Plant Ecology & Diversity

ISSN: 1755-0874 (Print) 1755-1668 (Online) Journal homepage: http://www.tandfonline.com/loi/tped20

Changes in plant taxonomic and functional diversity patterns following treeline advances in the South Urals Antonio Gazol, Pavel Moiseev & J. Julio Camarero To cite this article: Antonio Gazol, Pavel Moiseev & J. Julio Camarero (2017) Changes in plant taxonomic and functional diversity patterns following treeline advances in the South Urals, Plant Ecology & Diversity, 10:4, 283-292, DOI: 10.1080/17550874.2017.1400126 To link to this article: https://doi.org/10.1080/17550874.2017.1400126

View supplementary material

Published online: 13 Nov 2017.

Submit your article to this journal

Article views: 41

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=tped20 Download by: [2.152.229.44]

Date: 14 December 2017, At: 07:56

Plant Ecology & Diversity, 2017 Vol. 10, No. 4, 283–292, https://doi.org/10.1080/17550874.2017.1400126

ARTICLE Changes in plant taxonomic and functional diversity patterns following treeline advances in the South Urals Antonio Gazol

*a, Pavel Moiseev

b

and J. Julio Camarero

a

a

Departamento de Biodiversidad y Restauración, Instituto Pirenaico de Ecología (IPE-CSIC), Zaragoza, Spain; bLaboratory of Dendrochronology, Institute of Plant and Animal Ecology, Yekaterinburg, Russia

Downloaded by [2.152.229.44] at 07:56 14 December 2017

(Received 18 April 2017; accepted 30 October 2017) Background: Treeline ecotones represent environmental boundaries that fluctuate in space and time and thus induce changes in plant taxonomic and functional diversity. Aims: To study changes through time in taxonomic and functional plant diversity patterns along the treeline ecotone. Methods: In 2002, vegetation was sampled along a gradient from upper montane forest to the treeline–alpine transition in the South Ural Mountains, Russia. In 2014, vegetation was resampled and plant functional traits were collected. We studied spatial and temporal changes in plant species composition, functional composition and functional diversity. Results: Species composition and diversity changed along the elevational gradient. The functional composition in height, leaf area, specific leaf area and leaf nitrogen content decreased with elevation, whereas functional composition of leaf carbon content increased. We found a temporal shift towards shorter plants with smaller leaves in treeline sites. Functional richness varied in several traits along the elevational gradient, while functional dispersion showed a trend towards increased functional dispersion in height, specific leaf area and leaf nitrogen in the treeline–tundra transition. Conclusions: Tree encroachment across the treeline ecotone has resulted in a shift in plant species relative abundances and functional diversity, possibly affecting plant community assembly patterns. Keywords: alpine ecology; elevational gradient; functional composition; mountain forest; null model; plant community

Introduction Mountain regions intersect important environmental boundaries, such as the treelines whose elevations are expected to fluctuate in response to shifts or abrupt changes in the prevailing climate conditions (Körner 2012). Environmental conditions, such as snow cover, air and soil temperatures, wind regimes, radiation intensity and carbon and nitrogen cycles change abruptly across treeline ecotones (Holtmeier and Broll 1992). These environmental conditions provide a great variety of microhabitats at small spatial scales that determine plant community composition from treeless alpine vegetation to boreal or upper montane forests (Camarero and Gutiérrez 2002; Lenoir et al. 2008; Körner 2012). Mountain regions have warmed considerably over the last 100 years, and this warming has accelerated since the 1980s in mountains located at latitudes higher than 40° N (Kohler et al. 2014). This warming has enhanced an upward treeline shift in some areas and has resulted in important changes in ecosystem processes and functioning (Grace et al. 2002; Camarero and Gutiérrez 2004; Devi et al. 2008; Hagedorn et al. 2014). Nevertheless, there is no consensus on the influence that tree encroachment across the treeline ecotone and upward treeline advances may have on plant diversity and vegetation patterns since alpine communities are also characterised by temporal fluctuations in plant species composition often due to stochastic phenomena or abrupt climate events. For example, slow plant dispersal or failure

*Corresponding author. Email: [email protected] © 2017 Botanical Society of Scotland and Taylor & Francis

to establish can result in a slower rate of vegetation change in response to climate warming (Savage and Vellend 2014). In addition, the treeline is subject to abrupt changes in temperature and snow regimes that can trigger treeline advances or contractions and thus affect plant communities (Camarero and Gutiérrez 2002; Camarero et al. 2006). Several studies have reported an upward movement of plant species along elevational gradients and subsequent changes in community composition (e.g. Lenoir et al. 2008), while others have reported minor changes in vegetation patterns across the treeline ecotone (e.g. Pardo et al. 2013). Discrepancies among studies may arise as a consequence of local and regional contingencies or stochasticity determining treeline–vegetation dynamics (Harsch et al. 2009). However, further research considering how biotic factors, such as plant–plant interactions and abiotic constrains influence vegetation patterns across treelines is required to understand how upward treeline advances and denser tree cover in the treeline ecotone determine vegetation changes and alter ecosystems services, such as carbon sequestration or water regulation and biodiversity (Greenwood and Jump 2014). Alpine areas in many regions of the world such as Europe have been subjected to human land use for centuries (Gehrig-Fasel et al. 2007), and the cessation of traditional land use has recently impacted ecosystems (Vittoz et al. 2009). Thus, undisturbed areas where local human intervention has been low, such as the Urals

