Introduction
Tropical forests contain more than half of the Earth’s terrestrial species, store one quarter of the global terrestrial carbon and account for one-third of net primary productivity (Bonan, 2008; Wright, 2010). Therefore, any alteration in the structure or functioning of tropical forests has important consequences for biodiversity conservation, productivity and global carbon budgets. Currently, one of the most eminent changes occurring over the last three decades in tropical forests is the increase in liana abundance and biomass (Schnitzer & Bongers, 2011; Yorke et al., 2013; Schnitzer, 2018). Lianas (woody climbers) are non-self-supporting plants which rely on trees as supports to grow and to gain access to the forest canopy (Schnitzer & Bongers, 2002; Gerwing et al., 2006). They are a prominent growth form across lowland tropical forests where they may represent up to 40% of the woody stems and more than 25% of the woody species (Gentry, 1991). Lianas invest proportionally more in leaf mass than in stem cross-sectional area than trees, thus, deploying a large canopy of leaves on the host trees that they infest (Putz, 1984; Gerwing & Farias, 2000). Intense competition for above- and below-ground resources with lianas may limit tree recruitment, growth, reproduction and survival (Putz, 1984; Stevens, 1987; Clark & Clark, 1990; Schnitzer et al., 2000; Ingwell et al., 2010). Moreover, the negative effect of lianas is greater on late-successional- than on pioneer tree species (Schnitzer & Carson, 2010); this has enormous consequences for community richness and composition, as well as for ecosystem-level dynamics such as carbon storage (Schnitzer & Bongers, 2011; Schnitzer, 2018). Currently, at least eight studies have provided evidence for increases in liana abundance or biomass, or both, across Neotropical forests (Schnitzer & Bongers, 2011). However, not all studies support the liana increase hypothesis. For instance, only two studies have evaluated the lianas’ growth rates for the African continent and these did not find an increase in their abundance or biomass (Schnitzer & Bongers, 2011; Schnitzer, 2018, Bongers et al. 2020). One possible explanation for these different results across the tropics may be due to taxonomic and functional trait differences among the species present in each of those continents. In fact, Gallagher & Leishman (2012), in a global analysis for climbing plants, found a difference in the proportion of species having tendril or twining climbing mechanisms in Africa and the Americas. However, whether these taxonomic differences are related to specific morphological and physiological traits is still not clear. One reason is that most of the studies investigating the morphological and physiological traits in lianas have contrasted them with trees, while few have addressed differences amongst lianas differing in their climbing mechanisms (Carter & Teramura, 1988; Durigon et al., 2013). Consequently, the current predominant view considers lianas as a uniform group and, so far, no comprehensive work has been carried out to test if different climbing strategies may also account for functional trait variability among liana species (Medina et al., 2021). Recognizing functional and morphological differences among liana climbing mechanisms may be essential for understanding their spatial and temporal patterns of abundance across tropical forests, particularly if the increase in liana abundance and biomass is driven by taxa with a specific climbing mechanism and growth strategy. Therefore, if differences between groups of lianas exist and are not accounted for, this may lead to a misrepresentation of the role of lianas role in forest successional dynamics, which, in turn, will have important implications for our understanding of forest responses to climate change.
Lianas have long been attracting the interest of botanists because of their distinctive climbing mechanisms and growth strategies (Darwin, 1875; Isnard & Silk, 2009) (see Box 1). The vertical growth of lianas is facilitated via a large diversity of attaching systems (Putz, 1984; Nick & Rowe, 2004; Isnard & Silk, 2009; Speck & Burgert, 2011). Each type of climbing mechanism (see Fig. 1) may differ in carbon allocation and the mechanical support requirements necessary for vertical growth, consequently determining the differences in life history among the liana species. For example, it has been proposed that each type of climbing mechanism determines the maximum distance between supports and that climbing success depends on the maximum stem diameter support that they can colonize (Putz & Holbrook, 1991; Isnard & Silk, 2009; Rowe & Speck, 2004). Darwin, in his book “The climbing movement of plants” (1875) (the first study to investigate the ecological behavior of climbing plants), did not consider species which merely scramble over vegetation without any special organ. He described hook-climbers as having the least efficient climbing mechanism, followed by hook and root climbers, while he considered the more widespread twining and tendril climbers to be the most efficient. Those early observations are in accordance with modern-day measured mechanical properties of wood from different types of climbers. For instance, tendril and twining species generally have more flexible stems, while hook-climbers and scramblers retain relatively stiff stems throughout their lives (Isnard & Silk, 2009; Speck & Burgert, 2011; Rowe & Speck, 2004). Considering the lianas life history, one emerging question is whether an active climbing mechanism, which facilitates easier access to the upper canopy, is associated with a resource-acquisitive strategy via functional traits. The theoretical basis for this hypothesis relies on the fact that the type of climbing mechanism limits the maximum size of the support used for vertical growth (e.g., the stem diameter to which a specific climbing mechanism can attach and, consequently, climb on), as well as the light environment that can be explored (Darwin, 1875; Putz, 1984; Holbrook & Putz, 1991). Therefore, the success of certain types of climbing mechanisms will differ according to the forest structure and successional stages (Putz & Chai, 1987; DeWalt et al., 2000). Likewise, there is evidence that the proportion of species with different climbing mechanisms differ significantly between continents (Gallagher and Leishman, 2012), but the main drivers (climate, soil and forest structure) of these patterns are still not explored.
