Skip to main content
Download PDF

Biotic and abiotic drivers of intraspecific trait variation within plant populations of three herbaceous plant species along a latitudinal gradient

BMC Ecology volume 17, Article number: 38 (2017) Cite this article

Abstract

Background

The importance of intraspecific trait variation (ITV) is increasingly acknowledged among plant ecologists. However, our understanding of what drives ITV between individual plants (ITVBI) at the population level is still limited. Contrasting theoretical hypotheses state that ITVBI can be either suppressed (stress-reduced plasticity hypothesis) or enhanced (stress-induced variability hypothesis) under high abiotic stress. Similarly, other hypotheses predict either suppressed (niche packing hypothesis) or enhanced ITVBI (individual variation hypothesis) under high niche packing in species rich communities. In this study we assess the relative effects of both abiotic and biotic niche effects on ITVBI of four functional traits (leaf area, specific leaf area, plant height and seed mass), for three herbaceous plant species across a 2300 km long gradient in Europe. The study species were the slow colonizing Anemone nemorosa, a species with intermediate colonization rates, Milium effusum, and the fast colonizing, non-native Impatiens glandulifera.

Results

Climatic stress consistently increased ITVBI across species and traits. Soil nutrient stress, on the other hand, reduced ITVBI for A. nemorosa and I. glandulifera, but had a reversed effect for M. effusum. We furthermore observed a reversed effect of high niche packing on ITVBI for the fast colonizing non-native I. glandulifera (increased ITVBI), as compared to the slow colonizing native A. nemorosa and M. effusum (reduced ITVBI). Additionally, ITVBI in the fast colonizing species tended to be highest for the vegetative traits plant height and leaf area, but lowest for the measured generative trait seed mass.

Conclusions

This study shows that stress can both reduce and increase ITVBI, seemingly supporting both the stress-reduced plasticity and stress-induced variability hypotheses. Similarly, niche packing effects on ITVBI supported both the niche packing hypothesis and the individual variation hypothesis. These results clearly illustrates the importance of simultaneously evaluating both abiotic and biotic factors on ITVBI. This study adds to the growing realization that within-population trait variation should not be ignored and can provide valuable ecological insights.

Background

Functional traits determine plant species’ vital rates and fitness [1, 2]. Consequently, species under contrasting environmental conditions exhibit very different functional trait attributes due to diverging selective pressures [1, 3]. Interestingly, many plant species occur over relatively broad spatial scales and are thus exposed to strong abiotic and biotic gradients. This suggests that species exhibit large variation in their functional trait attributes across these gradients, caused by adaptation by natural selection or/and phenotypic plasticity [4]. The recent interest in intraspecific trait variation (ITV) from ecologists has detected considerable ITV in many species’ traits, even within populations (e.g. [5,6,7,8,9]). Since functional traits and their intraspecific variability are closely linked to plant community dynamics [10,11,12], ecosystem functions such as litter decomposition [13,14,15], ecosystem services such as disease risk reduction [15] and species responses to climate change [16,17,18], it is important to understand which factors drive ITV in plant species [4, 19].

Until now, most studies have focused on between-population ITV (e.g. [20,21,22]) and on quantifying the responsiveness of traits to environmental variation (due to trait plasticity) at the species level (e.g. [23, 24]). However, significant levels of trait variation are known to occur within populations as well (between individual ITV, ITVBI sensu Albert et al. [4]) (e.g. [5, 25]). Although several studies have focused on quantifying the range of ITVBI (e.g. [7, 21, 26, 27]), our understanding of what drives the extent of trait variation at the population level is still relatively limited (see for instance [17, 28,29,30,31]). Research at both the species and community level suggested that phenotypic plasticity and thus ITVBI are more constrained under unfavourable abiotic conditions (environmental stress) due to environmental filtering, resulting in a decreased ITVBI in cold and dry climates and in nutrient-poor sites (stress-reduced plasticity hypothesis) [29, 32, 33]. Other hypotheses state, however, that unfavourable conditions may trigger enhanced expression of phenotypic variability in traits, thus resulting in increased ITVBI in stressful environments (stress-induced variability hypothesis) [8, 34, 35]. Two additional hypotheses have been proposed concerning the possible effect of biotic drivers (and more specifically of competition) on ITVBI. First, niche theory predicts that highly diverse plant communities should be characterized by reduced ITVBI, due to increased interspecific competition and selection for reduced niche overlap (communities with low niche overlap and high niche density) (niche packing hypothesis) [8, 36]. Similarly, asymmetrical competition for light (i.e. taller species are disproportionately advantaged) in highly productive and competitive communities is expected to have comparable effects on ITVBI as high niche density (niche packing) [28, 29]. Second, the individual variation hypothesis assumes that all individuals in a community, irrespective of species identity, will exhibit different trait attributes to avoid (intraspecific) competition [10]. Consequently, this high within-species trait variation favours the maintenance of high species diversity, resulting in a positive correlation between ITVBI and species diversity [10, 36].

Some progress has been made in quantifying the extent of community-wide trait overlap/ITV and its dependency on species richness. Support exists for both the niche packing hypothesis, with decreased ITVBI in several traits, such as plant height and specific leaf area (SLA) under high species richness [8, 33] and for the individual variation hypothesis, with an increase in plant height ITVBI under high species richness [37]. Although species richness provides an easily measurable proxy to quantify community-level niche density, research suggests that measures of functional diversity at the community level more readily reflect niche-based processes [38]. This suggests that using the different components of functional diversity, namely functional richness, functional evenness and functional divergence [39], likely allow a more precise assessment of the importance of niche-based processes on species-level ITVBI. To our knowledge, no studies have evaluated species- or functional diversity effects on ITVBI patterns at the species level (however see [31, 40, 41]).

The relative importance of abiotic and biotic conditions on ITVBI is furthermore expected to change along macroecological gradients. While abiotic stress is expected to increase in environmentally unfavourable conditions, competition is usually more associated with environmentally favourable conditions (cf. the stress-gradient hypothesis, [42]). Additionally, the impact of stress and competition on ITVBI is most likely species- and trait-dependent. Indeed, research on invasive plant species has suggested that colonization rates (and thus invasion success) are higher for species that exhibit high ITV [43]. This suggests that species with high colonization capacity are more responsive to abiotic and biotic drivers of ITVBI. Regarding trait dependency, one could expect that the ITVBI of growth related traits are more responsive to competition-related biotic drivers than that of reproductive traits. Similarly, the ITVBI of traits that are known to be closely associated with stress responses (e.g. specific leaf area) might be more responsive to abiotic variation [3].

