Materials and Methods
Plant material and localities
122 herbaceous species were considered from two major plant functional types (forbs, n = 94; graminoids, n = 28) across a broad range of water availability conditions in European temperate grasslands (from dry through mesic to wet grasslands; Supporting Information S1). Species were sampled in the area of the White Carpathians Mts. (SE Czech Republic) and in Ohrazení (S Czech Republic), In both areas, the mean annual temperature is 7–9 °C, mean annual precipitation is 600–920 mm and they have been traditionally managed by mowing once a year in June or July (for details see Lepš 2014; Mudrák et al. 2019). These grasslands are renowned for high species diversity (Chytrý et al. 2015). The dominant species in the White Carpathian Mts. meadows areBromus erectus , Molinia arundinacea , and Carex montana , with common forbs such as Agrimonia eupatoria ,Centaurea jacea , Geranium sanguineum , Hypericum maculatum , Leucanthemum vulgare agg. and Vincetoxicum hirundinaria . In the Ohrazení meadow, the dominant grasses areMolinia caerulea and Holcus lanatus , alongside with about 10 species of sedges and common forbs such as Angelica sylvestris , Betonica officinalis , Galium boreale ,Potentilla erecta , Ranunculus acris and Lychnis flos-cuculi .
Trait measurements
πtlp was measured for all 122 species in late May to early June 2016 following the protocol described in detail in (Májekováet al. 2019). Prior to measurements, plants were rehydrated overnight in dark cool room with aboveground parts sealed in a plastic bag. For the 101 species from the White Carpathians, specific leaf area (SLA; m2 kg–1), leaf dry matter content (LDMC; mg g–1) and vegetative height (height; cm) were measured on rehydrated plants at the same time as πtlp following the standardized protocol (Pérez-Harguindeguy et al. 2013). For a subset of 20 species from the White Carpathians, δ13C was measured in a previous campaign in autumn 2006 (described in detail in de Bello et al.2012). For the 37 species from Ohrazení, SLA, LDMC, height and δ13C were measured in a previous campaign in June 2013 (described in detail in (Mudrák et al. 2019) following the standardized protocol (Pérez-Harguindeguy et al. 2013).
Statistical analyses
We used a one-way ANOVA to test for the differences in traits between the plant functional types (PFTs; forb and graminoid). We tested the coordination between πtlp and other functional traits (height, SLA, LDMC and δ13C) with standardised major axis regression, using the package SMATR (Warton et al., 2012). First, we considered a model with only traits (hereafter ‘All’). Second, we added PFTs with comparison of slopes of the trait relationships between the forbs and graminoids. Third, we considered the phylogenetic relatedness among species. For this, an age-calibrated phylogeny of species (Durka & Michalski 2012) was used to test whether traits exhibited a phylogenetic signal by calculating Pagel’s lambda (Pagel 1999) using the package phytools (Revell 2012). Since all traits (except for SLA) showed a phylogenetic signal, we applied phylogenetically independent contrasts (Felsenstein 1985) implemented in the package ape (Paradis et al. 2004) to account for phylogenetic relatedness among species. The contrasts where then used instead of the raw data in the standardised major axis analyses, with intercept forced through zero (Garland et al. 1992). It needs to be noted that the PFTs considered here as forbs and graminoids are phylogenetically monophyletic and these two PFTs mainly reflect phylogeny (i.e. a graminoid clade vs. ’the rest’). The goal was not to use phylogeny as addition to the PFTs, but rather as two alternative methods that are commonly used in trait-based ecology. Height was log10-transformed for all analyses to meet the assumption of the homogeneity of variances. All analyses were performed in the R software (R Core Team 2016).