The influence of forest structure on the abundance, biomass, and composition of lianas in tropical forest fragments

doi:10.21203/rs.3.rs-3410725/v1

Download PDF

Research Article

The influence of forest structure on the abundance, biomass, and composition of lianas in tropical forest fragments

https://doi.org/10.21203/rs.3.rs-3410725/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Deforestation in the Amazon creates fragmented landscapes and increases the exposure of forest remnants to altered microclimates, leading to forest structural changes that can alter vegetation density and the forest's vertical profile. Trees are crucial to defining tropical forest structure, but lianas contribute as much as 25% of species and can intercept 10–20% of the total light in forest. While much is known about the effects of forest fragmentation on trees, much less is known about lianas. Our study aimed to understand how the liana structure and composition community respond to forest fragmentation, especially forest-structural changes and distance to forest edge, in Central Amazonia. We used data from 1,336 individuals (diameter-at-breast-height ≥ 2 cm) of 172 liana species recorded in 2 ha of forest. Then, we tested the relationship of abundance, biomass, number of species and species composition as a function of forest structural parameters obtained through Terrestrial LiDAR, a 3D-remote-sensing technique, and distance to forest edge. At sites with a lower density of canopy vegetation, liana abundance and species richness both increased. However, increases in liana biomass and changes in liana species composition occurred only near forest edges. We conclude that unanalyzed factors, such as microclimatic variation and intrinsic characteristics of lianas, may be affecting liana community composition. In the long term, the effect of reduced canopy density on liana abundance and climate change may cause further changes in liana species composition in forest fragments.

tropical forest

Amazonian rainforest

forest structure

forest fragmentation

community structure

In tropical forests, such as Amazonian rainforest, increasing deforestation has led to increasingly fragmented landscapes, often by dividing large forest fragments into smaller forest patches, and increasing the forest area exposed to edge effects (Haddad et al. 2015; Montibeller et al. 2020; Fischer et al. 2021). Forest edges tend to have higher intensity and turbulence of winds and exposure to solar radiation, which results in a desiccated drier and warmer environment (Ferreira and Laurance 1997; Briant et al. 2010). Microclimatic changes modify forest structure and dynamics due to increased mortality rates, mainly of large trees (Laurance et al. 2000), thereby frequently creating more gaps and natural regeneration (Fischer et al. 2021). Forest fragmentation leads to changes in the vegetation distribution along the vertical profile of forests. Forest edges tend to have lower structural complexity as compared with the forest interior, with a higher density of vegetation in the understory than in the upper canopy (Maeda et al. 2022) and a more accelerated leaf production and branch growth of understory trees (Nunes et al. 2022). Changes are seen in the tree growth patterns, in which understory species increase and canopy species decrease basal area (Albiero-Júnior et al. 2019, 2021), in addition to a decrease in mean canopy height and an increase in the abundance of pioneer species (Almeida et al. 2019).

Remote sensing can be used to understand the complexity of forest structure and the impact of natural and anthropogenic disturbances on forest structure (Fahey et al., 2019). LiDAR (Light Detection and Ranging), in particular, allows us to measure the 3D forest structural traits such as plant area density (PAD) and indices of plant vegetation (PAI), leaf area (LAI) and wood area (WAI). These traits are closely linked to ecological processes and help us to understand the distribution of vegetation with a three-dimensional perspective (Almeida et al. 2019; Calders et al. 2020; Maeda et al. 2022; Nunes et al. 2022; Blanchard et al. 2023). However, most forest structure studies focus on the spatial allocation of vegetation, without distinguishing plant types, and its influence on tree community (Blanchard et al. 2022).

Lianas are important structural components of tropical forests contributing ~ 25% to all woody species (Schnitzer and Bongers 2002; Campbell et al. 2014), and despite that, they have received less attention at these studies. Using LiDAR data from Barro Colorado Island, Schnitzer et al. (2021) quantified canopy disturbance and related it to increasing liana abundance, and Rodríguez-Ronderos at al. (2016) showed that lianas can comprise 10 to 20% of plant area index in tropical forest, i.e., how much of the total light is intercepted by lianas. Currently, there are studies developing techniques to extract the lianas from remote sensing data, but that is still in early stages (Sánchez-Azofeifa et al. 2017; Krishna et al. 2019; Han and Sánchez-Azofeifa 2022).

The liana growth habit is characterized by a stem with little structural supporting tissue (Rocha et al. 2022), requiring these woody plants to derive physical support from trees to access the forest upper canopy. The liana-tree interaction leads to competition for abiotic resources above- and below- ground, reducing the growth rate, recruitment, and survival of host trees and may impact the carbon stock of forests (Schnitzer and Bongers 2002; Van der Heijden et al. 2014; Schnitzer 2018; Estrada-Villegas and Schnitzer 2018; Meunier et al. 2022). Despite this, there is no evidence that there is a species-specific relationship between lianas and trees (Garrido-Perez and Burnham 2010; Li et al. 2022; Ofadu-Bamfo et al. 2022), but some trees are more susceptible to the presence of lianas (Laurance et al. 2001; Song et al. 2023), which could cause selective tree mortality.

Unfortunately, studies on liana ecology often lack the refinement of taxonomic identification of individuals, which prevents us from understanding how species respond ecologically to environmental disturbances, and how much they contribute to the structure, or even on the changes caused in the composition of the liana community (Burnham 2004). Considering taxonomic identification, Piovesan et al. (2022) found that congeneric liana species of the genus Machaerium have different density patterns in fragmented forest environments compared to what is commonly found in the liana community in Amazonia. A similar pattern was found in subtropical secondary forests of southwest China (Yuan et al. 2009) and in degraded tropical forests of Ghana (Ofosu-Bamfo et al. 2022). Considering these findings, we ask in this study: How do the structural forest characteristics, changed by forest fragmentation, and distance to the forest edge affect liana community structure and composition? Does distance to the forest edge and forest structure, have the same degree of influence on liana community?

Structural changes caused by forest fragmentation impact several aspects of the tree assemblage (Blanchard et al. 2023), and because lianas are important forest components, we hypothesize that lianas are also strongly affected by these changes. The “new” forest structure may define the spatial distribution, structure, and composition of the liana assemblage. Previous studies have found a positive relationship between liana proliferation and forest disturbance, with liana biomass inversely associated with tree biomass (Laurance et al. 2001; Campbell et al. 2014; Schnitzer et al. 2021). Hence with respect to liana structure, we predict an increasing of liana abundance and biomass in more degraded locals, associated with more thinner trees, open canopy, and higher proportion of vegetation at lower layers of forest, and with proximity of the forest edge.

Furthermore, it is reasonable to expect that not all liana species respond positively to the process of forest fragmentation and associated forest disturbances (Schnitzer 2018; Piovesan et al. 2022). Thus, to liana composition, we hypothesize that some species would respond positively or negatively to structural changes in the forest and distance to edge, whereas most species may show a neutral pattern, i.e., generalist species able to develop in all of environments. Therefore, we expect no homogeneous species response to modifications in the vertical structure of the forest (Piovesan et al. 2022). Finally, we hypothesize that forest structure has more influence on liana assemblage than distance to the edge, because forest structure mediates edge effects and is affected by other forest disturbances that may be independent of forest fragmentation.