Here, we assess the relative effects of both abiotic (stress-related) and biotic (competition-related) variation on ITVBI of four functional traits (leaf area, SLA, plant height and seed mass), for three herbaceous plant species (Anemone nemorosa, Milium effusum and Impatiens glandulifera) across a 2300 km long macroecological gradient in Europe. These traits form the basis of the ‘leaf-height-seed’ (LHS) plant ecology strategy scheme and have been shown to correlate with a vast number of other functional and demographic traits, and with ecosystem processes [44, 45]. We evaluated the effect of abiotic stress by quantifying climatic variables (temperature and precipitation), soil conditions (soil N, soil P and pH) and the community level stress-signature [46, 47]. The importance of biotic factors was assessed using species richness and functional diversity as proxies for niche packing, and the mean abundance weighted, community level ‘functional competition signature’ based on the C-S-R plant functional type system as a proxy for asymmetrical competition strength [46, 47].

Our three study species were specifically selected for their contrasting colonization rates, to allow the assessment of colonization rate effect on ITVBI patterns. The study species were the slow colonizing Anemone nemorosa, a species with intermediate colonization rates, Milium effusum, and the fast colonizing, for the region invasive alien Impatiens glandulifera. Using this set-up, we aim at addressing the following research questions:

  1. 1.

    Does high abiotic stress result in decreased or increased ITVBI for the different traits (stress-reduced plasticity vs. stress-induced variability hypothesis) resulting in either linear or quadratic relationships between ITVBI and abiotic predictors? Similarly, does high biotic competition (niche density and/or asymmetrical competition) lead to decreased or increased ITVBI for the different traits (niche packing vs. individual variation hypothesis) resulting in linear relationships between ITVBI and biotic predictors?

  2. 2.

    Is the ITVBI of the growth related trait plant height more affected by competition and the ITVBI of the stress response related trait SLA more by abiotic factors, compared to the other traits?

  3. 3.

    Does the non-native, fast colonizer I. glandulifera exhibit larger ITVBI than the slower colonizing native A. nemorosa and M. effusum, and do biotic and abiotic drivers explain a larger proportion of the variation in ITVBI for I. glandulifera than for the A. nemorosa and M. effusum, following the expected patterns for fast colonizing species?

Methods

Study species

In this study we quantified ITVBI for three herbaceous species across Europe, namely A. nemorosa, M. effusum and I. glandulifera. Anemone nemorosa L. (Ranunculaceae) is a widespread European spring ephemeral forest perennial (geophyte). The species flowers in spring, is insect-pollinated and also reproduces vegetatively through rhizomes. Individual ramets produce ten to 30 seeds per year, which are adapted to myrmecochory, resulting in a relatively low colonization rate [22, 48]. Milium effusum L. (Poaceae) is also a widespread European forest understory perennial, but differs from A. nemorosa by a hemicryptophytic life form, a later flowering period (early summer), adaptation to wind pollination, higher seed production (100–300 per individual per year), predominantly barochorous and epizoochorous dispersal, and a limited ability of vegetative reproduction through short stolons [17, 22, 49]. All these traits suggest an intermediate colonization rate for M. effusum. Impatiens glandulifera Royle (Balsaminaceae) was introduced to Europe in the 1800s from its native range in western Himalaya and subsequently became strongly invasive [50]. The species is a competitive annual of up to 2.5 m high. In its invaded range it mainly grows in riparian habitat, but has also been found to spread along road verges in more northern locations. The species produces a large number of insect-pollinated flowers in late summer, followed by up to 2500 seeds per individual per year. Reproduction is fully dependent on seed germination, with dispersal facilitated both by fruit explosion (ballistochory) and water (hydrochory) [50, 51], making the species a very fast colonizer into unoccupied habitats. All three species predominantly occur in shaded (wooded or wood-edge) habitats in Europe, which are characterized by buffered temperatures, high air humidity and high soil moisture [48,49,50]. For this reason, we expect these species to mainly experience stress from high temperatures and low water availability (i.e. rainfall) along the studied gradient.

Study area

Populations were sampled in seven (A. nemorosa and M. effusum) and six (I. glandulifera) regions along a 2300 km macroecological gradient, ranging from Amiens, France (49.90°N, 2.30°E) in the south to Abisko, Sweden (68.35°N, 18.83°E) in the north (Table 1). In each selected region, up to six populations were randomly selected within a 20 × 20 km area, resulting in a total of 37, 39 and 34 populations for A. nemorosa, M. effusum and I. glandulifera, respectively (Table 1). All populations of A. nemorosa and M. effusum were sampled in 2008 and occurred in ancient deciduous forests (see [22] for more details). The populations of I. glandulifera were sampled in 2011 and occurred along river banks and road verges, within (recent) forests or along forest edges (see [52] for more details). No field permissions were necessary for the collection of the plant samples in this study. Plant material was formally identified by JB, SAOC, GD, AK, PDF and BJG is their respective study regions.

Table 1 Study region overview with average intraspecific trait variation levels
Full size table

Functional traits of the study species

Leaf area, plant height and seed mass were measured for all three studied species, while specific leaf area (SLA) was not measured for I. glandulifera. For A. nemorosa and M. effusum, plant height was measured and all seeds and leaves of 15 randomly selected individual plants per population were collected within a 25 × 25 m area for both species at seed maturity. Seed mass was defined as the mean air-dry seed mass per seed of all seeds per individual for A. nemorosa and mean air-dry seed mass per seed of 50 randomly selected seeds per individual for M. effusum. Plant height, seed mass and SLA measurements were performed according to standardized protocols [22, 53]. Leaf area of the flattened leaves was measured using a Li-Cor Portable Area Meter Li-3000 (Li-Cor Biosciences, NE, USA) after 72 h air-drying (50 °C) [22]. SLA was hence calculated using dry leaf area due to logistic constraints. We nonetheless expect this ‘shrinkage bias’ to be constant across individuals within a species, hence justifying its use for ITV comparisons [22].

For I. glandulifera, plant height was measured and ten capsules (fruits) were collected for 30 randomly selected individual plants per population within a 25 × 25 m area. Seed mass was calculated as the mean dry seed mass of all seeds of ten capsules per individual. Leaf area of I. glandulifera was estimated as the product of leaf length and leaf width of the largest leaf of each individual plant. This has previously been shown to be a reasonable proxy for leaf area [17]. In all species, only healthy, non-damaged individuals were selected for functional trait measurements. Within-population functional trait variation (ITVBI) was subsequently quantified by calculating both the standard deviation (SD) and the (dimensionless) coefficient of variation (CV) for each trait per population [7, 17].

Abiotic variables

Mean annual precipitation (MAP) for the period 1982–2002 was retrieved from the closest city (all < 40 km from the sampled populations) to each sampling region from the Climate-Data.org model [54] (cities given in Table 1) (cf. [55]). As a temperature proxy we used the number of growing degree hours (GDH) above 5 °C between 1st January and the day of trait collection, rather than mean annual temperature, since GDH is considered to be more tightly related to plant development [55]. GDH was calculated as defined in Lindsey and Newman [56] based on observed daily minimum and maximum temperatures during the sampling year, which was 2008 for A. nemorosa and M. effusum and 2011 for I. glandulifera, obtained for the closest city to each sampling region from NOAA [57].

For each A. nemorosa and M. effusum population, five 4-cm deep soil cores were collected and pooled to measure soil pH (determined from a solution of 10 g of soil and 25 ml of 0.01 M CaCl2 with a standard glass electrode), plant available phosphorus (P) (extraction with ammonium lactate and photometric determination by flow injection analysis and flame atomic absorption spectrophotometry, SpectrAA-220, Varian) and total soil nitrogen (N) (elemental analyzer; %) [22]. For each I. glandulifera population, plant available soil nitrogen (nitrate and ammonium), rather than total soil N was measured during 6 weeks of the growing season (mid-June until the end of July 2011) using four anion and cation plant root simulator (PRS) soil probes (Western Ag Innovations Inc., Saskatoon, Canada). After retrieval from the field, the probes were shipped to the analytical laboratory of Western Ag Innovations for analysis. There, inorganic nitrogen levels (NH4 + and NO3 −) were quantified by colorimetry using a Technicon Autoanalyzer II. The NO3 − analysis was slightly modified through the addition of NaOH to the NH4Cl reagent in order to neutralize the sample solution before its entry into the Cd-reduction column (Western Ag Labs 2003).

Biotic variables

Community composition was determined for each population of the three study species by recording the cover-abundance (% cover) of all (understory) vascular plant species present within the 25 × 25 m area. For functional diversity calculations, we selected ten functional plant traits related to species growth, survival and reproduction, namely plant height, life span, rosette type, leaf area, specific leaf area, leaf dry matter content, age of first flowering, reproductive type, seed number and seed mass (Additional file 1). Trait attributes for all co-occurring species were extracted from several online databases, with an average data availability of 92% (Additional file 1) [58, 59]. Plant height, leaf area, seed number and seed mass of co-occurring species were logarithmically transformed to reduce the influence of extreme values in subsequent functional diversity calculations.

The resulting plot × species (111 × 167) and species × trait (167 × 10) matrices were used to calculate species richness and three measures of functional diversity for the plant communities present at each population of the three study species, namely functional richness, functional evenness and functional divergence [60] with the ‘FD’ package in R version 3.2.4 [61]. These measures addressed the three independent components of functional diversity, respectively; the amount of filled trait space, the regularity of the distribution of abundance in the trait space, and the spread or divergence in the distribution of abundance within the trait space [39]. Note that for each given study site, we did not include the respective study species (A. nemorosa, M. effusum or I. glandulifera), nor its functional trait attributes in the species richness and functional diversity calculations. Prior to functional diversity calculations, all traits were z-transformed and missing trait values were replaced using multivariate imputation with chain equations (MICE [62]), using the ‘mice’ R package [63]. This method uses predictive mean matching with five imputations, with the average of these five imputed value used to replace the missing value and has been shown to produce superior results compared to other imputation methods for database plant traits [64].

Additionally, we calculated the mean abundance weighted ‘functional competition signature’ (C-sign.) and ‘functional stress signature’ (S-sign.) based on the C-S-R plant functional type system [46], following the method of Hunt et al. [47]. The C-S-R functional type of each species was collected from Hunt et al. [47] and the BiolFlor database [58]. Since this C-signature is based on the present species communities, we believe that they are adequate indicators of the overall community-level strength of the asymmetrical competition for resources such as light at these locations. Similarly, we believe that the S-signature presents a proxy of the extent of (abiotic) stress experienced by each community. The full dataset is included in Additional file 2.

Statistical analysis

First, we performed Spearman rank correlations between latitude on the one hand, and all abiotic (MAP, GDH, soil N, soil P, pH, S-sign.) and biotic predictor (species richness, C-sign. and functional richness, evenness and divergence) on the other hand. Secondly, we explored the correlation between both measures of ITVBI (SD and CV) on the one hand, and population-level mean trait values, on the other hand, using linear mixed-effect models with study region as a random intercept factor. This random intercept was included to account for the dependence of populations within each region. As expected, SD was strongly correlated with trait means for most traits, justifying the use of CV as a trait mean-independent proxy for ITVBI (Additional file 3). Next, we assessed correlations between the ITVBI measures of all traits, for each study species separately, using Spearman rank correlations.

We assessed the effects of the measured abiotic and biotic factors along the gradient on between individual functional trait variation per population with linear mixed-effect models. Note that for these models we used trait CV as a measure of ITVBI for all traits. An individual mixed-effect model was constructed for each population level functional ITVBI as response variable, for each study species separately, with study region as a random intercept factor and all abiotic, stress-related (MAP, GDH, soil N, soil P, pH, S-sign.) and biotic/competition-related (species richness, C-sign. and functional richness, evenness and divergence) predictors as fixed effects (predictors). Quadratic terms for the stress-related predictors (MAP, GDH, soil N, soil P, pH, S-sign.) were also included as fixed effects. Since the optimal growing conditions (lowest stress) can potentially occur at intermediate levels of the assessed abiotic variables, quadratic relationships could be expected under both the stress-reduced plasticity hypothesis (convex parabolic relation) and stress-induced variability hypothesis (concave parabolic relation). Note that quadratic relationships are not expected for the biotic predictors since competition is expected to vary unidirectionally with the assessed competition-related biotic predictors.

Models were run after scaling (standardizing) all variables, as such allowing parameter (beta coefficient) comparisons. Final models were obtained using stepwise model reduction based on the Akaike’s information criterion (AIC) for models based on ML estimations (Table 1). All final model output was based on REML estimations. Marginal and conditional R2-coefficients, respectively describing the proportion of variance explained by the fixed factors and the variance explained by both the fixed and random factors, were calculated for each final model (cf. [65]). Several response variables and predictors were transformed (prior to standardization) to satisfy both the normal distribution of residuals and independence of residuals model assumptions (Table 2). Before the linear mixed-effect model analyses, all predictors were checked for multicollinearity using variation inflation factors (VIF) for each study species separately, but no collinearity problems were observed with a conservative VIF threshold of five. All linear mixed-effect models were performed with the ‘nlme’ package [66] in R 3.3.3., all other statistical analyses were performed in SPSS 21.0 (SPSS Inc., Chicago, IL, US).

Table 2 Parameter estimates of the performed reduced linear mixed-effect models for each functional trait ITVBI measure separately
Full size table

Results

The extent of between individual intraspecific trait variation (ITVBI) was variable across traits, with on average the lowest variability for SLA [across species mean CV = 11.7 ± 4.0% (SD)], followed by plant height (mean CV = 15.6 ± 5.3%) and seed mass (mean CV = 21.6 ± 6.1%), and the highest variability for leaf area (mean CV = 31.5 ± 9.7%). Interestingly, the relative ranking of trait variabilities (CV) was consistent across all three study species (Table 1). When comparing the ITVBI (CV) for the different traits across species, it was not consistently the same study species exhibiting the highest ITVBI, with the highest mean variability in plant height and leaf area for I. glandulifera, the highest mean variability in SLA for M. effusum and the highest mean variability in seed mass for A. nemorosa (Table 1). ITVBI was positively correlated among all traits for A. nemorosa, except between SLA and leaf area and between SLA and seed mass. For M. effusum positive correlations only occurred between seed mass, SLA and leaf area ITVBI, while for I. glandulifera, only leaf area ITVBI correlated positively with all other traits (Additional file 4).