Plant functional types have been used by ecologists to aggregate the enormous number of plant species into a tractable number of functionally similar classes (Ustin & Gamon, 2010). This classification implies an ecological basis to differentiate between groups of plants responding similarly to changes in their environments (Diaz & Cabido, 1997; Lavorel & Garnier, 2002), while there is also strong support for significant differences in key traits among plants of different functional types (Wright et al., 2004, 2005; Verheijen et al., 2016); such classification is necessary in order to reduce the need for specific knowledge about each underlying species functional trait. Plant traits are suitable tools for describing the different functional aspects of plants and their relationships to environmental conditions. The various growth and survival strategies of plants in response to abiotic and biotic determinants are reflected by their set of trait values (Westoby et al., 2002; Reich, 2014; Verheijen et al., 2016). An increasing number of global overviews have been reported for various functional groups of plants (e.g., Wright et al., 2005a, b; Poorter et al., 2012; Verheijen et al., 2016) and it becomes clear that several questions characterizing lianas as a group still need to be answered (Wylka et al., 2013; Schnitzer, 2018; Medina et al., 2021). Although the morphology, anatomy and physiology of the climbing organs used by lianas to reach the canopy have been relatively well studied (for a review see Isnard & Silk, 2009), much less effort has been made to understand if there is a relationship between the type of climbing mechanism and the functional traits related to whole-plant performance, such as the leaf and wood economic spectra traits (Wright et al., 2004; Chave et al., 2009; Reich, 2014). If such links exist, this information would be especially needed to properly account for the role of lianas in forest dynamics via Dynamic Global Vegetation Models (DGVMs) (Verbeeck et al., 2016; Schnitzer et al., 2016, Meunier et al., 2021). Similarly, the few studies which have characterized the drivers of liana abundance and species richness have focused on only a few environmental variables and have mainly compared lianas to trees (Schnitzer, 2005), focused on specific liana clades and biogeographical realms such as the Neotropics (Meyer et al., 2020), or on a particular type of climbing mechanism (Durigon et al., 2013). Therefore, we still do not know how the environment (e.g., climate and soil) and the forest structure drive the proportion of species (richness) and abundance (density) of lianas regarding their main climbing strategies across the large spatial scales.
Classifications of the climbing mechanisms used by lianas may vary substantially throughout literature, ranging from three categories (Vaughn & Bowling, 2011) to as high as nine in some cases (Addo-Fordjour & Rahmad, 2015), as recently reviewed by Sperotto et al. (2020). In this study, we followed the recently proposed classification of climbing mechanisms from Sperotto et al. (2020) to establish a framework which considers a continuum from highly-specialized obligate lianas (active climbing) to species with an intermediate behavior (passive climbing) and devoid of highly specialized attachment systems (Fig. 1). Here we analyzed the functional strategies and distribution of lianas, classifying lianas into two main types of climbing mechanisms: active and passive. In general, active climbing comprises twining, tendrils, prehensile branches, twining petioles and prehensile peduncles and inflorescences, while passive climbing comprises scrambling (or clambering, leaning), hooks or grapnels and adhesive roots.
