversity in our study and other subtropical forests. It is well known that liana diversity decreases with increasing latitude and increases with temperature at both local and global scales [3,6,36]. Variation in liana species richness was best explained by minimum temperature and frost events along a latitudinal gradient of temperate rainforest in South America [37]. Reduced liana density and diversity for colder zones may result mainly from embolisms in the wide vessel elements of lianas due to freezing [3,38,39]. Therefore, it is not surprising that tropical forests harbor much higher liana density and diversity than the CBL. Though Campanello et al. [21] and Kusumoto et al. (2008)’s [12] studies were also conducted in subtropical areas, the average temperatures for the coldest month for these two sites were about 18 °C, much higher than that of our study forest (10 °C) [12] (https://www.worlddata.info/america/argentina/climate-misiones.php, accessed on 20 April 2022; https://www.osgeo.cn/post/f31b7, accessed on 20 April 2022). Possibly owing to their warmer climate, liana diversity and density in these two forests were higher or similar to those of the studied forest plot, although our forest has a half-year annual dry season, which was thought to promote liana expansion [40].
Liana diversity and density decreased with increasing latitude and moisture supply in both tropical and temperate forests [3,8,41]. The mechanism driving these large-scale liana distributions might also operate at local scales. Accordingly, we expected lianas in this subtropical forest to exhibit preference for drier and hotter habitats. However, we found large variations among families in liana distribution with respect to topographic variables related to moisture supply and temperature. In our study, liana abundance increased with convexity for five of the seven families, consistent with the expectation of more lianas in drier habitats. However, the most abundant family, Annonaceae, exhibited the reversed trend, with more individuals in concave habitats than convex habitats. Similar to Annonaceae, total liana abundance was higher in concave sites than convex sites in a watershed in southwestern Japan [12]. In Central Brazil, liana density was remarkably higher in flooded forests which were also concave in topography [14]. Furthermore, decreasing liana abundances with sin(aspect) at the right half of the sin(aspect) axis were observed for all except one family (Annonaceae), in contradiction with the expectation that lianas prefer drier and hotter habitat conditions in this forest. Similarly, we expected quadrats at high elevations to be drier and have more lianas. However, five families of lianas showed convex while the remaining two groups (Vitaceae and Others) showed decreasing relationships with elevation, suggesting that most lianas were distributed in relatively low-to-medium elevations, which tend to have higher moisture supply. In general, our analysis of habitat preference revealed differentiated habitat conditions preferred by different liana families, consistent with previous fine-scale studies [34,42]. In a Bornean lowland tropical forest with alluvial, sandstone, and kerangas habitat types, 30 of 42 (71%) liana species showed significant habitat association with at least one habitat type, and liana species composition differed greatly among the three habitat types [34]. In the northern part of Okinawa Island, southern Japan, liana distributions in relation to topography also varied among species [42]. Taken together, at fine scales, topographic preference of lianas varied with liana taxonomic identity and habitat heterogeneity may help maintain liana diversity in local forest communities. What needs to be kept in mind when interpreting this result is that these topographic variables are not direct measures of temperature and moisture content or humidity, though topographic variables are among the most important factors influencing water availability [43,44] and solar radiation [45] at fine scales.
Previous studies suggested that the lianas and trees do not establish species-specific interactions [46,47,48]. Consistent with this, our study found that 22 tree species in this forest associated positively with all three most abundant liana species studied and made up large proportions of tree species positively associated with them. Though liana–tree associations were not species-specific, some tree species were more likely to be hosts of lianas [11]. For example, white oak in northern temperate forests in the USA [18], Liquidambar styraciflua (sweet gum) in floodplain forests in the southeastern United States [19,20], and Allophylus edulis and Cordia ecalyculata in a subtropical Atlantic forest [21]. Our point pattern analysis also found that tree species differed in the number of liana species positively associated with them. Furthermore, the liana species that affiliated to a tree species were not random, either. For example, Acer rubrum was more likely than other trees to have Vitis aestivalis (summer grape) present, and both Pinus banksiana and Quercus velutina (black oak) were more likely to have V. riparia (riverbank grape) present [18]. Our study also found that 1, 9, and 16 tree species showed positive associations only with Ampcan, Fisold, and Caldie, respectively. On the other hand, some trees, such as Acer saccharum, were less likely than expected to host a liana [18]. In our study, tree species tended to negatively associate with different liana species. Particularly, there was only one tree species negatively associated with all three liana species. Former studies found that trees with smooth bark, long leaves, low wood density, high trunk/branch flexibility, high level of self-pruning, and tall, branch-free bole-height were more able to resist liana colonization or to shed lianas, resulting in lower liana colonization [16,17,49,50,51]. Therefore, there might be a functional basis underlying the liana–tree associations. Whatever the mechanisms are, distribution of trees explained only 21.8% of variance in liana distribution in the RDA analysis, indicating relatively weak liana–tree associations in this subtropical forest. Our analysis provided preliminary knowledge on possible liana preference of tree species. Direct surveys of liana colonization are needed for those tree species showing significant associations with liana species to test whether they are more/less likely to be infested by lianas. An analysis based on direct surveys and related traits of trees and lianas is essential to reveal the mechanisms for liana–tree interactions [11].

