Differentiation in stem and leaf traits among sympatric lianas, scandent shrubs and trees in a subalpine cold temperate forest

Abstract

The scandent shrub plant form is a variant of liana that has upright and self-supporting stems when young but later becomes a climber. We aimed to explore the associations of stem and leaf traits among sympatric lianas, scandent shrubs and trees, and the effects of growth form and leaf habit on variation in stem or leaf traits. We measured 16 functional traits related to stem xylem anatomy, leaf morphology and nutrient stoichiometry in eight liana, eight scandent shrub and 21 tree species co-occurring in a subalpine cold temperate forest at an elevation of 2600-3200 m in Southwest China. Overall, lianas, scandent shrubs and trees were ordered along a fast-slow continuum of stem and leaf functional traits, with some traits overlapping. We found a consistent pattern of lianas > scandent shrubs > trees for hydraulically weighted vessel diameter, maximum vessel diameter and theoretical hydraulic conductivity. Vessel density and sapwood density showed a pattern of lianas = scandent shrubs < trees, and lianas < scandent shrubs = trees, respectively. Lianas had significantly higher specific leaf area and lower carbon concentration than co-occurring trees, with scandent shrubs showing intermediate values that overlapped with lianas and trees. The differentiation among lianas, scandent shrubs and trees was mainly explained by variation in stem traits. Additionally, deciduous lianas were positioned at the fast end of the trait spectrum, and evergreen trees at the slow end of the spectrum. Our results showed for the first time clear differentiation in stem and leaf traits among sympatric liana, scandent shrub and tree species in a subalpine cold temperate forest. This work will contribute to understanding the mechanisms responsible for variation in ecological strategies of different growth forms of woody plants.

Keywords: ecological strategy; functional traits; growth form; high elevation; leaf habit; nutrient stoichiometry; woody vine.

Figure 1.

Differences in means (±SE) of hydraulically weighted vessel diameter (D_h; a), maximum vessel diameter (D_max; b), theoretical hydraulic conductivity (K_t; c), sapwood density (WD; d), vessel density (VD; e), SLA (f), LD (g), carbon concentration (C; h), phosphorus concentration (P; i) and N/P ratio (j) among eight liana (L), eight scandent shrub (SS) and 21 tree (T) species. Different letters above bars indicate significant differences (P < 0.05) among groups according to one-way ANOVA (Tukey’s HSD).

Figure 2.

The log–log relationships of stem theoretical hydraulic conductivity (K_t) with (a) wood density (WD), (b) VD and (c) hydraulic weighted vessel diameter (D_h) across eight lianas (red circles), eight scandent shrubs (blue triangles) and 21 trees (green squares). See Table 3 for regression statistics.

Figure 3.

The log–log relationships of SLA with (a) LT, (b) foliar carbon concentration (C), (c) stable carbon isotopic composition (δ¹³C) and (d) potassium concentration (K) across eight lianas (red circles), eight scandent shrubs (blue triangles) and 21 trees (green squares). See Table 3 for regression statistics.

Figure 4.

The log–log relationships (a) between SLA and wood density (WD) and the relationships of theoretical hydraulic conductivity (K_t) with (b) foliar carbon concentration (C), (c) SLA and (d) potassium concentration (K) across eight lianas (red circles), eight scandent shrubs (blue triangles) and 21 trees (green squares). See Table 3 for regression statistics.

Figure 5.

Position of (a) traits and (b) species on first two axes of principal component analysis for all traits included in this study. In (b) lines were drawn around groups of species with different growth forms. D-L: deciduous lianas; E-L: evergreen lianas; D-S: deciduous scandent shrubs; E-S: evergreen scandent shrubs, D-T: deciduous trees; E-T: evergreen trees. Species codes were shown in Table 1. See Table 2 for trait abbreviations.

Figure 6.

The fraction of variance in stem and leaf traits explained by growth form and leaf habit based on two-way MANOVA. See Table 2 for trait abbreviations.