The current perspective regarding the ecology of lianas remains somewhat limited because previous analyses of liana ecology have largely focused on the differences between lianas and trees and, so far, no major synthesis has considered the life history differences among lianas in terms of their climbing mechanisms. Here we synthesize data for lianas regarding the main functional traits used to characterize plant form and function and, thus, classify the liana species according to their main type of climbing mechanisms (Fig. 1, Box 1): active and passive climbing. Furthermore, we integrate the liana species climbing mechanism data with a standardized pantropical plot sampling of the species richness and abundance, in addition to remote sensing data (climate and soil) to explore the following hypotheses:
(1) There is an association between the climbing mechanism and functional traits, therefore, we expect that functional traits differ between active and passive climbing lianas. More specifically, we expect that species with active climbing, that forage vertically through a short-term vigorous growth, show functional traits towards the acquisitive end of the global trait spectra.
(2) The magnitude and direction of the associations between the main drivers (climate, soil and forest structure) of the relative abundance and richness of lianas differ between active and passive climbing species (Table 1). We expect that taller forests will favor active climbers, whereas shorter forests will favor passive climbers. We expect that climate and soil fertility show a positive relationship to species richness for both active and passive climbing, while abundance may respond differently to those variables because recruitment is also affected by forest structure (availability of suitable supports).
Here, we first show that functional trait differences are significant between the two main categories of liana climbing mechanisms: active and passive climbing (Sperotto et al., 2020). Subsequently, we show that the factors driving abundance and species richness of lianas across tropical forests differ significantly between active and passive climbing species. Finally, we discuss the need of reconsidering the view of lianas as a single, plant functional type, given the importance of the climbing mechanism to the life history and ecology of lianas.
Box 1. The definition of climbing mechanisms and their importance in the life-history of liana.
Lianas are a particular type of climbing plant that develop woody stems and have a non-self-supporting habit so they rely upon an external support for vertical growth towards the canopy (Rowe and Speck, 2004; Schnitzer & Bongers, 2002). In general, the ontogeny of lianas is marked by two distinct phases: firstly, the erect phase, where, after germinating, they can grow vertically adopting a self-supporting habit and can reach up to 2 m height without the need for an external support and, secondly, the climbing phase (Putz & Holbrook, 1991), which is marked by a rapid increase in the internode distance with low investment in ramification and foliar expansion until a suitable support is found (Hegarty, 1991; Isnard & Silk, 2009; Ichihashi and Tateno, 2015). This marked plasticity in response to the availability of support is associated with several changes in the anatomy, morphology and physiology of the leaves, stems and branches; this is a key feature in the life history of many liana species (Putz, 1984; Gentry, 1991; Isnard & Silk, 2009; Chen et al., 2014). The need for finding a support for vertical growth is also an important determinant of the spatial distribution of lianas inside forests (Putz, 1984; Putz & Holbrook, 1991; Schnitzer & Bongers, 2002). For example, tendril species are restricted to attach to trees with smaller diameters than twiner species (Putz, 1984; Putz and Chai, 1987), therefore, in forests with high rates of fragmentation or gap formation, where more small trees are present, tendril species tend to show higher abundance (DeWalt et al., 2005; Letcher & Chazdon, 2009).
The transition between the self- and non-self-supporting phase described above is also heavily dependent on the type of climbing mechanism that lianas can use to attach to the support structures. Lianas exhibit a diversity of morphological and anatomical variation (Putz & Holbrook, 1991; Isnard & Silk, 2009; Rowe & Speck, 2004) related to the type of attachment and the organs (e.g., leaf, branch or stem) used to adhere to the support structures (climbing strategy). In part, this diversity results from the unique demand to maintain stems that are both flexible to twist, but also strong enough to climb and span across different hosts without breaking during the search for light (Rowe & Speck, 2004; Isnard & Silk, 2009).
Species with an active climbing mechanism display a support-searching behavior such as circumnutation, an endogenous growth-related rhythmic movement in which leader shoots (i.e., shoots produced to search for supports) sweep through the air in arcs (Darwin, 1875; Hegarty, 1991; Carlquist, 1991; Putz et al., 1991; Isnard & Silk, 2009). This greatly increases the possibility of the plant finding suitable supports. On the other hand, species with a passive climbing mechanism do not actively search for support, rather, they merely grow over the host plants without any searching movements such as circumnutation. Additionally, species with a passive climbing mechanism may have adhesive roots or specialized grapnels or hooks to ensure their attachment to supports.