5. Conclusions

Liana habitat preference varies with their taxonomic identity at local scales, and there were significant spatial association between liana species and tree species. Liana species tend to have similar tree attractors but different tree detractors. Our study indicates that both fine-scale habitat conditions and tree species composition intervene in determining the diversity and distribution of lianas in subtropical forests. Habitat heterogeneity in terms of both topography and tree species may contribute to liana diversity maintenance. This study provides a primary understanding of liana diversity and distribution in subtropical forests and contributes to future planning on liana studies and diversity conservation in subtropical forests in scenarios of liana proliferation under climate change.

Author Contributions

Conceptualization, B.L. and Y.B.; methodology, B.L. and Y.Z.; software, F.L.; validation, A.A., Z.Y. and C.C.; formal analysis, Y.B.; investigation, B.L., Y.Z. and F.L.; resources, B.L., Y.Z. and F.L.; data curation, B.L.; writing—original draft preparation, Y.B.; writing—review and editing, B.L., Y.B. and A.A.; visualization, Z.Y. and A.A.; supervision, Y.B.; project administration, B.L., Y.Z. and F.L.; funding acquisition, B.L., Y.Z. and F.L. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by General Data Management Project for Forest Plots in the Chebaling National Nature Reserve and Data Collection Project for Forest Resource Monitoring in Chebaling National Nature Reserve. The APC was funded by General Data Management Project for Forest Plots in the Chebaling National Nature Reserve.

Institutional Review Board Statement

Not applicable as this study does not involve humans and animals.

Informed Consent Statement

Not applicable as this study does not involve humans.

Data Availability Statement

Li, B; Zhang, Y.; Luan, F.; Yuan, Z.; Ali, A.; Chu, C.; Bin, Y.; Topographic Variables and coordinates of three liana and 82 tree species in Chebaling 20 ha plot. Science Data Bank, 2022 [2022-08-23]. http://doi.org/10.57760/sciencedb.02213.