Study area

The study was carried out in a permanent plot system of the Biological Dynamics of Forest Fragments Project (BDFFP – INPA) located in the Area of Relevant Ecological Interest BDFFP (ARIE – BDFFP), 80 km North of Manaus, Amazonas, Brazil. In addition to continuous primary forest control areas, the ARIE maintains experimental forest fragments of 1, 10, and 100 ha and has 69 1-ha permanent plots spread across all reserves (Lovejoy et al. 1986; Laurance et al. 2011, Fig. 1).

The permanent plots are subdivided into 25 subplots (20 x 20 m). For this study, we concentrated data collection and analysis on a smaller area comprising ten subplots in five permanent plots in the Colosso and Dimona sampling sites of ARIE BDFFP, which are ca. 25 km apart.

The matrix surrounding the forest fragments comprises old pastures and regenerating secondary forests. Maintenance of the isolation of the forest fragments is performed by clear-cutting a 100 m wide strip of secondary forest around each forest fragment (Lovejoy et al. 1986; Laurance et al. 2018).

The climate is classified as tropical humid (Af) (Koeppen, 1948), with a mean annual temperature of 26 ºC and annual rainfall of 2500–2800 mm. The rainy season extends from November to May/June, while the driest season is from July to October (monthly average less than 100 mm of rain) (Laurance et al. 2011). The soil is classified as Alic Yellow Latosol, the most common latosols in the Amazon basin, predominantly clayey, with high water retention capacity, acidic, low fertility, and high aluminum concentration (Fearnside and Leal-Filho, 2020).

The forest is terra-firme type (non-flooded) and under undisturbed conditions has a vertical structure of four strata: understory, sub-canopy, canopy, and emergent trees. The canopy reaches 25 to 35 m, with emergent trees up to more than 45 m (Albiero-Júnior et al. 2019). Tree composition (≥ 10 cm DBH) is highly diverse, with more than 280 spp. ha^− 1 (Laurance et al. 2010). Lianas also follow a pattern of high diversity, with an estimated 160 spp. ha^− 1 (Piovesan et al. 2022).

Database and sampling design

The questions addressed in this study are possible because we could overlay data, mainly because of advances in liana inventories in the permanent plots of the Biological Dynamics of Forest Fragments Project (PDBFF). That work put efforts into the taxonomic identification of individuals, and we combined it with three-dimensional and high-resolution data of the same forest structure obtained by ground-based LiDAR remote sensing (Terrestrial Laser Scanning - TLS).

To explore whether the forest structure affects lianas community structure, we combined and evaluated three databases: (1) inventory of lianas, (2) TLS forest structure survey, and (3) demography of 10–20 DBH trees. Data overlap was restricted to the TLS forest structure survey coverage.

(1) Inventory of lianas

In 1997 and 2014, 33,896 lianas with DBH ≥ 2 cm were located, measured, and tagged in the 69 permanent plots of the BDFFP, but only in 2016 the botanical material was collected and identified to the species level. By 2023, 43 plots of 1 ha were monitored, in which 21,500 individuals were classified into 40 families and at least 360 spp. with an average density of 500 (Robyn Burnham, unpublished data). The species list may increase by as much as 20% because the material collected is still in the process of taxonomic identification (Robyn Burnham, unpublished data). In 2017 and 2018, this collecting and identification inventory was performed in the forest fragments plots of Colosso, and in 2019, the work continued to Dimona.

The diameter measurement protocol was adapted from other protocols described by Gerwing et al. (2006) and Schnitzer et al. (2008). From the data, we estimated the above-ground biomass (AGB) using an allometric equation exclusive for lianas, proposed by Schnitzer et al. (2006): \(\text{A}\text{G}\text{B} =\text{e}\text{x}\text{p}[-1.484+2.657 \times \text{l}\text{n}(\text{D}\text{A}\text{P}\left)\right]\). This equation uses only breast height (DBH); the AGB is expressed in kg of dry mass.

(2) Forest structure survey

In 2019, forest structure data were acquired by Terrestrial Laser Scanning (TLS), using a RIEGL VZ-400i instrument in an area of 100 x 32 m in 5 permanent plots evaluated in this study. The point clouds were processed and registered using RiSCAN PRO software version 2.9 (see Supplementary Material).

After that, voxelization was performed with the AMPVox software, in which the total volume of the point cloud of each area was divided into voxel (regular grids in three-dimensional space) of 1 m³, and the values of plant area density (PAD, m² m^− 3) were estimated for each m³. From the PAD the other traits of the forest structure were calculated.

The values used for vertical structure traits were obtained from averages for approximately 20 x 16 m areas. First, the plant area index (PAI, m² m^− 2) was estimated for each vertical column (X, Y coordinates), which is a combination of the leaf and woody area indexes, such as branches and trunks, and is obtained by the sum of the PAD values in each voxel column. PAI regulates the capacity of the forest to capture light and exchange gases and is, therefore, related to forest productivity. High values may indicate little light penetration under the canopy (Reich 2012; Ma et al. 2021). The PAI can be stratified, so we derived the PAI for a range from 15 to 30 m height and considered it as the canopy; for the range from 0 to 15 m height, we considered it to be understory, thus simplifying the forest complexity into two strata. We then averaged the PAI values for each stratum (Nunes et al. 2022).

Furthermore, we also calculated the canopy ratio (CR) using the equation: CR = (RH98 - RH25) ÷ RH98. RH25 and RH98 represent the relative heights (in m) at which 25% and 98% of the vegetation is in the voxel column (more details in Maeda et al. 2022). The CR reveals a proportional relationship of plant material distribution in the vertical gradient of the forest. Values smaller than 0.74 indicate a higher allocation of plant material in the upper strata. In comparison, values larger than 0.74 indicate that the allocation of material is more strongly localized in the lower strata (Maeda et al. 2022), possibly indicating clearings along the forest profile.

Lianescent growth is characterized by a liana stem being rooted in one location, yet its crown may be meters away. Thus, we added to the database all lianas present in an additional 4 m beyond the edge in each area covered by the TLS survey. In this case, each sampling area covered 100 x 40 m (Fig. 1d). The habit of the lianas can be apparently erratic or stimulated by light entry paths through the vertical structure of the forest, which are highly variable in time and space. Therefore, we extended our analyses to a subplot with twice the area of the TLS survey, by joining four subplots, creating subplots of 40 x 40 m. Only lianas whose rooting point was within the subplots were considered. Each stem that is marked with a unique identification tag was considered an independent individual.

(3) Tree demography: Between 2015 and 2016, the tree individuals were collected, measured, and identified to the species level. In the study sub-plots, there were 1,108 trees, of which 60.7% were individuals with DBH between 10 and 20 cm (Supplementary Table 1). The density of thin trees indicates the capacity for physical support: the higher the density, the greater the potential for lianas to climb into the forest canopy and thus obtain better light conditions for development and reproduction.