Downloaded by [2.152.229.44] at 07:56 14 December 2017

284

A. Gazol et al.

(Moiseev et al. 2004; Solly et al. 2017), are fundamental to understand how mountain vegetation responds to climate fluctuations (Camarero et al. 2006). Under natural conditions, changes in vegetation patterns along elevational gradients are mainly driven by temperature lapse rates (Körner 2003). However, other factors, such as precipitation regimes, topography, soil nutrient availability and anthropogenic influences can modify mountain vegetation (Greenwood and Jump 2014). The increase in cover of woody vegetation may lead to deeper snow packs, which raises soil temperatures in winter, and thus may facilitate woody plant growth (Sturm et al. 2005). For example, Kammer et al. (2009) have reported that snow depth abruptly increased from 20 cm to more than 80 cm from low-stature alpine vegetation to open forest in the slopes of the Mali Iremel situated in the Southern Ural Mountains. In addition, the increase in tree cover reduces light levels on the forest floor that may impact community composition by excluding shade-intolerant plant species (Hofgaard and Wilmann 2002; Greenwood and Jump 2014). Thus, plant species coexistence along an elevational gradient may depend on the ability of species to tolerate the abiotic conditions prevailing in each elevation belt (Weiher and Keddy 1995; Cornwell and Ackerly 2009; De Bello et al. 2013; Suding et al. 2015), but also on their ability to interact with neighbours (Callaway et al. 2002), and on their capacity to respond to changes in abiotic and biotic conditions (Grassein et al. 2014). Plant functional traits are measurable features of plant anatomical and morphological characteristics that reflect how plants vary in ecological strategies of growth, resource use and response to environmental conditions (Westoby and Wright 2006). As functional traits capture information on habitat requirements, it is expected that plant species arrays across the treeline ecotone may display different functional traits in response to a strong abiotic filtering. Thus, the mean trait value or functional composition of plant communities (community-weighted mean, Lavorel et al. 2008) across the elevation range should vary between different elevation belts. However, plant communities can vary not only in the mean trait value but also in their functional diversity, i.e. the variability in trait differences between coexisting species (Laliberté and Legendre 2010). These patterns in functional diversity, when compared to the expectations quantified with appropriate null models, may help to understand factors driving community assembly along environmental gradients (Adler et al. 2013; Mason et al. 2013). In harsh environments, abiotic conditions are expected to reduce functional diversity (Weiher and Keddy 1995), whereas in productive environments under favourable abiotic conditions niche complementarity among species will be required to reduce competition and enable the coexistence of a great number of plant species resulting in greater functional diversity (Chesson 2000). Thus, functional diversity should decrease across the treeline ecotone with elevation. Nevertheless, low

functional diversity may arise as a consequence of biotic interactions if species share similar resource-use strategies (equalising fitness processes; cf. Chesson 2000). Conversely, Spasojevic and Suding (2012) have shown that large values of functional diversity in stressful environments, in which abiotic filtering is expected to decrease functional diversity, can arise as consequence of facilitative interactions among functionally dissimilar plant species. However, patterns in functional composition and diversity across elevation gradients depend on the species pool size that determine the functional traits of the plant species that can coexist in each community. In this study, we tracked the main spatio-temporal changes in plant community assembly patterns across the treeline ecotone. Specifically, we resampled vegetation patterns along an elevational gradient located in the south Ural Mountains after 12 years of the initial survey. We used functional diversity metrics combined with null models to study the variation in community structure across an elevation gradient and over time. Treeline advances result in changes in above-ground and belowground conditions by reducing light values and increasing snow depth. These changes should modify the presence and relative abundance of plant species. Therefore, we hypothesised that the replacement in plant species composition mainly occurs across the treeline ecotone and along the elevational gradient, whereas changes in the relative abundances of species will be observed through time within each elevational level of the ecotone.

Materials and methods The study site was located in the Mali Iremel Massif, southern Urals, Russia (54° 32ʹ N, 58° 51ʹ E; see Figure 1). In this area the treeline ecotone is situated between 1200 m and 1400 m a.s.l. The vegetation of the treeline ecotone and the alpine zone is largely natural and not impacted by local anthropogenic disturbances such as

Figure 1. General view of the alpine vegetation (locally called mountain tundra) near a treeline located in the South Urals, Russia.

Downloaded by [2.152.229.44] at 07:56 14 December 2017

Changes in plant taxonomic and functional diversity patterns grazing, fires and logging. Abies sibirica Ledeb. and Picea obovata Ledeb. forest grows in valleys and slopes up to 1250 m. The lower part of the treeline is dominated by open P. obovata forests and meadows formed by tall herbaceous vegetation. By contrast, the upper part of the treeline is dominated by crooked spruce and birch forests (Betula tortuosa Ledeb.), moss-rich short grass meadows, shrubs (Juniperus sibirica Burgsd., Salix spp.) and mossgrass heaths. At elevations higher than 1400 m, alpine vegetation, formed by moss-dwarf shrub and moss-lichen heath communities is dominant on gentle slopes, plateaux and terraces. According to long term daily climate data from the Taganai-Gora weather station, located 115 km north-east of the study area at 1114 m a.s.l., January mean monthly temperature is −14.5°С, and July mean monthly temperature is 12.3°С (for 1933–2001). Precipitation ranges from 600 to 1300 mm, with an average of 936 mm (Gurskaya et al. 2016). Finally, the average snow depth in late April is about 123 cm (Moiseev et al. 2004). Local snow depth measurements over the 2005–2015 period showed a pronounced influence of trees on the snow cover in the Mali Iremel Massif. At the treeline, drift snow accumulates under the tree canopy and on the leeward side of tree islands (50–100 cm; 89 cm mean snow depth in late March) vs. 20 cm over alpine vegetation. Winter soil temperatures (taken at 10 cm depth) were around 0°C in the upper montane forest, ranged from −1° C to 0°C to in the tree islands and from −7°C to −1°C in the alpine vegetation. In summer, soil temperatures reached up to +10°C in both the alpine vegetation and at the treeline, being 1°C colder under trees (Kammer et al. 2009). Bedrock across the treeline ecotone is homogenous and consist of quarzite blocks. The soils are Haplic Cambisols without any signs of podzolisation. Soils B horizons were found almost directly above the blocky parent material because soils had very thin C horizons. Soils are acidic with pH values ranging between 3.4 and 3.8 at all elevations. Stone content in mineral soils is <10%. The organic layers were thicker at the alpine vegetation (10 cm under trees, 7 cm under the alpine vegetation) than in the forest (8 cm under trees, 2.5 cm in the open land). At the alpine vegetation above the treeline situated at 1400 m, organic layers were dominated by the Oa horizon (6 cm), which was absent in the forest at 1260 m a.s.l. (Kammer et al. 2009). The upward expansion and encroachment of treeline ecotones in the Southern Urals during the past 50–110 years has been linked to climatic factors, such as warmer conditions (Hagedorn et al. 2014). The available meteorological data provide evidence that the climate has become warmer and more humid during the past 170 years in the study region, especially in winter months (Hagedorn et al. 2014). For example, Moiseev et al. (2016) have reported an increase of 0.2°C in summer temperature (May to August) and an increase of 1.8°C during the cold season (November to March) for the Zlatoust weather