Acknowledgments

We are very grateful to Chebaling National Nature Reserve for providing accommodation during the plot census. We also owe much to Huagu Ye and Yihua Tong for helping with the species identification in the plot. We thank Yongcai Zhao, Liafu Huang and many other field workers for joining the plot census in Chebaling Nature Reserve in 2017.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Schnitzer, S.A. Increasing liana abundance and biomass in neotropical forests: Causes and consequences. In Ecology of Lianas; Schnitzer, S.A., Bongers, F., Burnham, R.J., Putz, F.E., Eds.; Wiley-Blackwell Publishing: Oxford, UK, 2015; pp. 451–464. [Google Scholar]
  2. Schnitzer, S.A.; Bongers, F. Increasing liana abundance and biomass in tropical forests: Emerging patterns and putative mechanisms. Ecol. Lett. 2011, 14, 397–406. [Google Scholar] [CrossRef] [PubMed]
  3. Schnitzer, S.A. A mechanistic explanation for global patterns of liana abundance and distribution. Am. Nat. 2005, 166, 262–276. [Google Scholar] [CrossRef] [PubMed]
  4. Jongkind, C.C.H.; Hawthorne, W.D. A botanical synopsis of lianes and other forest climbers. In Forest Climbing Plants of West Africa: Diversity, Ecology and Management; Bongers, F., Parren, M.P.E., Traore, D., Eds.; CABI Publishing: Oxfordshire, UK, 2005; pp. 19–39. [Google Scholar]
  5. Cai, Y.L.; Song, Y.C. Diversity of vines in subtropical zone of East China. J Wuhan Bot. Res. 2000, 18, 390–396. [Google Scholar]
  6. Gentry, A.G. The distribution and evolution of climbing plants. In The Biology of Vines; Putz, F.E., Mooney, H.A., Eds.; Cambridge University Press: Cambridge, UK, 1991; pp. 73–97. [Google Scholar]
  7. Schnitzer, S.A.; Carson, W.P. Treefall gaps and maintenance of species diversity in a tropical forest. Ecology 2001, 82, 913–919. [Google Scholar] [CrossRef]
  8. Schnitzer, S.A.; Bongers, F. The ecology of lianas and their role in forests. Trends Ecol. Evol. 2002, 17, 223–230. [Google Scholar] [CrossRef]
  9. Estrada-Villegas, S.; Schnitzer, S.A. A comprehensive synthesis of liana removal experiments in tropical forests. Biotropica 2018, 50, 729–739. [Google Scholar] [CrossRef]
  10. Schnitzer, S.A. Testing ecological theory with lianas. New Phytol. 2018, 220, 366–380. [Google Scholar] [CrossRef]
  11. Visser, M.D.; Wright, S.J.; Muller-Landau, H.C.; Jongejans, E.; Comita, L.S.; de Kroon, H.; Schnitzer, S. Tree species vary widely in their tolerance for liana infestation: A case study of differential host response to generalist parasites. J. Ecol. 2018, 106, 781–794. [Google Scholar] [CrossRef]
  12. Kusumoto, B.; Enoki, T.; Watanabe, Y. Community structure and topographic distribution of lianas in a watershed on Okinawa, south-western Japan. J. Trop. Ecol. 2008, 24, 675–683. [Google Scholar] [CrossRef]
  13. Joly, C.A.; Crawford, R.M.M. Variation in Tolerance and Metabolic Responses to Flooding in some Tropical Trees. J. Exp. Bot. 1982, 33, 799–809. [Google Scholar] [CrossRef]
  14. de Oliveira, E.A.; Marimon, B.S.; Feldpausch, T.R.; Colli, G.R.; Marimon-Junior, B.H.; Lloyd, J.; Lenza, E.; Maracahipes, L.; Oliveira-Santos, C.; Phillips, O.L. Diversity, abundance and distribution of lianas of the Cerrado–Amazonian forest transition, Brazil. Plant Ecol. Divers. 2014, 7, 231–240. [Google Scholar] [CrossRef]
  15. Jirka, S.; McDonald, A.J.; Johnson, M.S.; Feldpausch, T.R.; Couto, E.G.; Riha, S.J. Relationships between soil hydrology and forest structure and composition in the southern Brazilian Amazon. J. Veg. Sci. 2007, 18, 183–194. [Google Scholar] [CrossRef]
  16. Putz, F.E. How trees avoid and shed lianas. Biotropica 1984, 16, 19–23. [Google Scholar] [CrossRef]