Statistical analysis

In addition to the independent variables referring to forest structural parameters: (1) total PAI; (2) understory PAI (0 to 15 m); (3) canopy PAI (15 to 30 m); (4) canopy ratio (CR); (5) density of thin trees (with DBH between 10 and 20 cm). We also considered (6) distance to the edge (m). We performed analyses in R statistical software (v 4.2.1; R Core Team 2022) and the results were considered significant when P-values were ≤ 0.05.

There was a strong correlation of distance to edge with total PAI and understory PAI (indicated by Pearson correlation, Supplementary Fig. 1), as well as a moderate but significant (GLM; t = -4.09; p < 0.001; r² 27.8%) correlation between canopy PAI and understory PAI (-0.52). In the second step, multicollinearity among the independent variables was tested with the performance package (Lüdecke et al. 2021), and we only kept the variables whose VIF ≤ 2 (Supplementary Table 1), thus the variables total PAI and understory PAI were removed from the models.

We selected regression models based on the lowest Akaike Information Criterion (AIC) values and, when ΔAIC < 2, selected the model with the fewest degree of freedom. All models were fitted with the DHARMa package (Hartig 2022), and R² values were obtained by the r2 function from the performance package. For the selected multiple regression models, we standardized the variables on a standard deviation scale to compare the effect size of the independent variables on the statistical model.

We tested the relationship of liana density and biomass with forest structure parameters and distance to the edge. For liana abundance, we used the Generalized Linear Mixed Model (GLMM) with the glmmTMB package (Books et al. 2017) and the Negative Binomial 2 distribution (log link). We also used GLMM with Poisson distribution (log link) for alpha diversity. In both models, only forest fragment size was included as a random factor.

For liana biomass, we used the Generalized Linear Model (GLM) because the conditional R² in the GLMM models were very close (difference of less than 1%) to marginal, so there was no variation related to any of the spatial autocorrelation variables tested (site, reserve size, and permanent plots). The distribution used was Gamma (log link). We also excluded the biomass of one individual that it was an outlier (Abuta velutina - DBH 44.7 cm). With this individual in the subplot, the regression model had poor quality and it did not fit statistically, presenting significant dispersion (dispersion = 2.70; p = 0.01). Without it, the model was fitted and had a non-significant dispersion (dispersion = 1.49; p = 0.16).

To evaluate liana composition, we estimated the variation in diversity β among subplots using multivariate indirect ordination of Principal Coordinates Analysis (PCoA; Legendre and Legendre 1998), with the Bray-Curtis dissimilarity matrix (relative abundance per subplot). The first two axes were tested as a function of forest structure parameters and distance to edge, using GLMM with Gaussian distribution (identity link). To understand how the species were related to the significant independent variable of the model, we performed a direct ordination with the “poncho” function (Dambros 2020). The functions used were “deconstand” e “vegdist” from the vegan package (Oksanen et al. 2022) and “cmdscale” from R's native statistical functions.

Floristic and phytosociological aspects of lianas

We characterized the floristic and phytosociology, aspects of liana community structure and composition, from 1,336 individuals with an average density of 668 ind. ha^− 1 (485–872.5 ind. ha^− 1), of which 82% were identified to species and 1.3% (18 ind.) not identified to family. We recorded 123 spp. and 49 morphotypes representing 69 genera and 26 plant families. Our study encompasses ca. 50% of the liana diversity so far encountered at the BDFFP (see the species list in Supplementary Table 7).

Bignoniaceae, with 32 spp. and 408 ind. was the most diverse and abundant family, followed by Fabaceae (29 spp. and 363 ind.), Celastraceae (15 spp. and 80 ind.), Connaraceae (5 spp. and 73 ind.), Dilleniaceae (5 spp. and 39 ind.), and Apocynaceae (12 spp. 38 ind.). The most diverse genus was Adenocalymma (Bignoniaceae, 12 spp. and 157 ind.), followed by Strychnos (Loganiaceae, 10 spp. and 20 ind.), Machaerium (Fabaceae, 9 spp. and 112 ind.), Moutabea (Polygalaceae, 8 morphotypes and 27 ind.) and Coccoloba (Polygonaceae 8 morphotypes and 14 ind.).

The rank order of genus changes when we consider the highest frequency of occurrence. Adenocalymma has the highest frequency (90%), followed by Machaerium, Deguelia, Bignonia, Rourea and Fridericia, with Bignonia comprising only Bignonia aequinoctialis. Each one of these six genera was present in more than 52% of the sub-plots, with all of them has absolute density above 35.5 \({\text{i}\text{n}\text{d}. \text{h}\text{a}}^{-1}\) (Supplementary Table 8).

The highest absolute density was Bignonia aequinoctialis (48.5 \({\text{i}\text{n}\text{d}. \text{h}\text{a}}^{-1}\)), representing 8.1% of all the individuals identified. It is followed by Deguelia negrensis (44.5 \({\text{i}\text{n}\text{d}. \text{h}\text{a}}^{-1}\)), Rourea paraensis (26 \({\text{i}\text{n}\text{d}. \text{h}\text{a}}^{-1}\)), Fridericia prancei (22.5 \({\text{i}\text{n}\text{d}. \text{h}\text{a}}^{-1}\)) and Machaerium hoehneanum (18.5 \({\text{i}\text{n}\text{d}. \text{h}\text{a}}^{-1}\)). These five species contribute around 25% of all individuals, and 17 species collectively contribute 50% (Supplementary Table 9). The first three species with the highest density also have the highest absolute frequency, present in at least half of the subplots. About 79.6% species are considered rare, as they occur less than 5 \({\text{i}\text{n}\text{d}. \text{h}\text{a}}^{-1}\) and are naturally infrequent (Supplementary Table 9), as they are present in less than 10% of the subplots. They follow the same pattern of the hectares from which they are drawn, thus we conclude that our subplots are representative.

Forest structure and liana community

The liana community structure, represented by absolute abundance, was only related to canopy PAI (Z = -3.39; p < 0.001) and distance to the edge (Z = -2.20; p = 0.028), as indicated by the results of the most parsimonious model (model 1 in Supplementary Table 3). Thus, we observed a greater abundance of lianas in sites with low canopy PAI values, and as there was a structural increase or greater complexity in the canopy structure, the abundance of lianas decreased (Fig. 2a). The same negative relationship occurs with distance to the edge: the closer to the edge, the greater the abundance of lianas (Fig. 2b). However, standardizing the fixed variables in the model, the strength of the effect of canopy PAI (b = -0.40) was 1.6 times greater than that of distance from the edge (b = -0.25). However, the fixed variables explained only 22.5% of the abundance of lianas in the forests studied, while fragment size was responsible for explaining 25.2%.

On the other hand, the absolute biomass of lianas was only related to distance from the edge and not to the forest structure parameters (model 1 in Supplementary Table 4). The closer to the edge, the higher the absolute biomass of lianas (Fig. 2d; t = -2.193; p = 0.033; r² = 12.9%).