285

station (55° 10ʹ 00ʺ N, 59° 40ʹ 00ʺ E) from the mid-nineteenth to the early twenty-first century. Vegetation sampling Vegetation was sampled in 2002 and in 2014 from the uppermost closed forest (>50% tree cover) to the treeline where only scattered (groups) trees of at least 2-m tall are present (<7% tree cover) in an alpine vegetation matrix. First, three regions where defined across the treeline ecotone: the uppermost closed upper montane forest (elevational level V), the closed forest limit (less than 50% cover; elevational level III) and the treeline and open tundra (elevational level I; see more details in Hagedorn et al. 2014). At each of these elevation levels, we established three 20 m × 20 m plots in 2002. In each plot, vegetation was sampled in 16 0.5 m × 0.5 m squares arranged in a 10-m-long cross placed in each corner of the plots. The plots and squares were marked permanently and resurveyed in 2014. Vegetation sampling was made in summer coinciding with the peak of the growing season (July–August). We identified all vascular plant species present in the plot and visually estimated their cover (percentage scale). The second author, who is expert of the flora of the Urals, participated in the two surveys to estimate plant cover which allowed excluding inter-observer bias. Plant material collection In 2014, we measured five plant traits that are expected to reflect key aspects of plant functioning and adaptations to the environment (Cornelissen et al. 2003; Moles et al. 2009). In each elevation zone, we sampled trait data for the 25 most abundant plant species detected in previous surveys. Common species along the entire gradient were sampled in the three elevational levels (e.g. Calamagrostis uralensis, Polygonum alpinum). Plant height, leaf area, specific leaf area (SLA), leaf nitrogen and carbon concentrations were measured by using standardised protocols (Cornelissen et al. 2003). Height is related with competitive ability and aboveground resource acquisition (Moles et al. 2009). Leaf area is linked to energy consumption and it is expected that environmental stress (e.g. short growing season due to low temperatures) will reduce leaf size (Cornelissen et al. 2003). SLA is a measure of photosynthetic capacity and it encodes information on growth rate and tissue longevity (Cornelissen et al. 2003). Finally, leaf N concentration, and C concentration to a lesser extent, reflect leaf mineral nutrition and are connected with soil nutrient levels and resource-use efficiency (Welker et al. 2005). Plant height was measured as the length from ground level to the highest photosynthetically active tissue in 20 individuals of each species. For leaf traits, we collected fully formed adult leaves from 12 individuals, stored them in paper bags, in which were placed in watertight containers to avoid desiccation until they could be weighed and

286

A. Gazol et al.

Downloaded by [2.152.229.44] at 07:56 14 December 2017

scanned to measure leaf area and SLA. Leaf area was measured as the one-sided projected surface area of a whole leaf based on scanned images, using the software ImageJ (Rasband 2007). SLA was calculated as the ratio of leaf area to dry mass (mm2 mg−1). Leaf N and leaf C concentration were measured with an elemental analyser (Elementar VarioMAX N/CM, Elementar Analysensysteme GmbH, Hanau, Germany). Data analyses We used non-metric multidimensional scaling (NMDS; Legendre and Legendre 1998) to visualise patterns in plant community composition. The Bray–Curtis dissimilarity metric was used to calculate the similarity between samples in species composition. Diversity patterns across the elevational gradient were represented by: (1) plot-wise species richness; (2) the inverse of the Simpson dominance index (Hill 1973), as well as (3) the total number of species found in each elevational level. In order to describe functional composition in each plot we calculated community weighted mean (CWM) trait values (Lavorel et al. 2008). CWMs are calculated by weighting the mean trait value of each species by the relative abundance of each species in each plot. This metric can be used to quantify changes in mean community trait values along elevational gradients (De Bello et al. 2013). We calculated CWM for each plant trait. To control for the inter-specific trait variation along the elevational gradient, we only considered the mean trait for the species collected in each particular elevational level. For example, a species collected in the three elevations were represented by three mean trait values (Appendix S1). Relative abundances for each species were calculated using the percent cover estimated for each species in each plot. In order to avoid a skewed distribution of values, mean trait values of leaf size, leaf N content and plant height were log-transformed prior the calculations of the CWM values. We expressed functional diversity as functional richness (FRic) and functional dispersion (FDis) following Laliberté and Legendre (2010). In multidimensional trait space, FRic represents the convex hull volume of the species present in the community and FDis is the mean distance of each species, weighted by relative abundances, to the centroid of all species in the community. While there are currently many metrics of functional diversity available, FRic and FDis were the most appropriate indices because they represent patterns in functional diversity that can be attributed to changes in species composition and species relative abundance, respectively (Laliberté and Legendre 2010; Mason et al. 2013). We calculated FRic and FDis for each trait individually and for all combined traits. In order to assess the processes governing community assembly we compared observed values of FRic and FDis with null models (Mason et al. 2013). For the standardisation of FRic, we randomised the species occurrence matrix (999 permutations) while keeping community species

richness and frequency constant (Mason et al. 2013). For the standardisation of FDis, we randomised relative abundances across species within communities (Mason et al. 2013). For each randomisation, the functional diversity of the pooled traits and individual traits were calculated as above. Hence, any differences observed between the measured and simulated values were only generated by processes governing species richness (FRic) and species abundance (FDis), respectively (Mason et al. 2013). Finally, we used the standardised effect size (SES) to express FRic (SESFRic) and FDis (SESFDis) relative to that expected by chance: SES ¼