  17. van der Heijden, G.M.F.; Healey, J.R.; Phillips, O.L.; van der Heijden, G.M.F.; Healey, J.R.; Phillips, O.L. Infestation of trees by lianas in a tropical forest in Amazonian Peru. J. Veg. Sci. 2008, 19, 747–756. [Google Scholar] [CrossRef]
  18. Leicht-Young, S.; Pavlovic, N.B.; Frohnapple, K.J.; Grundel, R. Liana habitat and host preferences in northern temperate forests. For. Ecol. Manag. 2010, 260, 1467–1477. [Google Scholar] [CrossRef]
  19. Allen, B.P.; Pauley, E.F.; Sharitz, R.R. Hurricane impacts on liana populations in an old-growth southeastern bottomland forest. J. Torrey Bot. Soc. 1997, 124, 34–42. [Google Scholar] [CrossRef]
  20. Allen, B.P.; Sharitz, R.R.; Goebel, P.C. Twelve years post-hurricane liana dynamics in an old-growth southeastern floodplain forest. For. Ecol. Manag. 2005, 218, 259–269. [Google Scholar] [CrossRef]
  21. Campanello, P.I.; Garibaldi, J.F.; Gatti, M.G.; Goldstein, G. Lianas in a subtropical Atlantic Forest: Host preference and tree growth. For. Ecol. Manag. 2007, 242, 250–259. [Google Scholar] [CrossRef]
  22. Li, B.; Wu, Z.; Bin, Y.; Xiao, R.; Song, X.; Shu, Z.; Chu, C. Community Composition and Structure of the Chebaling Forest Dynamic Plot in a Mid-Subtropical Evergreen Broadleaved Forest; Guangdong Science & Technology Press: Guangzhou, China, 2020. [Google Scholar]
  23. Zhou, G.Y.; Wei, X.H.; Tan, Y.P.; Liu, S.G.; Huang, Y.H.; Yan, J.H.; Zhang, D.Q.; Che, Q.M.; Wang, J.X.; Meng, Z.; et al. Quantifying the hydrological responses to climate change in an intact forested small watershed in southern china. Global Change Biol. 2011, 17, 3736–3746. [Google Scholar] [CrossRef]
  24. Harms, K.E.; Condit, R.; Hubbell, S.P.; Foster, R.B. Habitat associations of trees and shrubs in a 50-ha neotropical forest plot. J. Ecol. 2001, 89, 947–959. [Google Scholar] [CrossRef]
  25. Wu, Z.; Wang, Z.; Luan, F.; Shu, Z.; Li, B. Community composition and floral characteristics of the Chebaling 20 hm2 forest dynamic plot in a mid-subtropical evergreen broad-leaved forest. For. Environ. Sci. 2021, 37, 86–91, (In Chinese with an English abstract). [Google Scholar]
  26. Schnitzer, S.A.; Rutishauser, S.; Aguilar, S. Supplemental protocol for liana censuses. For. Ecol. Manag. 2008, 255, 1044–1049. [Google Scholar] [CrossRef] [Green Version]
  27. Wiegand, T.; Moloney, K.A. Rings, circles, and null-models for point pattern analysis in ecology. Oikos 2004, 104, 209–229. [Google Scholar] [CrossRef]
  28. R Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2021; Available online: https://www.Rproject.org/ (accessed on 12 March 2021).
  29. Stoyan, D.; Stoyan, H. Fractals, Random Shapes, and Point Fields: Methods of Geometrical Statistics; Wiley: Chichester, UK, 1994. [Google Scholar]
  30. Legendre, P.; Legendre, L. Numerical Ecology, 3rd English ed.; Elsevier: Amsterdam, The Netherlands, 2012. [Google Scholar]
  31. Sakamoto, Y.; Ishiguro, M.; Kitagawa, G. Akaike Information Criterion Statistics; D. Reidel Publishing Company: Dordrecht, The Netherlands, 1986. [Google Scholar]
  32. Nabe-Nielsen, J. Diversity and distribution of lianas in a neotropical rain forest. J. Trop. Ecol. 2001, 17, 1–19. [Google Scholar] [CrossRef]
  33. Burnham, R.J. Dominance, diversity and distribution of lianas in Yasunı´, Ecuador: Who is on top? J. Trop. Ecol. 2002, 18, 845–864. [Google Scholar] [CrossRef]
  34. DeWalt, S.J.; Chave, J. Structure and biomass of four lowland Neotropical forests. Biotropica 2004, 36, 7–19. [Google Scholar] [CrossRef]
  35. Albuquerque, L.B.; Velázquez, A.; Mayorga-Saucedo, R. Solanaceae composition, pollination and seed dispersal syndromes in Mexican mountain cloud forest. Acta. Bot. Bras. 2006, 20, 599–613. [Google Scholar] [CrossRef]
  36. Hu, L.; Li, M.; Li, Z. Geographical and environmental gradients of lianas and vines in China. Glob. Ecol. Biogeogr. 2010, 19, 554–561. [Google Scholar] [CrossRef]