Due to the lianescent growth habit, which can appear to be erratic, we decided to analyze, even if we lost degrees of freedom, whether our results would change if we increased the sample area by doubling the size of the 40 x 40 m subplots. However, the most parsimonious model for the density and biomass of lianas was the null model, so by increasing the size of the subplot, the structural parameters and distance from the edge were no longer predictors that explained the variation in the abundance and biomass of lianas (Supplementary Table 5).

Forest structure and taxonomic composition of the liana community

Following our assessment of the influence of forest structure on liana community structure, we also investigated if the taxonomic composition of the community could be associated with or affected by forest structural parameters and distance to the forest edge.

For β-diversity, PCoA based on the presence and abundance of all liana species captured 20.8% of the variation in species composition. The subplots in Dimona were more clustered, indicating greater similarity in composition. In contrast, in Colosso, the subplots were dispersed, indicating a species composition difference within this reserve.

The most parsimonious model (model 2 in Supplementary Table 6) showed that PCoA 1 is strongly related to the sampling sites (Fig. 2f; Z = -6.332; p < 0.001; Supplementary Information). Therefore, the species composition between the sites is significantly different. It also showed a significant relationship with distance to the edge (Fig. 2c; Z = -2.56; p = 0.01). PCoA 2 showed no significant relationship with the forest's structural parameters.

For α diversity, represented by the diversity of species in the subplots, the most parsimonious model was the one related only to the canopy PAI (Fig. 2e; Z = -2.47; P = 0.014), showing a negative relationship, similar to results on the abundance of lianas.

We directly ordered the species composition to better understand the relationship between species and distance to the edge. However, this time, we only included species with absolute density ≥ 5 \({\text{i}\text{n}\text{d}. \text{h}\text{a}}^{-1}\)(35 spp.), as the distribution and rarity of the other species would make any plausible interpretation impossible. The result demonstrated that, based on distance to the edge, some species are more common in forest interiors or in forest edges within the forest fragment. While the most abundant species, Bignonia aequinoctialis, Deguelia negrensis, and Rourea paraensis, were distributed across the entire gradient of distance to the edge, their highest relative abundance is found in the interior of the forest fragments (Fig. 3).

In this study, we investigated the effects of forest structure on the liana community and the effect of distance to the forest edge in a fragmented tropical forest. We found that changes in forest structure arising from forest fragmentation, particularly canopy PAI, strongly influenced the species composition of lianas. High PAI values could indicate lower light under the canopy (Ma et al. 2021), higher vegetation indices in the understory (Maeda et al. 2022; Supplementary Fig. 1), and more significant structural complexity, so the restricted occurrence of some species in these sites could indicate a lower tolerance to shading (Lomwong et al. 2023).

However, distance from the edge had a direct or indirect influence, mediated by canopy PAI, on the liana community's abundance, biomass, and number of species (α diversity). In contrast, distance from the edge only influenced composition (diversity β).

High diversity and site effect

Even considering the spatial restriction of the TLS coverage, which provided detailed data on the BDFFP fragment’s forest structure, we included a significant fraction of the local diversity. In this study, we evaluated 123 spp. and 49 morphotypes, corresponding to half of the species and 65% of the families recorded throughout the BDFFP, even though the area sampled represents only 4.6% of the entire area inventoried by the Liana project. When we compare the number of species found in our study area, which attained a larger area than Laurance et al. (2001), the number of species was almost double (47.7%) for 2.88 ha in the BDFFP. And for a region 80 km away, at the Ducke Reserve, our study comprises a third (33.1%) of the total lianas species inventoried there (Rocha et al. 2022).

We found a strong site effect during exploring and analyzing the species composition data (Supplementary Information). This effect has already been described for trees, palms, and bats in the BDFFP (Laurance et al. 2017). Before forest fragmentation, the sites located to the East, West, and Central areas of the Conservation Unit already showed differences in species composition and dynamics, and the successional trajectory after forest fragmentation made these parameters even more distinct. This process has been attributed to the different matrices of each site (Laurance et al. 2017). We propose two other possible explanations to understand the divergence in liana composition based on the distribution and dispersal of species by drainage basins and by different topographic characteristics.

The sampling sites are in different hydrographic basins. The drainage of Colosso is via the rivers Preto da Eva and Urubu (Uatumã basin) while for Dimona drainage is to the Cuieiras river (Rio Negro basin) (Amazonas 2019; Costa et al. 2023). In the Ducke Reserve (80 km away from the location of this study), there is also a difference in composition among various taxonomic groups, mainly trees, shrubs, lianas, birds, frogs, and fish. This difference was explained by the distributions of these organisms in the different drainage basins, with rare species or those with a restricted distribution occurring in only one or another drainage basin (Costa et al. 2008), roughly the same basins as the BDFFP.

Another possible explanation raised, but not tested, is the difference in altitude of the terrains, which may consequently be related to a hydrological, soil texture and nutrient gradient (Schietti et al. 2014), with Dimona at lower elevation and Colosso at a higher elevation. For the BDFFP areas, Laurance et al. (2001) found that the abundance of lianas varied with soil edaphic factors, such as fertility, texture, and slope. In the Ducke Reserve, terrain height above nearest drainage (HAND) is a compositional descriptor for two of the most representative liana families: Bignoniaceae and Fabaceae (Schietti et al. 2014; Gerolamo et al. 2022; Rocha et al. 2022).

In the studies mentioned above, Bignonia aequinoctialis (Bignoniaceae) was found at shorter distances from the water table. In our study, around 91% of the B. aequinoctialis occurred in the Dimona reserve. The species Adenocalymma validum (Bignoniaceae) is only present at Colosso and is found at larger distances from HAND. This is also true for Senegalia altiscandens (Fabaceae), which has a high abundance in plots at Colosso as compared to all 43 ha of the BDFFP (Robyn Burnham, unpublished data).

Canopy structure and distance to forest edge affect liana community

Forest edge creation is one of the crucial factors in understanding the structure and composition of the liana assemblage in fragmented tropical forests, with direct and indirect relationships mediated by forest structure altered by forest fragmentation. Certainly, edge creation is one of the transformative factors driving forest structural changes (Maeda et al. 2022), leading to a collapse in tree biomass (Laurance et al. 2017), as well as changes in microclimate patterns found in forest fragments (Camargo and Kapos 1995; Nunes et al. 2022). There is evidence that these transformations are persistent and can be detected even if the process of forest fragmentation occurred decades ago, which is the case in our study area (Almeida et al. 2019; Maeda et al. 2022).

Our study reinforces the negative relationship between the absolute abundance of lianas and distance from the edge, which has been repeatedly recorded (Laurance et al. 2001; Magrach et al. 2014; Campbell et al. 2018; Ofosu-Bamfo et al. 2022; Piovesan et al. 2022). However, we also found that liana abundance is slightly more related to canopy structure than to distance to edge.