Obs  Exp σ Exp

(1)

where Obs is the observed functional diversity for each trait in each plot, Exp is the mean value of the randomisations and Exp its standard deviation of these randomised values. In order to test for the differences in community composition between habitat types we used Permutational Multivariate Analysis of Variance (PERMANOVA; McArdle and Anderson 2001). To account for the nested structure of the data PERMANOVA analyses were performed in two steps. In a first step, we tested for the influence of elevational level, time, and its interaction on species composition while controlling by the effect of sampling plot (48 plots per elevational) using the adonis function of the vegan package (Oksanen et al. 2013). In a second step, we tested for the influence of elevational on species composition while controlling for the effect of sampling plot using the function nested.npmanova of the BiodiversityR package (Kindt and Coe 2005). We applied linear mixed-effects models (LME; Pinheiro et al. 2015) to study the variation in taxonomic and functional composition and diversity along the elevational gradient and as a function of time. In each model, elevation, time and their interaction were considered fixed factors and plot was regarded as a random factor. To identify differences between the three elevations and two sampling occasions we carried out post-hoc analyses. Particularly, we tested for the differences in time within each level and for the differences in elevation within each sampling year. In addition, we tested whether there were significant differences in functional composition and diversity between plots with tree cover above or below the 50% threshold (Appendix S2). In this case, we only considered two elevations (III, open forest limit; and V, uppermost subalpine closed forest) since trees were absent in the open alpine vegetation. For the comparisons in each level and sampling year the Kruskal–Wallis statistic test was applied. All calculations were made in the R environment (R Core Team 2015). CWMs and functional diversity measures were calculated using the functions funcomp and dbFD in the package FD (Laliberté et al. 2014), NMDS

Changes in plant taxonomic and functional diversity patterns analyses were performed using the function metaMDS in the package vegan (Oksanen et al. 2013) and LMEs were performed using the function lme in the nlme package (Pinheiro et al. 2015).

Downloaded by [2.152.229.44] at 07:56 14 December 2017

Results We found clear changes through time in species composition along the elevational gradient (Figure 2). The first axis of the NMDS separated the vegetation of the treeline (level I) from the vegetation in the open forest limit (level III) and uppermost montane forest (elevational level V). The second axis, less important, separated vegetation between the two forest levels mentioned earlier. There were no evident changes in plant composition between 2002 and 2014 at the treeline as indicated by the location of the grey and black ellipses in Figure 2. However, in forests levels III and V vegetation became more heterogeneous as indicated by the ellipses showing the centroid of vegetation ordination in each elevation (grey and black ellipses in Figure 2). In the uppermost elevational, species such as Carex bigelovii, Festuca igoschiniae, and Vaccinium uliginosum had reduced abundance, whereas others, such as Polygonum viviparum, Luzula multiflora, Veratrum lobelianum and Vaccinium vitis-idaea have increased their cover (see Appendix S3 and S4). In elevational levels III

Figure 2. Temporal change in plant species composition in different elevation belts, southern Urals, Russia. Plotted scores correspond to the first and second axes of a Non-metric Multidimensional Scaling (NMDS 1, NMDS 2) based on plant species composition. Different symbols and colours are used for each elevation belt and sampling time: treeline and open tundra, I – triangles (2002 – grey; 2014 – black); closed forest limit, III – squares (2002 – light green; 2014 – dark green); closed upper montane forest, V – circles (2002 – light blue; 2014 – dark blue). The ellipses show the hypothetical centroid of species composition in 2002 (grey dashed line) and 2014 (black dashed line).

287

and V, species, such as Calamagrostis langsdorfii, Geranium sylvaticum and Polygonum bistorta were not detected or have decreased in abundance, whereas others such as Calamagrostis uralensis, Campanula rotundifolia, Stellaria nemorum and Vaccinium myrtillus have increased markedly (Appendix S3 and S4). Plot-wise species richness and diversity decreased with elevation, whereas species pool (i.e. total number of species per elevation) increased (Figure 3). In 2002, species richness was significantly greater at the treeline (level I) and at the open forest limit (level III) than in the upper montane forest (level V). In the case of diversity, only the uppermost elevation differed significantly from the lowermost. In 2014, both plant richness and diversity were significantly greater at the highest and intermediate elevations as compared with the lowest. When compared the two samplings, the only significant difference was found for the species richness in level V which decreased through time, while no differences were found between levels I and III. We found significant differences in CWM for several of the traits studied (Figure 4; Table 1; Appendix S5). Functional composition in height, LA, SLA, leaf N and C contents was significantly lower in level I than in levels III and V, whereas no differences were found between levels III and V. At level I, we found that the functional composition in all the traits excluding leaf C content has changed significantly from 2002 to 2014. In level III, we found significant differences in leaf traits (excluding leaf C concentration). In level V, we only found significant differences in LA and leaf N content. When separating plots according to tree coverage in levels III and V we found significant changes in functional composition for several traits (Table 2; Appendix S6). We also detected significant changes in functional diversity along the elevational gradient (Figure 4; Table 1). We found a lower than expected FRic in plant height in level I in contrast with levels III and V. Reverse patterns were observed for LA and leaf N content for which we found a larger than expected functional diversity in the uppermost level I. FRic in SLA and leaf C content showed little variation along the elevational gradient. In terms of functional dispersion, we found a change from lower than expected functional diversity in height and LA to greater than expected in level I. For the rest of levels, changes were not so clear. When separating plots according to tree coverage in levels III and V we found significant changes in FRic and dispersion for several traits (Table 2; Appendix S6). Discussion We found changes in the occurrence of species and their relative abundance between the two sampling campaigns across the treeline ecotone in the South Urals. We also found major changes in species composition between the treeline and the upper montane forest and the open forest limit. However, our results show that changes in species

A. Gazol et al.

Downloaded by [2.152.229.44] at 07:56 14 December 2017

288

Figure 3. Plant species diversity observed along the elevational gradient across the treeline and across time, southern Urals, Russia. For each elevational belt (I, treeline and open tundra; closed forest limit, III; closed upper montane forest, V), the diversity (inverse of the Simpson dominance index), species richness and total number of species in the level (species pool) are shown for the two sampling years (2002 – grey bars; 2014 – black bars). See the corresponding statistics in Table 1.

composition across time are more apparent in the upper montane forest and the open forest limit than in the treeline ecotone, where they seem to be minor (Figure 2). These changes in species composition can be attributed to natural vegetation dynamics and climate fluctuations

since the human interference is rather scarce in the region (Moiseev et al. 2004; Solly et al. 2017). Our results partially confirm the hypothesis that major changes in community assembly in space are related to changes in species composition (turnover), whereas temporal changes are mainly influenced by changes in the occurrence and relative abundance of the species present at each elevational. We also found that the increase in tree cover in the closed and open forests is a major factor driving species richness and composition in the lower part of the treeline ecotone (Camarero et al. 2006; Vittoz et al. 2009). Conversely, changes in community assembly in alpine vegetation near the uppermost part of the treeline ecotone respond to other factors, probably related to the recently observed rise in air temperatures (Pauli et al. 2012). Overall, our results support previous findings suggesting that trait patterns and associated assembly processes depend on the plant trait considered (Spasojevic and Suding 2012). Our findings suggest that communities in the closed and open upper montane forest are composed of more productive species than the communities dominating the treeline–alpine transition as demonstrated by the changes in functional composition of height, leaf area and leaf N content (Weiher and Keddy 1995; Welker et al. 2005; Spasojevic and Suding 2012). In addition, the duration of the growing season rapidly decreases with elevation, resulting in communities dominated by less productive plants (De Bello et al. 2013). These results confirm the presence of a marked abiotic filtering across the treeline ecotone, similar to other severe abiotic filters observed in other forest–alpine ecotones (Spasojevic and Suding 2012). We found that FRic in plant height increases from the more stressful tundra towards the closed forest. In addition, when controlling for the influence of tree cover, FRic in SLA shows clear differences between the closed forest and alpine communities. Changes in plant functional diversity along elevational gradients have been observed by other authors as well. For example, De Bello et al. (2013) found a decrease in FRic in leaf area, SLA and height with the increase in elevation in the Alps. Conversely, Pescador et al. (2015) found an increase in plant height and leaf area with the increase in elevation in Mediterranean mountains. The changes in FRic along the Mali Iremel transect may respond to the decrease in tree cover and increase in snow depth with elevation (Holtmeier and Broll 1992; Sturm et al. 2005; Kammer et al. 2009). Thus, we hypothesise that the presence of a greater than expected FRic in height, SLA, and leaf N content in the lower part of the treeline as compared to the upper part is related to an increase in competition for light caused by the presence of tall trees. This is in line with the limiting similarly hypothesis that states that species may differ in their resource-use strategy for key resources such as light in order to coexist (Stubbs and Wilson 2004). Nevertheless, a reduced FRic in leaf area in the lower part of the treeline ecotone as compared to the upper part

Changes in plant taxonomic and functional diversity patterns Table 1.

Spatial and temporal changes in plant functional composition and diversity in treeline plots from the southern Urals, Russia.

Downloaded by [2.152.229.44] at 07:56 14 December 2017

Differences in space (2002/2014)

CWM Height LA SLA Leaf N Leaf C FRic Height LA SLA Leaf N Leaf C FDis Height LA SLA Leaf N Leaf C

289

Differences in time

I vs. III

I vs. V

III vs. V

I vs. I

III vs. III

V vs. V

-12.10/-13.05 -12.48/-13.60 -10.25/-11.97 -10.55/-0.87 5.21/6.93

-10.62/-14.26 -19.53/-19.50 -17.90/-18.02 -13.71/-5.49 6.18/7.33

1.48/-0.94 -7.05/-5.62 -7.65/-5.79 -3.16/-4.53 0.97/0.29

3.75 4.04 3.58 -9.21 -0.58

2.16 1.92 0.64 5.34 2.03

-0.88 4.09 3.39 3.33 1.12

-3.13/-2.26 -0.45/0.89 0.74/0.21 8.37/8.53 1.74/1.44

-5.52/-3.30 3.52/5.45 -0.71/-1.30 3.59/3.27 -0.25/-1.12

-2.39/-0.97 3.97/4.46 -1.46/-1.49 -4.77/-5.26 -0.25/-2.53

-2.85 -2.04 0.26 -0.87 0.64

-1.52 0.15 -0.60 -0.35 0.12

0.56 1.08 0.73 -1.31 -0.92

0.38/5.35 1.18/2.03 2.93/5.54 -2.45/1.42 3.57/2.74

-0.89/3.49 -1.28/0.80 4.71/7.01 0.20/3.97 -0.51/-3.14

-1.26/-1.90 -2.45/-1.25 1.78/1.35 2.66/2.45 -4.08/-5.80

-5.82 -0.51 -3.46 -4.86 1.08

0.41 0.52 -0.31 -0.54 0.09

-0.43 1.88 -0.75 -0.70 -2.39

Abbreviations of plant traits: LA: leaf area; SLA: specific leaf area; leaf C and N: leaf carbon and nitrogen contents. The values of several plant traits, functional composition (CWM), and standardised effect sizes in functional Richness (FRic) and Functional dispersion (FDis) are compared among three elevational levels (I: treeline; III: open forest limit; and V: closed upper montane forest) and between sampling years (2002 vs. 2014). For each elevational level, we show the z-statistic testing for the presence of significant differences between each level in each sampling year. The first value indicates if there are differences in 2002, and the second one if there are differences in 2014. The last three columns show if there are differences within each level between two– sampling years. Significant values are shown in bold (P < 0.05).

Figure 4. Plant functional composition and diversity at the treeline, South Urals, Russia. For each elevational belt (I, treeline and open tundra; closed forest limit, III – ; closed upper montane forest, V), the community-weighted mean and standardised effect size on functional richness (SESFRic) and Functional Dispersion (SESFDis) are shown for the two sampling years (2002 – grey bars; 2014 – black bars). See the corresponding statistics in Table 1.

290

A. Gazol et al.

Table 2. Comparisons of plant functional composition and diversity in treeline–ecotone areas with different tree canopy cover, southern Urals, Russia. Level III

Downloaded by [2.152.229.44] at 07:56 14 December 2017

2002

CWM Height LA SLA Leaf N Leaf C FRic All traits Height LA SLA Leaf N Leaf C FDis All traits Height LA SLA Leaf N Leaf C

Level V

2014

K

P

K

5.82 3.62 6.71 0.81 0.37

0.016 0.057 0.01 0.368 0.541

3.46 8.7 8.42 0.19 0.02

5.63 0.77 0.03 2.06 1.31 0.07

0.018 0.38 0.868 0.151 0.252 0.798

4.23 4.61 3.7 6.83 0.03 0.25

0.04 0.032 0.054 0.009 0.868 0.617

P

2002 K

2014 P

K

P

0.063 13.02 0 0.003 12.58 0 0.004 0.01 0.918 0.661 4.96 0.026 0.878 12.14 0

4.64 8.33 3.23 6.51 7.17

0.031 0.004 0.072 0.011 0.007

0.58 2.41 0 5.94 0.02 0.97

0.446 0.121 0.961 0.015 0.883 0.325

0.08 0.21 1.38 4.69 7.52 0.29

0.773 0.65 0.24 0.03 0.006 0.592

0.47 1.01 0.01 2.02 1.96 0.37

0.492 0.315 0.933 0.155 0.161 0.544

2.97 2.8 2.04 0.92 0.79 1.58

0.085 0.094 0.153 0.337 0.375 0.209

0.86 0.1 2.14 4.51 0.04 2.65

0.353 0.757 0.143 0.034 0.837 0.103

0 0.07 0.25 3.14 0.05 0.82

1 0.79 0.616 0.076 0.83 0.366

Abbreviations of plant traits: LA: leaf area; SLA: specific leaf area; leaf C and N: leaf carbon and nitrogen contents. The results show significant differences in functional composition (CWM) and standardised effect sizes in functional richness and dispersion (FRic and FDis, respectively) between plots with different tree cover considering several plant traits, two sampling periods (2002, 2004) and two elevational levels (III: open forest limit; and V: closed upper montane forest). Specifically, we tested whether there were significant differences between plots with tree cover above or below the 50% threshold. For the comparisons in each level and sampling year the Kurskall–Wallis statistic test and associated probability levels are shown. Significant differences (P < 0.05) are highlighted in bold.

may indicate that assembly patterns for this trait are not driven by light competition. In this sense, the lower than expected FRic in leaf area will respond to the competitive exclusion of species with small leaf areas as a consequence of the strong competition for light in the lower part of the treeline ecotone, or increased functional diversity in the upper part of the ecotone due to strong competition for soil resources. In this sense, Solly et al. (2017) have reported that the roots of shrubs and herbaceous plants increased with elevation in the South Urals and suggested that it might be related to a strong competition for below ground resources. Nevertheless, caution is required when interpreting the results as we did not measure environmental conditions directly (Adler et al. 2013). We found contrasting results regarding the temporal changes of vegetation. First, we found a decrease in plot-wise species richness and diversity across all levels which is in accordance with previously reported results (Wilson and Nilsson 2009). Second, the total number of species found in the upper montane forest and the open

forest limit increased along time, which suggests an upward movement of some plant species (Venn et al. 2012; Savage and Vellend 2014). These discrepancies between local and regional scales (i.e. sampled squares and total number of species in the plots) have been previously reported and attributed to the great resilience of alpine plant communities at the local scale (Vittoz et al. 2009; Venn et al. 2012). We also found that changes in plot-wise species richness and diversity in the South Urals are more apparent at the lowermost elevational near the subalpine forest limit in line with previous studies (Klanderud and Birks 2003; Lenoir et al. 2008; Wilson and Nilsson 2009). Our results show that there are strong changes in the presence and relative abundance of some plant species across the treeline ecotone through time which can be attributed to different processes. First, the year-to-year change in climate conditions, including the occurrence of climate extremes, such as frosts, can influence the germination and growth of several alpine plant species (Körner 2003). Second, stochasticity can also influence the composition of alpine plant communities (Yang et al. 2013). Finally, there might be a change in plant assembly processes as a consequence of upward treeline advances. We showed indirectly that community-weighted means in height, leaf area, SLA and leaf N content decreased from 2002 to 2014. In addition, we observed an increase in functional diversity in height, SLA and leaf N content which can be attributed to an increase in competition (Stubbs and Wilson 2004). In our study, temporal changes in functional composition are the consequence of the change in the occurrence and relative abundance of some key plant species since we have not considered intraspecific trait variability. Moreover, trait values were only measured in 2014 and not in 2002 which makes impossible to detect changes in species-specific trait values along time. For instance, the reduction in height and leaf area observed at the treeline is due to the rapid increase in abundance of the shrub Vacciminium vitis-idaea and the reduction of its congeneric V. uliginosum. At the other two elevations, the changes in functional composition were mainly driven by the reduction in abundance of Polygonum bistorta, and the increase of graminoids such as Calamagrostis uralensis. These results emphasise how changes in the presence and relative abundance of some dominant plant species can have important changes in the functional composition at the community level. This also highlights the need to exercise caution when interpreting the results of temporal changes in community assembly, since sampling effort and accuracy can influence those patterns (Vittoz et al. 2009; Savage and Vellend 2014). In addition, the fluctuations in species relative abundances can also arise as a consequence of short-term climate oscillations, including climate extremes, such as cold spells, and we cannot discard the effect of such events since we only have two sampling campaigns. Thus, a long-term monitoring of these treeline ecotones, together with high-precision climate data recorded in situ are required to track changes in community assembly along time.

Changes in plant taxonomic and functional diversity patterns

Downloaded by [2.152.229.44] at 07:56 14 December 2017

We have provided insight on plant community assembly processes at treeline ecotones by characterising spatio-temporal changes in vegetation patterns, combined with the measurement of multiple plant traits. All in all, our results suggest that changes have occurred in vegetation patterns in the South Urals treeline ecotone. These changes are complex and require the consideration of multiple aspects of vegetation. However, the lack of anthropogenic management in the region ensure that these changes are mainly driven by natural vegetation dynamics. Species turnover along the elevational gradient is the main process governing community assembly in space. Nevertheless, different assembly processes operate simultaneously on different traits. Temporal changes are much more difficult to interpret and respond mainly to shifts in the occurrence and relative abundance of dominant or key plant species that have a stronger weight in functional composition and diversity metrics. Disclosure statement No potential conflict of interest was reported by the authors.

Funding This work was support by the TreeClim ERA.Net RUS Pilot Joint Call for Collaborative S&T Projects (funded under the 7th European Framework Programme for Research and Development). Antonio Gazol was supported by a Postdoctoral grant form the Ministerio de Economía y Competitividad (Contrato Formación Postdoctoral MINECO – FPDI 640 201316600). We also thank the support of the project CGL201126654 (Ministerio de Economía y Competitividad).

Supplemental data Supplemental data for this article can be accessed here.

ORCID Antonio Gazol

http://orcid.org/0000-0001-5902-9543

Pavel Moiseev

http://orcid.org/0000-0003-4808-295X

J. Julio Camarero

http://orcid.org/0000-0003-2436-2922

Notes on contributors Antonio Gazol is a plant ecologist who uses different approaches to disentangle how biotic and abiotic factors influence spatiotemporal plant species patterns. Pavel Moiseev aims to understand how climate changes influence on spatio-temporal variations of microclimatic and soil conditions and vegetation dynamics above the timberline. J. Julio Camarero is a dendroecologist, working on growth, regeneration, dieback and mortality processes in woody plants.

References Adler PB, Fajardo A, Kleinhesselink AR, Kraft NJB. 2013. Traitbased tests of coexistence mechanisms. Ecology Letters. 16:1294–1306.

291

Callaway RM, Brooker RW, Choler P, Kikvidze Z, Lortie CJ, Michalet R, Paolini L, Pugnaire FI, Newingham B, Aschehoug ET, et al. 2002. Positive interactions among alpine plants increase with stress. Nature. 417:844–848. Camarero JJ, Gutiérrez E. 2002. Plant species distribution across two contrasting treeline ecotones in the Spanish Pyrenees. Plant Ecology. 162:247–257. Camarero JJ, Gutiérrez E. 2004. Pace and pattern of recent treeline dynamics: response of ecotones to climatic variability in the Spanish Pyrenees. Climatic Change. 63:181–200. Camarero JJ, Gutiérrez E, Fortin MJ. 2006. Spatial patterns of plant richness across treeline ecotones in the Pyrenees reveal different locations for richness and tree cover boundaries. Global Ecology and Biogeography. 15:182–191. Chesson P. 2000. Mechanisms of maintenance of species diversity. Annual Review of Ecology, Evolution and Systematics. 31:343–366. Cornelissen JHC, Lavorel S, Garnier E, Diaz S, Buchmann N, Gurvich DE, Reich PB, Ter Steege H, Morgan HD, Van Der Heijden MGA, et al. 2003. A handbook of protocols for standardized and easy measurement of plant functional traits worldwide. Australian Journal of Botany. Vol. 51:335–380. Cornwell WK, Ackerly DD. 2009. Community assembly and shifts in plant trait distributions across an environmental gradient in coastal California. Ecological Monographs. 79:109–126. de Bello F, Lavorel S, Lavergne S, Albert CH, Boulangeat I, Mazel F, Thuiller W. 2013. Hierarchical effects of environmental filters on the functional structure of plant communities: a case study in the French Alps. Ecography. 36:393–402. Devi N, Hagedorn F, Moiseev P, Bugmann H, Shiyatov S, Mazepa V, Rigling A. 2008. Expanding forests and changing growth forms of Siberian larch at the Polar Urals treeline during the 20th century. Global Change Biology. 14:1581–1591. Gehrig-Fasel J, Guisan A, Zimmermann NE. 2007. Tree line shifts in the Swiss Alps: climate change or land abandonment? Journal of Vegetation Science. 18:571–582. Grace J, Berninger F, Nagy L. 2002. Impacts of climate change on the tree line. Annals of Botany. 90:537–544. Grassein F, Lavorel S, Till-Bottraud I. 2014. The importance of biotic interactions and local adaptation for plant response to environmental changes: field evidence along an elevational gradient. Global Change Biology. 20:1452–1460. Greenwood S, Jump AS. 2014. Consequences of treeline shifts for the diversity and function of high altitude ecosystems. Arctic, Antarctic, and Alpine Research. 46:829–840. Gurskaya M, Moiseev P, Wilmking M. 2016. Does slope exposure affect frost ring formation in Picea obovata growing at treeline in the Southern Urals? Silva Fennica. 50:1560. Hagedorn F, Shiyatov SG, Mazepa VS, Devi NM, Grigorev AA, Bartysh AA, Fomin VV, Kapralov DS, Terent’ev M, Bugman H, et al. 2014. Treeline advances along the Urals mountain range – driven by improved winter conditions? Global Change Biology. 20:3530–3543. Harsch MA, Hulme PE, McGlone MS, Duncan RP. 2009. Are treelines advancing? A global meta-analysis of treeline response to climate warming. Ecology Letters. 12:1040– 1049. Hill MO. 1973. Diversity and evenness: a unifying notation and its consequences. Ecology. 54:427–473. Hofgaard A, Wilmann B. 2002. Plant distribution pattern across the forest–tundra ecotone: the importance of treeline position. Écoscience. 9:375–385. Holtmeier FK, Broll G. 1992. The influence of tree islands and microtopography on pedoecological conditions in the forestalpine tundra ecotone on Niwot Ridge. Colorado Front Range, U.S.A. Arctic and Alpine Research Vol. 24:216–228. Kammer A, Hagedorn F, Shevchenko I, Leifeld J, Guggenberger G, Goryacheva T, Rigling A, Moiseev P. 2009. Treeline

Downloaded by [2.152.229.44] at 07:56 14 December 2017

292

A. Gazol et al.

shifts in the Ural mountains affect soil organic matter dynamics. Global Change Biology. 15:1570–1583. Kindt R, Coe R. 2005. Tree Diversity Analysis. A Manual and Software for Common Statistical Methods for Ecological and Biodiversity Studies. World Agroforestry Centre (ICRAF), Nairobi. Klanderud K, Birks HJB. 2003. Recent increases in species richness and shifts in altitudinal distributions of Norwegian mountain plants. The Holocene. 13:1–6. Kohler T, Wehrli A, Jurek M. (eds.) 2014. Mountains and climate change: A global concern. Sustainable mountain development series. Bern. Switzerland: Centre for Development and Environment (CDE), Swiss Agency for Development and Cooperation (SDC) and Geographica Bernensia. p. 136. Körner C. 2003. Alpine plant life. 2nd ed. Heidelberg: Springer. Körner C. 2012. Alpine treelines – functional ecology of the global high elevation tree limits. Basel: Springer. Laliberté E, Legendre P. 2010. A distance-based framework for measuring functional diversity from multiple traits. Ecology. 91:299–305. Laliberté E, Legendre P, Shipley B. 2014. FD: measuring functional diversity from multiple traits, and other tools for functional ecology. R Package Version. 1.0-12. http://CRAN.R-project. org/package=FD. (Accessed 06 November 2017). Lavorel S, Grigulis K, McIntyre S, Williams NSG, Garden D, Dorrough J, Berman S, Quetier F, Thebault A, Bonis A. 2008. Assessing functional diversity in the field - methodology matters! Functional Ecology. 22:134–147. Legendre P, Legendre L. 1998. Numerical ecology. Amsterdam: Elsevier. Lenoir J, Gegout JC, Marquet PA, De Ruffray P, Brisse H. 2008. A significant upward shift in plant species optimum elevation during the 20th century. Science. 320:1768–1771. Mason NWH, de Bello F, Mouillot D, Pavoine S, Dray S. 2013. A guide for using functional diversity indices to reveal changes in assembly processes along ecological gradients. Journal of Vegetation Science. 24:794–806. McArdle BH, Anderson MJ. 2001. Fitting multivariate models to community data: A comment on distance-based redundancy analysis. Ecology. 82:290–297. Moiseev PA, Bubnov MO, Devi NM, Ya NZ. 2016. Changes in the structure and phytomass of tree stands at the upper limit of their growth in the Southern Urals. Russian Journal of Ecology. 47:219–227. Moiseev PA, van der Meer M, Rigling A, Shevchenko IG. 2004. Effect of climatic changes on the formation of Siberian spruce generations in subglotsy tree stands of the Southern Urals. Russian Journal of Ecology. 35:135–143. Moles AT, Warton DI, Warman L, Swenson NG, Laffan SW, Zanne AE, Pitman A, Hemmings FA, Leishman MR. 2009. Global patterns in plant height. Journal of Ecology. 97:923–932. Oksanen J, Blanchet FG, Kindt R, Legendre P, Minchin PR, O’Hara B, Simpson GL, Solymos P, Stevens MHH, Wagner HH 2013. Package ‘vegan’, version 2.0-8. Pardo I, Camarero JJ, Gutiérrez E, García MB. 2013. Uncoupled changes in tree cover and field layer vegetation at two Pyrenean treelines over 11 years. Plant Ecology and Diversity. 6:355–364. Pauli H, Gottfried M, Dullinger S, Abdaladze O, Akhalkatsi M, Benito Alonso JL, Coldea G, Dick J, Erschbamer B, Calzado

RF, et al. 2012. Recent plant diversity changes on Europe’s mountain summits. Science. 336:353–355. Pescador DS, de Bello F, Valladares F, Escudero A. 2015. Plant trait variation along an altitudinal gradient in Mediterranean high mountain grasslands: controlling the species turnover effect. PLoS ONE. 10:e0118876. Pinheiro J, Bates D, DebRoy S, Sarkar D, Core Team R. 2015. nlme: linear and nonlinear mixed effects models. R package version 3.1–120. http://CRAN.R-project.org/package=nlme R Core Team. 2015. R: R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. http://www.R-project.org Rasband WS. 2007. ImageJ, U. S. Bethesda, Maryland, USA: National Institutes of Health. http://imagej.nih.gov/ij/. Savage J, Vellend M. 2014. Elevational shifts, biotic homogenization and time lags in vegetation change during 40 years of climate warming. Ecography. 37:1–010. Solly EF, Djukic I, Moiseev PA, Andreyashkina NI, Devi NM, Göransson H, Mazepa VS, Shiyatov SG, Trubina MR, Schweingruber FH, et al. 2017. Treeline advances and associated shifts in the ground vegetation alter fine root dynamics and mycelia production in the South and Polar Urals. Oecologia. 183:571–586. Spasojevic MJ, Suding KN. 2012. Inferring community assembly mechanisms from functional diversity patterns: the importance of multiple assembly processes. Journal of Ecology. 100:652–661. Stubbs WJ, Wilson JB. 2004. Evidence for limiting similarity in a sand dune community. Journal of Ecology. 92:557–567. Sturm M, Schimel J, Michaelson G, Welker JM, Oberbauer SF, Liston GE, Fahnestock J, Romanovsky VE. 2005. Winter biological processes could help convert arctic tundra to shrubland. Bioscience. 55:17–26. Suding KN, Farrer EC, King AJ, Kueppers L, Spasojevic MJ. 2015. Vegetation change at high elevation: scale dependence and interactive effects on Niwot Ridge. Plant Ecology and Diversity. 8:713–725. Venn S, Pickering CM, Green K. 2012. Short-term variation in species richness across an altitudinal gradient of alpine summits. Biodiversity and Conservation. 21:3157–3186. Vittoz P, Randin C, Dutoit A, Bonnet F, Hegg O. 2009. Low impact of climate change on subalpine grasslands in the Swiss Northern Alps. Global Change Biology. 15:209–220. Weiher E, Keddy PA. 1995. Assembly rules, null models, and trait dispersion – new questions front old patterns. Oikos. 74:159–164. Welker JM, Fahnestock JT, Sulliva PS, Chimner RA. 2005. Leaf mineral nutrition of Arctic plants in response to warming and deeper snow in northern Alaska. Oikos. 109:167−177. Westoby M, Wright IJ. 2006. Land-plant ecology on the basis of functional traits. Trends in Ecology and Evolution. 21:261– 268. Wilson SD, Nilsson C. 2009. Arctic alpine vegetation change over 20 years. Global Change Biology. 15:1676–1684. Yang Z, Guo H, Zhang J, Du G. 2013. Stochastic and deterministic processes together determine alpine meadow plant community composition on the Tibetan Plateau. Oecologia. 171:495–504.