  37. Lobos-Catalán, P.; Jiménez-Castillo, M. Temperature shapes liana diversity pattern along a latitudinal gradient in southern temperate rainforest. Plant Ecol. 2019, 220, 1109–1117. [Google Scholar] [CrossRef]
  38. Sperry, J.S.; Hollbrook, N.M.; Zimmerman, M.H.; Tyree, M.T. Spring filling of xylem vessels in wild grapevine. Plant Physiol. 1987, 83, 414–417. [Google Scholar] [CrossRef] [PubMed]
  39. Ewers, F.W.; Fisher, J.B.; Fichtner, K. Water flux and xylem structure in vines. In The Biology of Vines; Putz, F.E., Mooney, H.A., Eds.; Cambridge University Press: Cambridge, UK, 1991. [Google Scholar]
  40. DeWalt, S.J.; Schnitzer, S.A.; Chave, J.; Bongers, F.; Burnham, R.J.; Cai, Z.Q.; Chuyong, G.; Clark, D.B.; Ewango, C.E.N.; Gerwing, J.J.; et al. Annual rainfall and seasonality predict pan-tropical patterns of liana density and basal area. Biotropica 2010, 42, 309–317. [Google Scholar] [CrossRef]
  41. Teramura, A.H.; Gold, W.G.; Forseth, I.N. The biology of vines. In Physiological Ecology of Mesic, Temperate Woody Vines; Putz, F.E., Mooney, H.A., Eds.; Cambridge University Press: Cambridge, UK, 1991; pp. 245–285. [Google Scholar]
  42. Kusumoto, B.; Enoki, T.; Kubota, Y. Determinant factors influencing the spatial distributions of subtropical lianas are correlated with components of functional trait spectra. Ecol. Res. 2013, 28, 9–19. [Google Scholar] [CrossRef] [Green Version]
  43. Brubaker, S.C.; Jones, A.J.; Lewis, D.T.; Frank, K. Soil properties associated with landscape positions and management. Soil Sci. Soc. Am. J. 1993, 57, 235–239. [Google Scholar] [CrossRef]
  44. Gessler, P.E.; Chadwick, O.A.; Charmran, F.; Althouse, F.; Holmes, K. Modeling soil-landscape and ecosystem properties using terrain attributes. Soil Sci. Soc. Am. J. 2000, 64, 2046–2056. [Google Scholar] [CrossRef]
  45. Hayes, M.; Moody, A.; White, P.S.; Costanza, J.L. The influence of logging and topography on the distribution of spruce-fir forests near their Southern limits in Great Smoky Mountains National Park, USA. Plant Ecol. 2007, 189, 59–70. [Google Scholar] [CrossRef]
  46. Hegarty, E.E. Vine-host interactions. In The Biology of Vines; Putz, F.E., Mooney, H.A., Eds.; Cambridge University: Cambridge, UK, 1991; pp. 357–375. [Google Scholar]
  47. Carrasco-Urra, F.; Gianoli, E. Abundance of climbing plants in a southern temperate rain forest: Host tree characteristics or light availability? J. Veg. Sci. 2009, 20, 1155–1162. [Google Scholar] [CrossRef]
  48. Garrido-Perez, E.I.; Burnham, R. The evolution of host specificity in liana-tree interactions. Puente Biologico 2010, 3, 145–157. [Google Scholar]
  49. Balfour, D.A.; Bond, W.J. Factors limiting climber distribution and abundance in a Southern African forest. J. Ecol. 1993, 81, 93–100. [Google Scholar] [CrossRef]
  50. Campbell, E.J.F.; Newbery, D.M. Ecological relationships between lianas and trees in lowland Rain Forest in Sbah, East Malaysia. J. Trop. Ecol. 1993, 9, 469–490. [Google Scholar] [CrossRef]
  51. Putz, F. Lianas vs. trees. Biotropica 1980, 12, 224–225. [Google Scholar] [CrossRef]
Forests 13 01358 g001 550
Figure 1. The distribution of liana individuals in a 20 ha subtropical forest dynamic plot in Chebaling (CBL), China. Color and numbers on the contour lines denote elevation. Capital letters show the locations of lianas and indicate their families.
Figure 1. The distribution of liana individuals in a 20 ha subtropical forest dynamic plot in Chebaling (CBL), China. Color and numbers on the contour lines denote elevation. Capital letters show the locations of lianas and indicate their families.
Forests 13 01358 g001
Forests 13 01358 g002 550