Laurance et al. (2001), despite reinforcing the relationship with distance from the edge, already suggested that the relationship was not as strong as they expected, with tree biomass and the rate of change of tree biomass being more critical factors in that context, and these factors are related to the structure of the forest canopy. Considering the changes in forest structure caused by forest fragmentation and edge creation, distance to edge is not in and of itself what influences the structure of the liana assemblage: they are also influenced by forest structure (Blanchard et al. 2023).

Forest disturbances, such as the opening of clearings and consequent increased light penetration, can enhance the regrowth capacity of some species (Ledo and Schnitzer 2014; Rocha et al. 2020). In synergy with the process of forest fragmentation (Almeida et al. 2019), this could explain the greater abundance of lianas in these regions with less complex canopies, i.e., more open, and less dense canopies. Lianas would act as gap maintainers and niche constructors, seeing as in these regions, there is a higher abundance of lianas and a higher number of species (Schnitzer et al. 2021).

The success in increasing the abundance (Laurance et al. 2014) and density of stems (Rocha et al. 2020), while potentially spreading intensely and shading all or part of the tree canopy, reduces the growth rate of trees, keeping the canopy low. In this way, the increased abundance of lianas is influenced by, and influences, forest structure and tree species composition (Schnitzer et al. 2021).

Despite the influence of canopy PAI on the abundance of lianas, when we include taxonomic identity in the individuals (β diversity), this relationship disappears, and only the negative relationship with distance to edge remains. Thus, the direct effect of distance to edge may be mediated by other processes not analyzed in this study, such as the microclimatic variations that exist in the gradient of distance related to the edge (Camargo and Kapos 1995).

In addition to being one of the factors responsible for changes in forest structure after the death of large trees, the edge effect is also responsible for a decrease in humidity and an increase in temperature in forest fragments because of physical damage caused by strong winds and desiccation (Camargo and Kapos 1995). The edge effect can also be exacerbated by the effect of seasonality on the leaf phenology of canopy trees, reducing canopy vegetation and allowing higher light penetration into the lower strata, creating an even hotter environment (Nunes et al. 2022).

Functional traits related to the ability to tolerate microclimatic variations, such as variations in soil moisture and temperature, are essential factors in tree distribution, leading to a shift in forest dynamics (Engelbrecht et al. 2007). As we find different compositions of liana species in the gradient of distance to the edge, there may be functional groups whose characteristics generate different responses to microclimatic variations, modulating the occurrence of species. Understanding these responses seems important and alerts us to future approaches, especially related to climate change (Vogado et al. 2022), increased seasonality due to extreme events such as El Niño, and increased forest fragmentation.

Finally, comparing the results obtained for alpha and beta diversity with another study carried out in the same area, we found opposite patterns, in which 22 years ago, there were a more significant number of species at the edge, but the composition did not change in the gradient of distance to edge (Laurance et al. 2011). However, it is essential to note that the context of forest fragmentation at that time differs from what we find now. Although persistent, the effects of the forest fragmentation process have been attenuated over the years (Almeida et al. 2019), even with the periodic cutting of the matrix vegetation to maintain the isolation of the forest fragments.

In this way, it is possible that more than four decades after the beginning of the forest fragmentation process, the fragments have reached a more dynamic equilibrium, with the age of the edge being an essential factor in the dynamics of liana community (Laurance et al. 2001).

Biomass and support capacity for lianas

This work sheds light on a relevant factor in fragmented forests, the altered forest structure, to disentangle the indirect effects that influenced the structure and composition of the liana assemblage. The influence of the factors analyzed was low for all analyses performed, so other environmental factors may have a more substantial influence on the liana assemblage.

For liana biomass, studies are showing that it is related to edaphic characteristics, such as soil fertility and texture, and in the second instance, to the rate of change in above-ground tree biomass (Laurance et al. 2001), which may explain the relationship we found with distance to the forest edge. For the diversity and distribution of liana species, there is a strong relationship with the abiotic environmental characteristics, such as topographic gradient, water availability, and solar radiation, both in tropical forests (Schietti et al. 2014; Gerolamo et al. 2022; Rocha et al. 2022) and subtropical forests (Li et al. 2022).

We expected that the higher availability of thin trees would increase the physical support capacity for lianas to grow and develop, thus increasing the assemblage's abundance and biomass. However, we were unable to validate this pattern. Van der Heijden and Phillips (2008) found a positive but weak relationship between the abundance of lianas and the density of thin trees (DBH ≥ 10 cm), and the explanation for this finding was that, at a geographical scale, this predictor did not seem important, but perhaps locally it could be. In our study, we did not find this importance even locally.

Based on the literature, we concluded that the influence of the availability of physical support on the abundance and distribution of lianas tends to depend on a topographic gradient between lowlands and plateaus because of the functional traits of the stem (Rocha et al. 2022) and the life stage of the individual (Nogueira et al. 2011). In addition, the climbing guild is an essential parameter to analyze because there may be a significant relationship with tree size for the population of some species whose climbing strategy is based on tendrils and hooks (Laurance et al. 2001; Yang et al. 2018).

Lianas are a highly heterogeneous group of organisms that respond differently to forest fragmentation (Piovesan et al. 2022), have distinct functional and hydraulic traits (Meunier et al. 2021; Coppieters et al. 2022; Rocha et al. 2022) and growth strategies (Cai et al. 2007), so detecting community-level relationships is a significant challenge. Studies based on lower taxonomic levels, such as genus and species, combined with exploring functional traits, are the most promising path towards a better understanding of the other mechanisms involved in increasing the density and distribution of liana species.

In the context of fragmented forests in Central Amazonia, we tested whether forest structure affects the structure and composition of the liana assemblage and whether forest structure and distance to the edge have the same effect on lianas. We found that canopy plant area index was related to species richness and absolute abundance of lianas, with this relationship being slightly stronger than the distance to edge. Species composition was only influenced by distance to the edge, suggesting that other processes related to the creation of the edge, such as microclimatic variation, may influence the distribution of liana species.

We conclude that forest structural parameters related to light penetration impact the absolute abundance of lianas but not their composition. Furthermore, the age of the forest fragment edge is essential for analyzing aspects of the liana’s community. As the forest fragment ages, it may reach a more dynamic equilibrium. Therefore, the patterns of composition and richness may shift when compared to younger forest edges and fragments. Meanwhile, liana abundance seems to increase, not only due to distance to the edge but also due to canopy vegetation affected by forest fragmentation.

The effect of the lower density of canopy vegetation on the liana abundance, together with microclimatic changes and extreme events of regional climatic oscillations, may shift the composition of the liana species. These changes and oscillations may increase the average temperature and extend the dry season. The compositional shift of lianas may occur because of competition between coexisting species and increased species density. Newly added species with functional traits better adapted to new microclimate conditions, may prevail. The simple increase in density and dominance of just a few liana species may be far reaching on forest and ecosystem function. Under this future scenario, our knowledge of the response of lianas is extremely important to a holistic understanding of the forest dynamics.

Acknowledgments We thank the Biological Dynamics of Forest Fragments Project (BDFFP) and the National Institute for Amazonian Research (INPA) for providing access to the study sites, logistics, preliminary funding to RJB, and facilities. We thank Vicentini A for curating and organizing BDFFP tree data and Pequeno PACL for assistance in the statistical evaluations. This is study XXX of the Technical Series of the Biological Dynamics of Forest Fragments Project (BDFFP – INPA/STRI).

Funding This contribution is part of the master dissertation of NTM undertaken at the National Institute for Amazonian Research (INPA), with a studentship from Fundação de Amparo à Pesquisa do Amazonas (FAPEAM – process n.º 008/2021). EEM was funded by the Academy of Finland (decision numbers 318252, 319905, and 345472).

Declarations

Competing interests The authors have no competing interests to declare that are relevant to the content of this article

Ethics approval Ethics approval was not required for this study according to local legislation

Consent to participate Not applicable

Consent for publication Not applicable

Availability of data and material The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Code availability study is available from the corresponding author on reasonable request.

Albiero-Júnior A, Venegas-González A, Botosso PC, Roig FA, Camargo JLC, Tomazello-Filho M (2019) What is the temporal extension of edge effects on tree growth dynamics? A dendrochronological approach model using Scleronema micranthum (Ducke) Ducke trees of a fragmented forest in the Central Amazon. Ecol Indic 101:133–142. https://doi.org/10.1016/j.ecolind.2018.12.040
Albiero-Júnior A, Venegas-González A, Camargo JLC, Roig FA, Tomazello-Filho M (2021) Amazon forest fragmentation and edge effects temporarily favored understory and midstory tree growth. Trees - Structure and Function 35:2059–2068. https://doi.org/10.1007/S00468-021-02172-1/FIGURES/5
Almeida DRA, Stark SC, Schietti J, Camargo JLC, Amazonas NT, Gorgens EB et al (2019) Persistent effects of fragmentation on tropical rainforest canopy structure after 20 year of isolation. Ecol Appl 0:e01952. https://doi.org/10.1002/eap.1952
Amazonas. Secretaria de Estado do Meio Ambiente (2019) Plano estadual de recursos hídricos do Amazonas - PERH/AM - RT-03: diagnóstico, prognóstico e cenários futuros dos recursos hídricos do estado – relatório preliminar. Available on: http://meioambiente.am.gov.br/wp-content/uploads/2016/04/Produto-II-Diagn%C3%B3stico-Progn%C3%B3stico-e-Cen%C3%A1rios-volume-II.pdf
Blanchard G, Barbier N, Vieilledent G, Ibanez T, Hequet V, McCoy S et al (2023) UAV-Lidar reveals that canopy structure mediates the influence of edge effects on forest diversity, function and microclimate. J Ecol. https://doi.org/10.1111/1365-2745.14105
Books ME, Kristensen K, van Benthem KJ, Magnusson A, Berg CW, Nielsen A, Skaug HJ, Maechler M, Bolker BM (2017) glmmTMB Balances Speed and Flexibility Among Packages for Zero-inflated Generalized Linear Mixed Modeling. R J 9(2):378–400
Briant G, Gond V, Laurance SGW (2010) Habitat fragmentation and the desiccation of forest canopies: A case study from eastern Amazonia. Biol Conserv 143:2763–2769. https://doi.org/10.1016/j.biocon.2010.07.024
Burnham RJ (2002) Dominance, diversity and distribution of lianas in Yasuní, Ecuador: Who is on top? J Trop Ecol 18:845–864. https://doi.org/10.1017/S0266467402002559
Burnham RJ (2004) Alpha and beta diversity of Lianas in Yasuní, Ecuador. For Ecol Manage 190:43–55. https://doi.org/10.1016/j.foreco.2003.10.005
Cai Z-Q, Poorter L, Cao K-F, Bongers F (2007) Seedling Growth Strategies in Bauhinia Species: Comparing Lianas and Trees. Ann Bot 100:831–838. https://doi.org/10.1093/aob/mcm179
Calders K, Adams J, Armston J, Bartholomeus H, Bauwens S, Bentley LP et al (2020) Terrestrial laser scanning in forest ecology: Expanding the horizon. Remote Sens Environ 251:112102. https://doi.org/10.1016/J.RSE.2020.112102
Camargo JLC, Kapos V (1995) Complex Edge Effects on Soil Moisture and Microclimate in Central Amazonian Forest. Source: Journal of Tropical Ecology 11:205–221
Campbell M, Laurance WF, Magrach A (2014) Ecological effects of lianas in fragmented forests. In: Schnitzer S, Bongers F, Burnham R, Putz F (eds) Ecology of Lianas, 1st edn. JohnWiley & Sons, pp 443–450
Campbell MJ, Edwards W, Magrach A, Alamgir M, Porolak G, Mohandass D et al (2018) Edge disturbance drives liana abundance increase and alteration of liana–host tree interactions in tropical forest fragments. Ecol Evol 8:4237–4251. https://doi.org/10.1002/ECE3.3959
Chambers JQ, Asner GP, Morton DC, Anderson LO, Saatchi SS, Espírito-Santo FDB et al (2007) Regional ecosystem structure and function: ecological insights from remote sensing of tropical forests. Trends Ecol Evol 22:414–423. https://doi.org/10.1016/j.tree.2007.05.001
Coppieters K, Verbeeck H, Dequeker S, Powers JS, Vargas GG, Smith-Martin CM et al (2022) Two Co-occurring Liana Species Strongly Differ in Their Hydraulic Traits in a Water-Limited Neotropical Forest. Front Forests Global Change 5:836711. https://doi.org/10.3389/ffgc.2022.836711
Costa ES, Camargo JLC, Martins EMMA, Mendonça FC, Menezes BM (2023) Plano de manejo da Área De Relevante Interesse Ecológico Projeto Dinâmica Biológica De Fragmentos Florestais - ARIE PDBFF. Instituto Chico Mendes de Biodiversidade (ICMBio/MMA), Brasil. Available on: https://www.gov.br/icmbio/pt-br/assuntos/biodiversidade/unidade-de-conservacao/unidades-de-biomas/amazonia/lista-de-ucs/arie-projeto-dinamica-biologica-de-fragmentos-florestais/arquivos/minuta_plano_de_manejo_arie_pdbff_apos_portaria_de_aprovacao.pdf
Costa F, Castilho C, Drucker DP, Kinupp V, Nogueira A, Spironello W (2008) Flora. In: Oliveira M, Baccaro F, Braga-Neto R, Magnusson W (Org) (eds) Reserva Ducke: A biodiversidade amazônica através de uma grade. Áttema, Manaus, pp 21–30
Dambros CS (2020) csdambros/R-functions: First release. [https://github.com/csdambros/R-functions/tree/v1.0]. https://doi.org/10.5281/zenodo.3784397
Engelbrecht BMJ, Comita LS, Condit R, Kursar TA, Tyree MT, Turner BL et al (2007) Drought sensitivity shapes species distribution patterns in tropical forests. Nature 447:80–82. https://doi.org/10.1038/nature05747
Estrada-Villegas S, Schnitzer SA (2018) A comprehensive synthesis of liana removal experiments in tropical forests. Biotropica 50:729–739. https://doi.org/10.1111/BTP.12571
Fahey RT, Atkins JW, Gough CM, Hardiman BS, Nave LE, Tallant JM, Nadehoffer KJ, Vogel C, Scheuermann CM, Stuart-Haëntjens E, Haber LT, Fotis AT, Ricart R, Curtis PS (2019) Defining a spectrum of integrative trait-based vegetation canopy structural types. Ecol Lett 22:2049–2059. https://doi.org/10.1111/ele.13388
Fearnside PM, Leal-Filho N (2020) Solo e desenvolvimento na Amazônia: Lições do Projeto Dinâmica Biológica de Fragmentos Forestais. In: Fearnside P (ed) Destruição e Conservação da Floresta Amazônica. Editora Inpa, Manaus, pp 109–137
Ferreira LV, Laurance WF (1997) Effects of Forest Fragmentation on Mortality and Damage of Selected Trees in Central Amazonia. Biology (Basel) 11:797–801
Fischer R, Taubert F, Müller MS, Groeneveld J, Lehmann S, Wiegand T et al (2021) Accelerated forest fragmentation leads to critical increase in tropical forest edge area. Sci Adv 7:eabg7012. https://doi.org/10.1126/sciadv.abg7012
Garrido-Perez EI, Burnham RJ (2010) The evolution of host specificity in liana-tree interactions. Puente Biológico 3:145–157
Gerolamo CS, Costa FRC, Zuntini AR, Vicentini A, Lohmann LG, Schietti J, RochaEX, Angyalossy V, Nogueira A (2022) Hydro-Edaphic Gradient and Phylogenetic History Explain the Landscape Distribution of a Highly Diverse Clade of Lianas in the Brazilian Amazon. Front Forests Global Change 5:1–17. https://doi.org/10.3389/ffgc.2022.809904
Gerwing JJ, Schnitzer SA, Burnham RJ, Bongers F, Chave J, DeWalt SJ et al (2006) A Standard Protocol for Liana Censuses1. Biotropica 38:256–261. https://doi.org/10.1111/J.1744-7429.2006.00134.X
Haddad NM, Brudvig LA, Clobert J, Davies KF, Gonzalez A, Holt RD et al (2015) Habitat fragmentation and its lasting impact on Earth’s ecosystems. Sci Adv. https://doi.org/10.1126/SCIADV.1500052. 1:
Han T, Sánchez-Azofeifa GA (2022) Extraction of Liana Stems Using Geometric Features from Terrestrial Laser Scanning Point Clouds. Remote Sens (Basel) 14:4039. https://doi.org/10.3390/rs14164039
Hartig F (2022) _DHARMa: Residual Diagnostics for Hierarchical (Multi-Level / Mixed) Regression Models_. R package version 0.4.6, https://CRAN.R-project.org/package=DHARMa
Koeppen W (1948) Climatología: con un estudio de los climas de la Tierra. Fondo de Cultura Economica, Mexico City
Krishna SM, Bao Y, Calders K, Schnitzer SA, Verbeeck H (2019) Semi-automatic extraction of liana stems from terrestrial LiDAR point clouds of tropical rainforests. ISPRS J Photogrammetry Remote Sens 154:114–126. https://doi.org/10.1016/j.isprsjprs.2019.05.011
Laurance WF, Andrade AS, Magrach A, Camargo JLC, Valsko JJ, Campbell M et al (2014) Long-term changes in liana abundance and forest dynamics in undisturbed Amazonian forests. Ecology 95:1604–1611. https://doi.org/10.1890/13-1571.1
Laurance WF, Camargo JLC, Fearnside PM, Lovejoy TE, Williamson GB, Mesquita RCG et al (2018) An Amazonian rainforest and its fragments as a laboratory of global change. Biol Rev 93:223–247. https://doi.org/10.1111/BRV.12343
Laurance WF, Perez-Salicrup D, Delamônica P, Fearnside PM, D’Angelo S, Jerozolinski A et al (2001) Rain forest fragmentation and the structure of Amazonian liana communities. Ecology 82:105–116
Ledo A, Schnitzer SA (2014) Disturbance and clonal reproduction determine liana distribution and maintain liana diversity in a tropical forest. Ecology 95:2169–2178. https://doi.org/10.1890/13-1775.1
Li B, Zhang Y, Luan F, Yuan Z, Ali A, Chu C et al (2022) Habitat Conditions and Tree Species Shape Liana Distribution in a Subtropical Forest. Forests 13:1358. https://doi.org/10.3390/f13091358
Lomwong N, Chanthorn W, Nathalang A, Saenprasert R, Yaemphum S, Matmoon U et al (2023) Liana abundance and diversity increase along a successional gradient, even with homogeneous closed canopy. For Ecol Manage 534:120878. https://doi.org/10.1016/j.foreco.2023.120878
Lovejoy TE, Bierregaard RO, Rylands AB, Malcolm JR, Quintela C, Harper L et al (1986) Edge and other effects of isolation on Amazon forest fragments. In: Soulé ME (ed) Conservation Biology: The Science of Scarcity and Diversity. Sinauer, Sunderland, Massachusetts, pp 257–285
Lüdecke D, Ben-Shachar MS, Patil I, Waggoner P, Makowskiet D (2021) performance: An R Package for Assessment, Comparison and Testing of Statistical Models. J Open Source Softw 6(60):3139. https://doi.org/10.21105/joss.03139
Ma L, Zheng G, Ying Q, Hancock S, Ju W, Yu D (2021) Characterizing the three-dimensional spatiotemporal variation of forest photosynthetically active radiation using terrestrial laser scanning data. Agric For Meteorol 301:108346. https://doi.org/10.1016/j.agrformet.2021.108346
Maeda EE, Nunes MH, Calders K, de Moura YM, Raumonen P, Tuomisto H et al (2022) Shifts in structural diversity of Amazonian forest edges detected using terrestrial laser scanning. Remote Sens Environ 271:112895. https://doi.org/10.1016/J.RSE.2022.112895
Magrach A, Rodríguez-Pérez J, Campbell M, Laurance WF (2014) Edge effects shape the spatial distribution of lianas and epiphytic ferns in Australian tropical rain forest fragments. Appl Veg Sci 17:754–764. https://doi.org/10.1111/AVSC.12104
Meunier F, Verbeeck H, Cowdery B, Schnitzer SA, Smith-Martin CM, Powers JS et al (2021) Unraveling the relative role of light and water competition between lianas and trees in tropical forests: A vegetation model analysis. J Ecol 109:519–540. https://doi.org/10.1111/1365-2745.13540
Meunier F, Visser MD, Shiklomanov A, Dietze MC, Guzmán Q, JA, Sanchez-Azofeifa GA et al (2022) Liana optical traits increase tropical forest albedo and reduce ecosystem productivity. Glob Chang Biol 28:227–244. https://doi.org/10.1111/GCB.15928
Montibeller B, Kmoch A, Virro H, Mander Ü, Uuemaa E (2020) Increasing fragmentation of forest cover in Brazil’s Legal Amazon from 2001 to 2017. Sci Rep 10:5803. https://doi.org/10.1038/s41598-020-62591-x
Nogueira A, Costa FRC, Castilho CV (2011) Liana Abundance Patterns: The Role of Ecological Filters during Development. Biotropica 43:442–449. https://doi.org/10.1111/j.1744-7429.2010.00722.x
Nunes MH, Camargo JLC, Vincent G, Calders K, Oliveira RS, Huete A et al (2022) Forest fragmentation impacts the seasonality of Amazonian evergreen canopies. Nat Commun 13:917. https://doi.org/10.1038/s41467-022-28490-7
Ofosu-Bamfo B, Addo-Fordjour P, Belford EJD (2022) Edge disturbance shapes liana diversity and abundance but not liana-tree interaction network patterns in moist semi-deciduous forests. Ghana Ecol Evol 12:e8585. https://doi.org/10.1002/ece3.8585
Oksanen J, Simpson G, Blanchet F, Kindt R, Legendre P, Minchin P, O'Hara R, Solymos P, Stevens M, Szoecs E, Wagner H, Barbour M, Bedward M, Bolker B, Borcard D, Carvalho G, Chirico M, De Caceres M, Durand S, Evangelista H, FitzJohn R, Friendly M, Furneaux B, Hannigan G, Hill M, Lahti L, McGlinn D, Ouellette M, Ribeiro Cunha E, Smith T, Stier A, Ter Braak C, Weedon J (2022) _vegan: Community Ecology Package_. R package version 2.6-4, https://CRAN.R-project.org/package=vegan
Phillips OL, Vésquez Martínez R, Arroyo L, Baker TR, Killeen T, Lewis SL et al (2002) Increasing dominance of large lianas in Amazonian forests. Nat 2002 418:6899. https://doi.org/10.1038/nature00926
Piovesan PRR, Burnham RJ, Ferraz IDK, Camargo JLC (2022) Species density diverges after forest fragmentation in lianescent Machaerium Pers. (Fabaceae) in Central Amazonia. For Ecol Manage 519:120335. https://doi.org/10.1016/J.FORECO.2022.120335
R Core Team (2022) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/
Reich PB (2012) Key canopy traits drive forest productivity. Proceedings of the Royal Society B: Biological Sciences 279:2128–2134. https://doi.org/10.1098/RSPB.2011.2270
Rocha EX, Nogueira A, Costa FRC, Burnham RJ, Gerolamo CS, Honorato CF et al (2022) Liana functional assembly along the hydrological gradient in Central Amazonia. Oecologia 200:183–197. https://doi.org/10.1007/s00442-022-05258-w
Rocha EX, Schietti J, Gerolamo CS, Burnham RJ, Nogueira A (2020) Higher rates of liana regeneration after canopy fall drives species abundance patterns in central Amazonia. J Ecol 108:1311–1321. https://doi.org/10.1111/1365-2745.13345
Sánchez-Azofeifa GA, Guzmán-Quesada JA, Vega-Araya M, Campos-Vargas C, Durán SM, D’Souza N et al (2017) Can terrestrial laser scanners (TLSs) and hemispherical photographs predict tropical dry forest succession with liana abundance? Biogeosciences 14:977–988. https://doi.org/10.5194/bg-14-977-2017
Schietti J, Emilio T, Rennó CD, Drucker DP, Costa FRC, Nogueira A et al (2014) Vertical distance from drainage drives floristic composition changes in an Amazonian rainforest. Plant Ecol Divers 7:241–253. https://doi.org/10.1080/17550874.2013.783642
Schnitzer SA (2018) Testing ecological theory with lianas. New Phytol 220:366–380. https://doi.org/10.1111/NPH.15431
Schnitzer SA, Bongers F (2002) The ecology of lianas and their role in forests. Trends Ecol Evol 17:223–230. https://doi.org/10.1016/S0169-5347(02)02491-6
Schnitzer SA, DeFilippis DM, Visser M, Estrada-Villegas S, Rivera-Camaña R, Bernal B et al (2021) Local canopy disturbance as an explanation for long-term increases in liana abundance. Ecol Lett 24:2635–2647
Schnitzer SA, DeWalt SJ, Chave J (2006) Censusing and Measuring Lianas: A Quantitative Comparison of the Common Methods1. Biotropica 38:581–591. https://doi.org/10.1111/J.1744-7429.2006.00187.X
Schnitzer SA, Estrada-Villegas S, Wright SJ (2020) The response of lianas to 20 year of nutrient addition in a Panamanian forest. Ecology 101:e03190. https://doi.org/10.1002/ECY.3190
Schnitzer SA, Rutishauser S, Aguilar S (2008) Supplemental protocol for liana censuses. For Ecol Manage 255:1044–1049. https://doi.org/10.1016/j.foreco.2007.10.012
Song S, Schnitzer SA, Ding Y, Wang G, Chen L, Liu J et al (2023) Light-demanding tree species are more susceptible to lianas than shade-tolerant tree species in a subtropical secondary forest. J Ecol. https://doi.org/10.1111/1365-2745.14134
Van Der Heijden GMF, Phillips OL, Schnitzer SA (2014) Impacts of lianas on forest-level carbon storage and sequestration. In: Schnitzer S, Bongers F, Burnham R, Putz F (eds) Ecology of Lianas. John Wiley & Sons, New Jersey, pp 164–174
Vogado NO, Engert JE, Linde TL, Campbell MJ, Laurance WF, Liddell MJ (2022) Climate Change Affects Reproductive Phenology in Lianas of Australia’s Wet Tropics. Front Forests Global Change 5:787950. https://doi.org/10.3389/ffgc.2022.787950
Yang SZ, Fan H, Li KW, Ko TY (2018) How the diversity, abundance, size and climbing mechanisms of woody lianas are related to biotic and abiotic factors in a subtropical secondary forest. Taiwan Folia Geobot 53:77–88. https://doi.org/10.1007/S12224-017-9306-Z/FIGURES/5
Yuan C, Liu W, Tang CQ, Li X (2009) Species composition, diversity, and abundance of lianas in different secondary and primary forests in a subtropical mountainous area, SW China. Ecol Res 24:1361–1370. https://doi.org/10.1007/s11284

SUPPLEMENTARYINFORMATION.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

The influence of forest structure on the abundance, biomass, and composition of lianas in tropical forest fragments

Status:

Version 1

Abstract

Figures

Introduction

Materials and Methods

Study area

Database and sampling design

Statistical analysis

Results

Floristic and phytosociological aspects of lianas

Forest structure and liana community

Forest structure and taxonomic composition of the liana community

Discussion

High diversity and site effect

Canopy structure and distance to forest edge affect liana community

Biomass and support capacity for lianas

Conclusion

Declarations

References

Supplementary Files

Status:

Version 1