Several abiotic and biotic variables were significantly related to latitude. Both temperature and precipitation decreased with latitude, while species richness (for M. effusum), functional diversity (for A. nemorosa and M. effusum) and the functional competition signature (for M. effusum) increased with latitude (Additional file 5). Soil variables were mainly unrelated to latitude, with only for A. nemorosa a significant increase in soil nitrogen with latitude (Additional file 5).

Climatic variables significantly affected ITVBI. Mean annual precipitation (MAP) during 1982–2002 was negatively correlated with the CV of SLA of A. nemorosa and the CV of seed mass and SLA of M. effusum. MAP furthermore showed a convex parabolic relation with the CV’s of both plant height and leaf area of I. glandulifera (Table 2, Fig. 1a). Temperature (Growing degree hours, GDH) was positively correlated with the CV’s of plant height and seed mass of M. effusum and the CV of plant height of I. glandulifera. GDH also showed a convex parabolic relation with the CV of plant height of A. nemorosa (Table 2). Concerning the effects of the soil, we observed a positive correlation between soil nitrogen and the CV of plant height of A. nemorosa and a convex parabolic relation with the CV’s of both seed mass and SLA of M. effusum. We did also observe a clear response to increased stress levels, as quantified through the S-signature, with a positive correlation with the CV of SLA of A. nemorosa, a negative correlation with the CV’s of both plant height and leaf area of I. glandulifera and a convex parabolic relation with the CV of SLA of M. effusum (Table 2, Fig. 1b).

Fig. 1
figure1

Relationships between intraspecific trait variation (cf. ITVBI) and several abiotic and biotic factors. a Relationship between mean annual precipitation (MAP) and the log-transformed coefficient of variation (CV) of leaf area for Impatiens glandulifera, b relationship between stress-signature and the square-root-transformed CV of specific leaf area (SLA) for Milium effusum, c relationship between species richness and the CV of plant height for Anemone nemorosa, d relationship between functional evenness and the log-transformed CV of seed mass for Impatiens glandulifera. Regression lines present statistically significant linear relationships

Full size image

Several biotic factors also affected ITVBI. First, plant height variability in A. nemorosa was directly negatively affected by the increased niche density associated with increased species richness (significant species richness effect, Fig. 1c), while A. nemorosa’s SLA variability was negatively affected by increasing functional richness (Table 2). Second, functional evenness was negatively correlated with the CV’s of seed mass and SLA of A. nemorosa and positively correlated with the CV’s of both plant height and seed mass of I. glandulifera (Table 2, Fig. 1d). Functional divergence was furthermore negatively related to the CV of plant height of A. nemorosa and the CV of SLA of M. effusum and positively related to the CV of leaf area of I. glandulifera (Table 2). Finally, the proxy for vegetation-level asymmetrical competition strength (C-signature) was positively related to variation in both plant height and seed mass for A. nemorosa (Table 2).

Discussion

Stress and competition effects

Although responses tended to be mainly species-specific, we observed both decreased (stress-reduced plasticity hypothesis, [32]) and increased (stress-induced variability hypothesis, [34]) ITVBI under (abiotically) stressful conditions, although climatic stress (high temperature and low precipitation) nevertheless consistently increased ITVBI. Similarly, we observed both reduced (niche packing hypothesis, [36]) and enhanced (individual variation hypothesis, [10]) ITVBI under high community-level niche density.

Interestingly, both drought stress (low precipitation) and (niche) competition (functional diversity, functional competition signature) increased with latitude, although heat stress (high GDH) decreased with latitude. This was unexpected, since theory predicts competition and abiotic stress to trade-off across latitudes [33, 42]. This suggests that these plants are occurring in milder microclimates in climatically harsher locations, as such allowing them to circumvent more extreme macroclimatic stress conditions. This likely results in the occurrence of these species in more competitive, mesic communities in desirable, mild microsites in the harsher (drier) north [67]. This could furthermore explain the absence of a clear correlation between the biotically inferred mean community-level stress level (S-sign.) and climatic factors. These results further emphasize the importance of assessing both abiotic and biotic factors along macroecological gradients to get a full picture of potential drivers of ITVBI, since both are not always predictably related [30, 33].

Climate (temperature and precipitation) significantly affected ITVBI for several traits in each study species. The direction of these correlations were furthermore consistent across species and traits, with either reduction of trait variation at low temperatures (GDH) and high precipitation (MAP), or trait variation reduction at intermediate temperatures/precipitation levels (climatically favourable locations). These results strongly suggest the prevalence of the stress-induced variability hypothesis. Indeed, considering that these species predominantly occur in relatively cool (shaded) locations with high humidity and soil moisture [48,49,50], we expect climatic stress to be strongest at high GDH and low MAP levels. Although note that the observed convex parabolic relationships indicate that for A. nemorosa and I. glandulifera the climatically favourable locations seem to occur at intermediate levels of GDH and MAP, respectively. One soil variable (nitrogen) had a similar effect for several traits of M. effusum, with a reduction in ITVBI at intermediate soil nitrogen levels, again indicative of the stress-induced variability hypothesis. This pattern was also mirrored in the biotically inferred stress parameter (S-sign.), which is usually indicative of soil nutrient (stress) levels. For A. nemorosa and I. glandulifera however, soil nutrient-related stress seemed to mainly support the stress-reduced plasticity hypothesis, with lowest ITVBI at low soil nitrogen levels or high S-sign. levels, These results are largely in agreement with previous studies of nutrient effects on ITVBI [17, 29]. Overall, our results support the recent findings of inconsistent effects of environmental stress on ITVBI across traits and species [17, 30], and suggest that climatic factors and nutrient stress can differently affect ITVBI levels.

Concerning the proxies for niche-based effects on ITVBI, we can assume that communities with high species/functional richness and high functional evenness have high niche density. High functional divergence is usually expected to be a result of strong niche filtering during community assembly, and thus indicates the presence of strong niche-based effects within the community [68]. Our results show that for A. nemorosa and M. effusum, ITVBI is consistently reduced under high niche density. This is in agreement with the results of the meta-analysis of Siefert et al. [8], which looked at the effect of species richness on community-level ITVBI. The ITVBI of I. glandulifera, however, is enhanced under the same conditions, a pattern that has previously also been observed at the community-level for limestone grasslands on Öland (Sweden) [37]. This suggests that the niche packing hypothesis seemingly drives trait variation in the native A. nemorosa and M. effusum (cf. [8, 33, 36]), but that the individual variation hypothesis applies to the ITVBI of the invasive I. glandulifera (cf. [10, 36]). Additionally, community-level asymmetrical competition strength (expressed by the C-signature) led to increased ITVBI for two traits in A. nemorosa. Hence, contrary to our expectations, asymmetrical competition may trigger increased trait variation [28], while niche density reduced this variation for the same traits in the same species. Interestingly, Burns and Strauss [40] observed a similar increase in trait variation under increased interspecific (hierarchical) competition in an experimental setup. It has been argued that this trait variation may be an adaptive response to asymmetrical competition [17, 40].

We did not find clear support for our hypotheses that ITVBI of growth related traits, such as plant height, are more related to competition, or that ITVBI in stress-related traits, such as SLA, is best explained by abiotic variables. Indeed, ITVBI in both traits was affected by a combination of abiotic and biotic drivers. These patterns might be the result of linkage between several functional traits, rendering them non-independent (Additional file 4) [44]. Additionally, neither plant height nor SLA are likely to be solely related to either stress-related abiotic or competition-related biotic responses. Indeed, although plant height is often linked to competition strategies in plants, both light- and nutrient stress are known to affect plant height as well [8, 44]. Similarly, although SLA is often linked to stress responses in plants, it is also known to be affected by reduced light levels due to competition effects [3, 44].

Colonization rate

We partly found support for the expected higher ITVBI in species with faster colonization rates [43]: overall ITVBI (CV) values for the vegetative traits plant height and leaf area in the fast colonizing, invasive I. glandulifera were higher than those in the slow colonizing A. nemorosa and intermediate colonizer M. effusum. Unexpectedly, leaf area ITVBI was lowest for M. effusum, and not for A. nemorosa. This pattern of higher ITVBI for the fast colonizer was only apparent for vegetative traits, however, with a reversed pattern for the measured reproductive trait (seed mass). Since both A. nemorosa and M. effusum occur in relatively stable ancient forests, these slowly colonizing species have most likely experienced consistent, long term environmental selection on their functional trait set, thus resulting in more optimal mean trait attributes and reduced ITVBI [17, 22]. The invasive I. glandulifera, on the other hand, mainly occurs in more disturbed recent forests and forest edges, where these young, dynamic populations have been exposed to shorter periods of much less consistent, fluctuating environmental selection, thus explaining the occurrence of higher ITVBI [50, 51]. Alternatively, species with greater dispersal ability are likely exposed to a greater range of environmental variability, possibly leading to selection for increased phenotypic plasticity [43]. These effects are likely aggravated by differences in plant height between the three species. Indeed, while both A. nemorosa and M. effusum are short understory species buffered against environmental fluctuation by larger understory species, I. glandulifera is not buffered by these effects, since it usually is the dominant ‘canopy species’ of the forest understory. The reduced ITVBI variation in seed mass for I. glandulifera might be the result of stronger selection on seed traits for this species, compared to the partially clonal, perennial A. nemorosa and M. effusum. Indeed, since this species is annual, population persistence is fully dependent on annual seedling establishment.

There was no consistently stronger effect of either abiotic or biotic drivers on trait ITVBI of the fast colonizing I. glandulifera. We rather observed species-specific responses of the different abiotic and biotic drivers, making overall generalizations difficult. However, as previously mentioned, the native, slower colonizing species A. nemorosa and M. effusum were characterized by a reduction in ITVBI under increased niche density, as opposed to the exotic, fast colonizing I. glandulifera, which showed an increase in trait variation under high niche density. Such a potential for large trait variation under high competition intensity, through either plastic responses or genetic adaptation, might explain the invasive success of the non-native I. glandulifera in Europe [43].

Conclusions

We found clear effects of both stress-related abiotic drivers and competition-related biotic drivers on within-population intraspecific functional trait variation. Our results suggest that stress can both reduce and increase ITVBI, seemingly supporting both the stress-reduced plasticity and stress-induced variability hypotheses (cf. [32, 34]). Climatic stress consistently increased ITVBI across species and traits. Soil nutrient stress, on the other hand, reduced ITVBI for A. nemorosa and I. glandulifera, but not for M. effusum. Similarly, niche packing effects on ITVBI were both negative and positive, respectively supporting the niche packing hypothesis [36] and the individual variation hypothesis [10]. This clearly illustrates the importance of simultaneously evaluating both abiotic and biotic factors on ITVBI [30, 33].

In sum, ITVBI for several (vegetative) traits tended to be highest in the non-native, fast colonizing species. Second, we observed a reversed effect of niche density (functional diversity) on ITVBI for the fast colonizing invasive Impatiens glandulifera, as compared to the slow(er) colonizing native Anemone nemorosa and Milium effusum. Finally, our study adds to the growing realization that within-population trait variation should not be ignored in ecological studies and can even provide valuable ecological insights [4, 8, 12, 19, 22, 25].

Abbreviations

C-sign.:

functional competition signature

CV:

coefficient of variation

GDH:

growing degree hours

ITV:

intraspecific trait variation

ITVBI :

between individual intraspecific trait variation

MAP:

mean annual precipitation

S-sign.:

functional stress signature

SD:

standard deviation

SLA:

specific leaf area

soil N:

soil nitrogen

soil P:

soil phosphorous

References

  1. 1.

    Violle C, Navas MML, Vile D, Kazakou E, Fortunel C, Hummel I, et al. Let the concept of trait be functional! Oikos. 2007;116:882–93.

    Article  Google Scholar 

  2. 2.

    Adler PB, Salguero-Gómez R, Compagnoni A, Hsu JS, Ray-mukherjee J, Adler PB, et al. Functional traits explain variation in plant life history strategies. Proc Natl Acad Sci. 2014;111:740–5.

    CAS  Article  PubMed  Google Scholar 

  3. 3.

    Wright IJ, Reich PB, Cornelissen JHC, Falster DS, Groom PK, Hikosaka K, et al. Modulation of leaf economic traits and trait relationships by climate. Glob Ecol Biogeogr. 2005;14:411–21.

    Article  Google Scholar 

  4. 4.

    Albert CH, Grassein F, Schurr FM, Vieilledent G, Violle C. When and how should intraspecific variability be considered in trait-based plant ecology? Perspect Plant Ecol Evol Syst. 2011;13:217–25.

    Article  Google Scholar 

  5. 5.

    Messier J, McGill BJ, Lechowicz MJ. How do traits vary across ecological scales? A case for trait-based ecology. Ecol Lett. 2010;13:838–48.

    Article  PubMed  Google Scholar 

  6. 6.

    Kattge J, Díaz S, Lavorel S, Prentice IC, Leadley P, Bönisch G, et al. TRY—a global database of plant traits. Glob Chang Biol. 2011;17:2905–35.

    Article  PubMed Central  Google Scholar 

  7. 7.

    Mitchell RM, Bakker JD. Quantifying and comparing intraspecific functional trait variability: a case study with Hypochaeris radicata. Funct Ecol. 2014;28:258–69.

    Article  Google Scholar 

  8. 8.

    Siefert A, Violle C, Chalmandrier L, Cécile HA, Taudiere A, Fajardo A, et al. A global meta-analysis of the relative extent of intraspecific trait variation in plant communities. Ecol Lett. 2015;18:1406–19.

    Article  PubMed  Google Scholar 

  9. 9.

    Hart SP, Schreiber SJ, Levine JM, Coulson T. How variation between individuals affects species coexistence. Ecol Lett. 2016;19:825–38.

    Article  PubMed  Google Scholar 

  10. 10.

    Clark JS. Individuals and the variation needed for high species diversity in forest trees. Science. 2010;327:1129–32.

    CAS  Article  PubMed  Google Scholar 

  11. 11.

    Jung V, Violle C, Mondy C, Hoffmann L, Muller S. Intraspecific variability and trait-based community assembly. J Ecol. 2010;98:1134–40.

    Article  Google Scholar 

  12. 12.

    Chalmandrier L, Münkemüller T, Colace M-P, Renaud J, Aubert S, Carlson BZ, et al. Spatial scale and intraspecific trait variability mediate assembly rules in alpine grasslands. J Ecol. 2017;105:277–87.

    Article  Google Scholar 

  13. 13.

    Hooper DU, Chapin FS III, Ewel JJ, Hector A, Inchausti P, Lavorel S, et al. Effects of biodiversity on ecosystem functioning: a consensus of current knowledge. Ecol Monogr. 2005;75:3–35.

    Article  Google Scholar 

  14. 14.

    Lecerf A, Chauvet E. Intraspecific variability in leaf traits strongly affects alder leaf decomposition in a stream. Basic Appl Ecol. 2008;9:598–605.

    Article  Google Scholar 

  15. 15.

    Garrett KA, Zúñiga LN, Roncal E, Forbes GA, Mundt CC, Su Z, Nelson RJ. Intraspecific functional diversity in hosts and its effect on disease risk across a climatic gradient. Ecol Appl. 2009;19:1868–83.

    CAS  Article  PubMed  Google Scholar 

  16. 16.

    Hof C, Levinsky I, Araújo MB, Rahbek C. Rethinking species’ ability to cope with rapid climate change. Glob Chang Biol. 2011;17:2987–90.

    Article  Google Scholar 

  17. 17.

    Lemke IH, Kolb A, Graae BJ, De Frenne P, Acharya KP, Blandino C, et al. Patterns of phenotypic trait variation in two temperate forest herbs along a broad climatic gradient. Plant Ecol. 2015;216:1523–36.

    Article  Google Scholar 

  18. 18.

    Etterson JR, Schneider HE, Gorden NLS, Weber JJ. Evolutionary insights from studies of geographic variation: contemporary variation and looking to the future. Am J Bot. 2016;103:5–9.

    Article  PubMed  Google Scholar 

  19. 19.

    Bolnick DI, Amarasekare P, Araújo MS, Bürger R, Levine JM, Novak M, et al. Why intraspecific trait variation matters in community ecology. Trends Ecol Evol. 2011;26:183–92.

    Article  PubMed  PubMed Central  Google Scholar 

  20. 20.

    Albert CH, Thuiller W, Yoccoz NG, Soudant A, Boucher F, Saccone P, et al. Intraspecific functional variability: extent, structure and sources of variation. J Ecol. 2010;98:604–13.

    Article  Google Scholar 

  21. 21.

    Fajardo A, Piper FI. Intraspecific trait variation and covariation in a widespread tree species (Nothofagus pumilio) in southern Chile. New Phytol. 2011;189:259–71.

    Article  PubMed  Google Scholar 

  22. 22.

    De Frenne P, Graae BJ, Kolb A, Shevtsova A, Baeten L, Brunet J, et al. An intraspecific application of the leaf-height-seed ecology strategy scheme to forest herbs along a latitudinal gradient. Ecography. 2011;34:132–40.

    Article  Google Scholar 

  23. 23.

    Poorter H, Niinemets U, Poorter L, Wright IJ, Villar R. Causes and consequences of variation in leaf mass per area (LMA): a meta-analysis. New Phytol. 2009;182:565–88.

    Article  PubMed  Google Scholar 

  24. 24.

    Poorter H, Niinemets Ü, Walter A, Fiorani F, Schurr U. A method to construct dose-response curves for a wide range of environmental factors and plant traits by means of a meta-analysis of phenotypic data. J Exp Bot. 2010;61:2043–55.

    CAS  Article  PubMed  Google Scholar 

  25. 25.

    Auger S, Shipley B. Inter-specific and intra-specific trait variation along short environmental gradients in an old-growth temperate forest. J Veg Sci. 2013;24:419–28.

    Article  Google Scholar 

  26. 26.

    Albert CH, Thuiller W, Yoccoz NG, Douzet R, Aubert S, Lavorel S. A multi-trait approach reveals the structure and the relative importance of intra- vs. interspecific variability in plant traits. Funct Ecol. 2010;24:1192–201.

    Article  Google Scholar 

  27. 27.

    Carlucci MB, Debastiani VJ, Pillar VD, Duarte LDS. Between- and within-species trait variability and the assembly of sapling communities in forest patches. J Veg Sci. 2015;26:21–31.

    Article  Google Scholar 

  28. 28.

    Mason NWH, De Bello F, Doležal J, Lepš J. Niche overlap reveals the effects of competition, disturbance and contrasting assembly processes in experimental grassland communities. J Ecol. 2011;99:788–96.

    Article  Google Scholar 

  29. 29.

    Lemke IH, Kolb A, Diekmann MR. Region and site conditions affect phenotypic trait variation in five forest herbs. Acta Oecol. 2012;39:18–24.

    Article  Google Scholar 

  30. 30.

    Le Bagousse-Pinguet Y, Börger L, Quero J, Garcia-Gomez M, Soriano S, Maestre FT, et al. Traits of neighbouring plants and space limitation determine intraspecific trait variability in semi-arid shrublands. J Ecol. 2015;103:1647–57.

    Article  Google Scholar 

  31. 31.

    Herz K, Dietz S, Haider S, Jandt U, Scheel D, Bruelheide H. Drivers of intraspecific trait variation of grass and forb species in German meadows and pastures. J Veg Sci. 2017;28:705–16.

    Article  Google Scholar 

  32. 32.

    Valladares F, Gianoli E, Gómez JM. Ecological limits to plant phenotypic plasticity. New Phytol. 2007;176:749–63.

    Article  PubMed  Google Scholar 

  33. 33.

    Hulshof CM, Violle C, Spasojevic MJ, Mcgill B, Damschen E, Harrison S, et al. Intra-specific and inter-specific variation in specific leaf area reveal the importance of abiotic and biotic drivers of species diversity across elevation and latitude. J Veg Sci. 2013;24:921–31.

    Article  Google Scholar 

  34. 34.

    Hoffmann AA, Merilä J. Heritable variation and evolution under favourable and unfavourable conditions. Trends Ecol Evol. 1999;14:96–101.

    CAS  Article  PubMed  Google Scholar 

  35. 35.

    Rey O, Danchin E, Mirouze M, Loot C, Blanchet S. Adaptation to global change: a transposable element-epigenetics perspective. Trends Ecol Evol. 2016;31:514–26.

    Article  PubMed  Google Scholar 

  36. 36.

    Violle C, Enquist BJ, McGill BJ, Jiang L, Albert CH, Hulshof C, et al. The return of the variance: intraspecific variability in community ecology. Trends Ecol Evol. 2012;27:244–52.

    Article  PubMed  Google Scholar 

  37. 37.

    Le Bagousse-Pinguet Y, De Bello F, Vandewalle M, Leps J, Sykes MT. Species richness of limestone grasslands increases with trait overlap: evidence from within- and between-species functional diversity partitioning. J Ecol. 2014;102:466–74.

    Article  Google Scholar 

  38. 38.

    Cadotte MW, Carscadden K, Mirotchnick N. Beyond species: functional diversity and the maintenance of ecological processes and services. J Appl Ecol. 2011;48:1079–87.

    Article  Google Scholar 

  39. 39.

    Mason NWH, Mouillot D, Lee WG, Wilson JB. Functional richness, functional evenness and functional divergence: the primary components of functional diversity. Oikos. 2005;111:112–8.

    Article  Google Scholar 

  40. 40.

    Burns J, Strauss S. Effects of competition on phylogenetic signal and phenotypic plasticity in plant functional traits. Ecology. 2012;93:126–37.

    Article  Google Scholar 

  41. 41.

    Bennett JA, Riibak K, Tamme R, Lewis RJ, Pärtel M. The reciprocal relationship between competition and intraspecific trait variation. J Ecol. 2016;104:1410–20.

    Article  Google Scholar 

  42. 42.

    Bertness MD, Callaway R. Positive interactions in communities. Trends Ecol Evol. 1994;9:187–91.

    Article  Google Scholar 

  43. 43.

    Davidson AM, Jennions M, Nicotra AB. Do invasive species show higher phenotypic plasticity than native species and if so, is it adaptive? A meta-analysis. Ecol Lett. 2011;14:419–31.

    Article  PubMed  Google Scholar 

  44. 44.

    Westoby M. A leaf-height-seed (LHS) plant ecology strategy scheme. Plant Soil. 1998;199:213–27.

    CAS  Article  Google Scholar 

  45. 45.

    Garnier E, Navas ML. A trait-based approach to comparative functional plant ecology: concepts, methods and applications for agroecology. A review. Agron Sustain Dev. 2012;32:365–99.

    Article  Google Scholar 

  46. 46.

    Grime J. Evidence for the existence of three primary strategies in plants and its relevance to ecological and evolutionary theory. Am Nat. 1977;111:1169–94.

    Article  Google Scholar 

  47. 47.

    Hunt R, Hodgson J, Thompson K, Bungener P, Dunnett N, Askew A, et al. A new practical tool for deriving a functional signature for herbaceous vegetation. Appl Veg Sci. 2004;7:163–70.

    Article  Google Scholar 

  48. 48.

    Shirreffs D. Biological flora of the British Isles. Anemone Nemorosa L. J Ecol. 1985;73:1005–20.

    Article  Google Scholar 

  49. 49.

    De Frenne P, Brunet J, Cougnon M, Decocq G, Graae BJ, Hagenblad J, et al. Biological flora of the British Isles: Milium effusum. J Ecol. 2017;105:839–58.

    Article  Google Scholar 

  50. 50.

    Beerling DJ, Perrins JM. Biological flora of the British Isles. Impatiens glandulifera Royle (Impatiens roylei Walp.). J Ecol. 1993;81:367–82.

    Article  Google Scholar 

  51. 51.

    Greenwood P, Kuhn NJ. Does the invasive plant, Impatiens glandulifera, promote soil erosion along the riparian zone? An investigation on a small watercourse in northwest Switzerland. J Soils Sediments. 2014;14:637–50.

    Article  Google Scholar 

  52. 52.

    Acharya KP. Invasive species: genetics, characteristics and trait variation along a latitudinal gradient. Ph.D. thesis, Norwegian University of Science and Technology (NTNU), Norway. 2014.

  53. 53.

    Cornelissen JHC, Lavorel S, Garnier E, Díaz S, Buchmann N, Gurvich DE, et al. A handbook of protocols for standardised and easy measurement of plant functional traits worldwide. Aust J Bot. 2003;51:335–80.

    Article  Google Scholar 

  54. 54.

    Climate-data.org. Mean annual temperatures and mean annual precipitations. 2014. http://climate-data.org/. Accessed 9 May 2016.

  55. 55.

    De Frenne P, Graae BJ, Kolb A, Brunet J, Chabrerie O, Cousins SAO, et al. Significant effects of temperature on the reproductive output of the forest herb Anemone nemorosa L. For Ecol Manage. 2010;259:809–17.

    Article  Google Scholar 

  56. 56.

    Lindsey A, Newman J. Use of official weather data in spring time: temperature analysis of an Indiana phenological record. Ecology. 1956;37:812–23.

    Article  Google Scholar 

  57. 57.

    NOAA. National centers for environmental information. National oceanic and atmospheric administration. 2016. https://www.ncdc.noaa.gov/. Accessed 14 Sep 2016.

  58. 58.

    Klotz S, Kühn I, Durka W. BIOLFLOR-Eine Datenbank zu biologisch-ökologischen Merkmalen der Gefäßpflanzen in Deutschland. Schriftenreihe für Vegetationskunde. 2002;38:1–333.

    Google Scholar 

  59. 59.

    Kleyer M, Bekker R, Knevel IC, Bakker JP, Thompson K, Sonnenschein M, et al. The LEDA Traitbase: a database of life-history traits of the Northwest European flora. J Ecol. 2008;96:1266–74.

    Article  Google Scholar 

  60. 60.

    Villéger S, Mason NWH, Mouillot D. New multidimensional functional diversity indices for a multifaceted framework in functional ecology. Ecology. 2008;89:2290–301.

    Article  PubMed  Google Scholar 

  61. 61.

    Laliberté E, Shipley B. FD: measuring functional diversity from multiple traits, and other tools for functional ecology. R package version 1.0-12. 2014.

  62. 62.

    Azur MJ, Stuart EA, Frangakis C, Leaf PJ. Multiple imputation by chained equations: what is it and how does it work? Int J Methods Psychiatr Res. 2011;20:40–9.

    Article  PubMed  PubMed Central  Google Scholar 

  63. 63.

    van Buuren S, Groothuis-Oudshoorn K. Mice: multivariate imputation by chained equations in R. J Stat Softw. 2011;45:3.

    Article  Google Scholar 

  64. 64.

    Taugourdeau S, Villerd J, Plantureux S, Huguenin-Elie O, Amiaud B. Filling the gap in functional trait databases: use of ecological hypotheses to replace missing data. Ecol Evol. 2014;4:944–58.

    Article  PubMed  PubMed Central  Google Scholar 

  65. 65.

    Nakagawa S, Schielzeth H. A general and simple method for obtaining R2 from generalized linear mixed-effects models. Methods Ecol Evol. 2013;4:133–42.

    Article  Google Scholar 

  66. 66.

    Pinheiro J, Bates D, DebRoy S, Sarkar D, R Core Team. nlme: linear and nonlinear mixed effects models. R package version 3.1-131. 2017. https://cran.r-project.org/package=nlme. Accessed 16 Jul 2017.

  67. 67.

    Wasof S, Lenoir J, Gallet-Moron E, Jamoneau A, Brunet J, Cousins SAO, et al. Ecological niche shifts of understorey plants along a latitudinal gradient of temperate forests in north-western Europe. Glob Ecol Biogeogr. 2013;22:1130–40.

    Article  Google Scholar 

  68. 68.

    Pillar VD, Duarte LDS, Sosinski EE, Joner F. Discriminating trait-convergence and trait-divergence assembly patterns in ecological community gradients. J Veg Sci. 2009;20:334–48.

    Article  Google Scholar 

Download references

Authors’ contributions

KH performed statistical analyses and drafted the manuscript. KPA, JB, SAOC, GD, AK, IHL, JP, KV, PDF and BJG performed, or assisted sample collection and trait measurements. All authors, including JL and MH participated in the design of the study and helped to draft the manuscript. All authors read and approved the final manuscript.

Acknowledgements

We thank Mattias Ahlström, Prof. Martin Diekmann, Dr. Jaan Liira and Emelie Waldén for assistance in the field and/or valuable feedback on the manuscript.

Competing interests

The authors declare that they have no competing interests.

Availability of data and materials

The dataset supporting the results of this article is available in Additional file 2.

Consent for publication

Not applicable.

Ethics approval and consent to participate

Not applicable.

Funding

This work was financially supported for the data collection by the Research Foundation-Flanders (FWO) through funding of the Scientific Research network ‘FLEUR’ (http://www.fleur.ugent.be); and for the data analysis and writing of the manuscript by NTNU, in the sustainability strategic research area 2014–2023 [81617824].

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Author information

Affiliations

  1. Department of Biology, Norwegian University of Science and Technology, Høgskoleringen 5, 7034, Trondheim, Norway

    Kenny Helsen, Kamal P. Acharya & Bente J. Graae

  2. Southern Swedish Forest Research Centre, Swedish University of Agricultural Sciences, Box 49, 230 53, Alnarp, Sweden

    Jörg Brunet

  3. Department of Physical Geography and Quaternary Geology, Stockholm University, 106 91, Stockholm, Sweden

    Sara A. O. Cousins & Jan Plue

  4. Edysan (FRE 3498 CNRS), Centre National de la Recherche Scientifique/Université de Picardie Jules Verne, 1 rue des Louvels, 80037, Amiens Cedex, France

    Guillaume Decocq & Jonathan Lenoir

  5. Division Forest, Nature and Landscape Research, Department Earth and Environmental Sciences, University of Leuven, Celestijnenlaan 200E, 3001, Heverlee, Belgium

    Martin Hermy

  6. Vegetation Ecology and Conservation Biology, Institute of Ecology, FB 02, University of Bremen, Leobener Strasse 5, 28359, Bremen, Germany

    Annette Kolb & Isgard H. Lemke

  7. Forest & Nature Lab, Ghent University, Geraardsbergsesteenweg 267, 9090, Gontrode-Melle, Belgium

    Kris Verheyen & Pieter De Frenne

  8. Department of Plant Production, Ghent University, Proefhoevestraat 22, 9090, Melle, Belgium

    Pieter De Frenne

Authors
  1. Kenny Helsen
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Kamal P. Acharya
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Jörg Brunet
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Sara A. O. Cousins
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Guillaume Decocq
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Martin Hermy
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Annette Kolb
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Isgard H. Lemke
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Jonathan Lenoir
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Jan Plue
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Kris Verheyen
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Pieter De Frenne
    View author publications

    You can also search for this author in PubMed Google Scholar

  13. Bente J. Graae
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Kenny Helsen.

Additional files

Additional file 1.

Overview of the selected functional traits for the functional diversity analysis. Description, scale, percentage of available data for all species and main data sources are given for each trait.

Additional file 2.

The full dataset used for this study. Study species, study region, population code, all trait SD’s and CV’s, latitude and all abiotic and biotic predictors given for each sampled population of Anemone nemorosa, Milium effusum and Impatiens glandulifera. C-signature = mean abundance weighted functional competition signature, CV = coefficient of variation, GDH = growing degree hours, MAP = mean annual precipitation, NA = not available, S-signature = mean abundance weighted functional stress signature, SD = standard deviation, soil N = soil nitrogen, soil P = soil phosphorous.

Additional file 3.

Parameter estimates for the separate linear mixed-effect models between each trait ITVBI (SD and CV) (response) and population-level trait mean (predictor). Test statistic (F) and P value (before slash), and beta-coefficient (after slash) given for each model. CV = coefficient of variation, SLA = specific leaf area, SD = standard deviation. Significance: (*): 0.10 ≥ P-value > 0.05 *: 0.05 ≥ P-value > 0.01; **: 0.01 ≥ P-value > 0.001; ***: 0.001 ≥ P-value. a = predictor square root transformed, b = predictor logarithmic transformed, c = response square root transformed, d = response logarithmic transformed.

Additional file 4.

Pairwise spearman rank correlations between all trait ITVBI, for each study species separately. ITVBI is quantified as coefficient of variation (CV) for all traits. Spearman rank correlation coefficients given for each test. SLA = specific leaf area. Significance: (*): 0.10 ≥ P-value > 0.05 *: 0.05 ≥ P-value > 0.01; **: 0.01 ≥ P-value > 0.001; ***: 0.001 ≥ P-value.

Additional file 5.

Spearman rank correlations between latitude and the different climatic, soil and local biotic variables, for each study species separately. Spearman rank correlation coefficients given for each test. C-sign. = mean abundance weighted functional competition signature, GDH = growing degree hours, MAP = mean annual precipitation, S-sign. = mean abundance weighted functional stress signature, soil N = soil nitrogen, soil P = soil phosphorous. Significance: (*): 0.10 ≥ P-value > 0.05 *: 0.05 ≥ P-value > 0.01; **: 0.01 ≥ P-value > 0.001; ***: 0.001 ≥ P-value.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Helsen, K., Acharya, K.P., Brunet, J. et al. Biotic and abiotic drivers of intraspecific trait variation within plant populations of three herbaceous plant species along a latitudinal gradient. BMC Ecol 17, 38 (2017). https://doi.org/10.1186/s12898-017-0151-y

Download citation

Keywords

  • Anemone nemorosa
  • Between-individual ITV
  • Herbaceous plant species
  • Impatiens glandulifera
  • Individual variation hypothesis
  • Intraspecific trait variation
  • Latitudinal gradient
  • Milium effusum
  • Niche packing
  • Phenotypic plasticity

BMC Ecology

ISSN: 1472-6785

Contact us