Figure 2. The expected number of lianas in relation to topographic variables under their observed ranges in the 20 ha subtropical forest dynamic plot in China. These curves were all obtained by the final generalized linear model fitted for lianas abundance in a 20 m × 20 m quadrat with a log link function against the family which the lianas belong to, and the interactions of family with slope, sin(aspect), convexity, and elevation of the quadrat. A Poisson distribution was set for the response variable.
Figure 2. The expected number of lianas in relation to topographic variables under their observed ranges in the 20 ha subtropical forest dynamic plot in China. These curves were all obtained by the final generalized linear model fitted for lianas abundance in a 20 m × 20 m quadrat with a log link function against the family which the lianas belong to, and the interactions of family with slope, sin(aspect), convexity, and elevation of the quadrat. A Poisson distribution was set for the response variable.
Forests 13 01358 g002
Forests 13 01358 g003 550
Figure 3. Examples of pair correlation functions for liana species and tree species: negative at some scales while positive at others (a,b), positive associations at <five scales (c) and at ≥five scales (d), negative associations at <five scales (e) and ≥five scales (f).
Figure 3. Examples of pair correlation functions for liana species and tree species: negative at some scales while positive at others (a,b), positive associations at <five scales (c) and at ≥five scales (d), negative associations at <five scales (e) and ≥five scales (f).
Forests 13 01358 g003
Forests 13 01358 g004 550
Figure 4. Venn diagram for the number of tree species associated with three liana species in a subtropical forest. Panel (a) shows the number for positive associations and panel (b) for negative associations.
Figure 4. Venn diagram for the number of tree species associated with three liana species in a subtropical forest. Panel (a) shows the number for positive associations and panel (b) for negative associations.
Forests 13 01358 g004
Forests 13 01358 g005 550
Figure 5. Redundancy analysis for the distribution of three liana species with respect to 20 tree species in a 20 ha subtropical forest dynamic plot in Chebaling (CBL), China. Tree species (explanatory variable), liana species, and sites are indicated by arrows, red crosses, and open circles, respectively. RDA axis 1 and 2 contributed to 68.3% and 19.1% of the variance explained by the model, respectively. Tree species: Ps: Phoebe sheareri; Dc: Daphniphyllum calycinum; Pb: Phoebe bournei; Ca: Choerospondias axillaris; Sc: Schoepfia chinensis; Tm: Tarenna mollissima; Lg: Lithocarpus glaber; Rn: Rapanea neriifolia; Do: Daphniphyllum oldhamii; Sst: Symplocos stellaris; Cp: Cinnamomum parthenoxylon; Av: Antidesma venosum; Pp: Photinia prunifolia; Il: Ilex lohfauensis; Ssu: Styrax suberifolius; Dm: Diospyros morrisiana; Bf: Beilschmiedia fordii; Cau: Cinnamomum austrosinense; Df: Diplospora fruticosa; Af: Alniphyllum fortunei.
Figure 5. Redundancy analysis for the distribution of three liana species with respect to 20 tree species in a 20 ha subtropical forest dynamic plot in Chebaling (CBL), China. Tree species (explanatory variable), liana species, and sites are indicated by arrows, red crosses, and open circles, respectively. RDA axis 1 and 2 contributed to 68.3% and 19.1% of the variance explained by the model, respectively. Tree species: Ps: Phoebe sheareri; Dc: Daphniphyllum calycinum; Pb: Phoebe bournei; Ca: Choerospondias axillaris; Sc: Schoepfia chinensis; Tm: Tarenna mollissima; Lg: Lithocarpus glaber; Rn: Rapanea neriifolia; Do: Daphniphyllum oldhamii; Sst: Symplocos stellaris; Cp: Cinnamomum parthenoxylon; Av: Antidesma venosum; Pp: Photinia prunifolia; Il: Ilex lohfauensis; Ssu: Styrax suberifolius; Dm: Diospyros morrisiana; Bf: Beilschmiedia fordii; Cau: Cinnamomum austrosinense; Df: Diplospora fruticosa; Af: Alniphyllum fortunei.
Forests 13 01358 g005
Table
Table 1. The liana species in the 20 ha Chebaling plot.
Table 1. The liana species in the 20 ha Chebaling plot.
Species CodeLatin BinomialFamilyAbundance
TodasiToddalia asiaticaRutaceae6
UncrhyUncaria rhynchophyllaRubiaceae16
MaccocMaclura cochinchinensisMoraceae6
FisoldFissistigma oldhamiiAnnonaceae610
AmpcanAmpelopsis cantoniensisVitaceae182
KadcocKadsura coccineaSchisandraceae1
EmbribEmbelia ribes var. pachyphyllaPrimulaceae6
StrcatStrychnos cathayensisLoganiaceae3
MorparMorinda parvifoliaRubiaceae17
MelfusMelodinus fusiformisApocynaceae11
ActcalActinidia callosalActinidiaceae12
ActlatActinidia latifoliaActinidiaceae1
MilnitMillettia nitidaFabaceae2
ParlaeParthenocissus laetevirensVitaceae1
ElaglaElaeagnus glabraElaeagnaceae1
RhanapRhamnus napalensisRhamnaceae9
VersolVernonia solanifoliaAsteraceae1
AketriAkebia trifoliataLardizabalaceae1
EmblaeEmbelia laetaPrimulaceae49
DalhanDalbergia hanceiFabaceae29
StaoboStauntonia obovatifoliolaLardizabalaceae18
MuspubMussaenda pubescensRubiaceae1
CaecriCaesalpinia cristaFabaceae68
EmbrudEmbelia rudisPrimulaceae32
CaldieCallerya dielsianaFabaceae158
GneparGnetum parvifoliumGnetaceae64
Table
Table 2. The contribution of each variable to the reduction in deviance in the final model for liana abundance as a function of Family and its interactions with the linear and quadratic terms of four topographic variables, which were slope, sin(aspect), convexity, and elevation using a generalized linear model assuming a Poisson distribution for the response variable. “Family” was a factor, with levels of Annonaceae, Fabaceae, Gnetaceae, Primulaceae, Rubiaceae, Vitaceae, and Others. The level of Others included all families of lianas with fewer than 30 individuals in the plot.
Table 2. The contribution of each variable to the reduction in deviance in the final model for liana abundance as a function of Family and its interactions with the linear and quadratic terms of four topographic variables, which were slope, sin(aspect), convexity, and elevation using a generalized linear model assuming a Poisson distribution for the response variable. “Family” was a factor, with levels of Annonaceae, Fabaceae, Gnetaceae, Primulaceae, Rubiaceae, Vitaceae, and Others. The level of Others included all families of lianas with fewer than 30 individuals in the plot.
Degree of FreedomDevianceResidue Degree of FreedomResidue Deviance
NULL 34995073.9
Family61082.134933991.8
Family: slope798.8934863892.9
Family: sin(aspect)7116.9634793775.9
Family: convexity7126.2434723649.7
Family: elevation7442.4534653207.2
Family: sin2(aspect)725.3734583181.9
Family: elevation2749.9834513131.9
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Li, B.; Zhang, Y.; Luan, F.; Yuan, Z.; Ali, A.; Chu, C.; Bin, Y. Habitat Conditions and Tree Species Shape Liana Distribution in a Subtropical Forest. Forests 2022, 13, 1358. https://doi.org/10.3390/f13091358

AMA Style

Li B, Zhang Y, Luan F, Yuan Z, Ali A, Chu C, Bin Y. Habitat Conditions and Tree Species Shape Liana Distribution in a Subtropical Forest. Forests. 2022; 13(9):1358. https://doi.org/10.3390/f13091358

Chicago/Turabian Style

Li, Buhang, Yingming Zhang, Fuchen Luan, Zuoqiang Yuan, Arshad Ali, Chengjin Chu, and Yue Bin. 2022. "Habitat Conditions and Tree Species Shape Liana Distribution in a Subtropical Forest" Forests 13, no. 9: 1358. https://doi.org/10.3390/f13091358

APA Style

Li, B., Zhang, Y., Luan, F., Yuan, Z., Ali, A., Chu, C., & Bin, Y. (2022). Habitat Conditions and Tree Species Shape Liana Distribution in a Subtropical Forest. Forests, 13(9), 1358. https://doi.org/10.3390/f13091358

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop