SEARCH

SEARCH BY CITATION

Keywords:

  • Biogeography;
  • Cantabrian Range;
  • Community assembly;
  • Environmental filtering;
  • Species diversity;
  • Species pool;
  • Topography

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Author contributions
  10. References
  11. Supporting Information

Questions

Are species from different biogeographic groups (mediterranean, alpine and endemic) filtered in different ways by altitude and topography in alpine plant communities? What is the relative performance of these environmental factors at predicting the species diversity of the communities as a whole and of the geographic species groups?

Location

Picos de Europa, Cantabrian Range (Spain).

Methods

We sampled the presence and cover of vascular plants in 5-m radius plots on alpine grasslands between 1900 and 2500 m a.s.l. Five GIS-based terrain variables at 15 m × 15 m were used to model species richness and cover per plot using generalized and linear models, and the variation in species composition with redundancy analysis. The same analyses were repeated for the whole data set and for subsets of species from alpine, mediterranean and endemic distributions.

Results

The influence of altitude and topography on species richness, cover and composition differed for the whole data set and for the geographic species groups. Altitude was the main variable affecting floristic diversity in the communities as a whole, but the separate species groups were more influenced by slope, topographic wetness index and solar radiation. Richness and cover of mediterranean species showed the strongest relationships with topography. Alpine and endemic species showed relationships with topography for species cover and composition, but not for species richness.

Conclusions

In alpine landscapes, biogeographic deconstruction of the species pool can provide a better understanding of the influence of altitude and topography on local communities than analysis of the entire community alone. Furthermore, the strong influence of local topography on species groups improves our understanding of how alpine species will respond to climate change.


Nomenclature
Flora Iberica

(www.floraiberica.es)

Flora Europaea

(Tutin et al. 1964)

Flora Alpina

(Aeschimann et al. 2004)

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Author contributions
  10. References
  11. Supporting Information

Understanding the patterns and processes that determine the number and composition of co-occurring species is of major interest in community ecology (Vellend 2010). It is generally accepted that the regional species pool is influenced by broad-scale bioregional and evolutionary factors that interact with local factors and give rise to complex ecological processes, such as niche selection, ecological drift, dispersal or speciation (Freschet et al. 2011; Zobel et al. 2011; Hardy et al. 2012). Niche-based approaches stress that the response of plant assemblages to the environment can be viewed as a filter of the regional species pool according to biotic (species interactions) and abiotic (environmental) factors (Keddy 1992; Cingolani et al. 2007). In alpine communities, ′environmental filtering′ seems to be the dominant niche-based process relating climate with species (Kikvidze et al. 2005; de Bello et al. 2012a; Pottier et al. 2012). When alpine habitats are scarcely perturbed, competition is relaxed (Grime 2002; Bruun et al. 2006; Vonlanthen et al. 2006) and the harsh environmental conditions determine a major influence of abiotic filters (Callaway et al. 2002; Kammer & Möhl 2002; Chase & Myers 2011). Environmental filtering of the regional species pool can therefore be expected to exert a strong influence over the diversity and abundance of alpine communities along local ecological gradients. Understanding these patterns is of special interest because these habitats are biodiversity hotspots that are especially sensitive to the effects of climate change (Pauli et al. 2012).

The effect of environmental filters on species assemblages can be assessed with regard to different dimensions of diversity (Keddy 1992). Recently, many studies have focused on phylogenetic (Anderson et al. 2011) or functional traits (de Bello et al. 2012a), providing additional information to the more traditional studies based on taxonomical species names (Swenson 2011). However, a major drawback of these approaches is the difficulty of linking small- and large-scale biogeographic processes, partially since evolutionary history has resulted in different regional species pools (Chase & Myers 2011). Despite the need for more integrative analyses (Weiher et al. 2011) and the general contingencies associated with the community assembly theory (Lessard et al. 2012), it is still necessary to explore the effect of environmental filtering at different diversity levels (de Bello et al. 2012b). One example of an easy-to-apply method based on species information is biogeographic deconstruction, which separately analyses the geographical components of communities along environmental gradients (Real et al. 2008). Deconstruction procedures generally involve the separate analysis of species groups from a given community according to the ecological factors that best explain their diversity (Marquet et al. 2004), and have been used to assess community patterns in species with different habitats (Blamires et al. 2008) or ecological requirements (Azeria et al. 2011). In plant communities, it has been suggested that species with similar distributions respond similarly at large and local ecological gradients (Ferrer-Castán & Vetaas 2003). Since the high species diversity of mountain regions is strongly influenced by environmental filters (de Bello et al. 2012a), they are good models to assess whether species with different geographic ranges respond differently to local environmental factors. Analysing these patterns may provide useful information for understanding species assemblages in alpine communities.

The main abiotic factors affecting the composition of alpine plant communities are pH, altitude, topography and processes relating to soil development and composition (Vonlanthen et al. 2006; Schöb et al. 2008). Since the influence of pH is spatially determined by the dominant bedrock (Virtanen et al. 2002), local variation in species richness is ultimately driven by temperature along narrow altitudinal and topographic gradients (Kammer & Möhl 2002; Bruun et al. 2006). Increasing altitude is generally associated with a decrease in local species richness (Nogués-Bravo et al. 2008), but this trend is weaker at the highest altitudes due to the influence of topographic variation, which locally buffers temperature changes (Scherrer & Körner 2011). One of the most evident effects of topography is found at different exposures on mountain summits, affecting species distribution (Pickering & Green 2009), spatial structure (Gutiérrez-Girón & Gavilán 2010) and plant responses to climate change (Cannone et al. 2007; Pauli et al. 2007). Besides the summit effect, the influence of topography on alpine communities has been traditionally explained through the effect of surface heterogeneity, wind exposure and eco-hydrological settings along the so-called meso-topographic gradient (Billings 1974). In its simplest form, this gradient explains transitions from wind-exposed ridges to small micro-valleys, producing different ecological conditions related to temperature, snow accumulation and soil development (Walker et al. 2001; Choler 2005; Bruun et al. 2006). However, the high complexity of geomorphology and the combined effect of altitude and topography require more sophisticated models to explain the interactions between vegetation and topography in alpine land forms (Bruun et al. 2006).

Using GIS-based models to predict alpine vegetation patterns, Gottfried et al. (1998) highlighted the importance of curvature and roughness as complementary variables to altitude and slope for understanding species richness and the local distribution of species and plant communities. Furthermore, variables such as solar radiation or topographic indices are commonly used to investigate vegetation patterns of species composition in alpine environments (Barrio et al. 1997; Dirnböck et al. 2003a; Pfeffer et al. 2003). More recently, the importance of these variables has been empirically demonstrated by measuring the relationships between temperature and topographic variations (Fridley 2009; Scherrer & Körner 2011), suggesting that fine-scale topographic variables (measured with a resolution of between 1 and 25 m) are robust surrogates of ecological variability in alpine communities (Scherrer et al. 2010). However, few studies compare their ecological significance in diversity estimates, and examples dealing with this topic may shed light on patterns of community assembly along environmental filters.

In this work, we use biogeographic deconstruction to test the hypothesis that species groups from different geographic regions are filtered in different ways by altitude and topography in alpine plant communities. We selected the Cantabrian range (Spain) as the study area, an example of a biogeographic refuge for high-mountain species in southern Europe. This region is influenced by alpine conditions and the mediterranean climate to the south, and supports numerous cold-adapted species that form part of the postglaciation expansion (Taberlet et al. 1998). We anticipated that the alpine species would respond positively to altitude or snow-related topographic conditions, while mediterranean species would prefer different (warmer) conditions. The influence of local environmental gradients on the diversity and composition of these groups may differ from the overall patterns of the whole community, providing a complementary approach for the assessment of alpine community responses. A secondary aim of this study is to explore the relative performance of altitude and topography at predicting diversity patterns in the communities as a whole and in their geographic species groups.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Author contributions
  10. References
  11. Supporting Information

Study area

The study was conducted on the central massif of the Picos de Europa, an isolated alpine landscape in Western Europe with a substantial component of circumboreal mountain species. This calcareous massif comprises the highest elevation in the Cantabrian range (maximum altitude = 2654 m) and one of the main outposts of alpine flora and vegetation in Europe (Nagy 2006). The study was conducted on alpine grasslands that are relatively well developed at altitudes between 1900 and 2400 m, forming inter-connected patches interrupted by steep limestone and dolomite rocks (Fig. 1). Local geomorphology is related to Pleistocene glaciations, the Little Ice Age and current periglacial processes (Gonzalez Trueba et al. 2008; Moreno et al. 2010). The communities studied are biogeographically connected with the European mountain system (Alps, Pyrenees, Carpathians), showing a relatively high similarity with Pyrenean-related vegetation (Nava 1988; Peyre & Font 2011). Besides the presence of generalist non-mountain species, the floristic composition of the study vegetation is mainly represented by arctic–alpine species (e.g. Kobresia myosuroides), species endemic to the mountains of northern Spain (e.g. Armeria cantabrica) and high-mountain mediterranean species (e.g. Senecio boissieri), indicating historic biogeographical connections with the mountains of the Iberian Peninsula (Peredo et al. 2009).

image

Figure 1. Study area and topographic landscape. (a) Location of the Picos de Europa National Park in Spain (SP), and (b) the calcareous massif representing the study area. Red points and green patches (c) represent the distribution of the sample plots and the approximate cover of alpine grasslands, respectively.

Download figure to PowerPoint

Field sampling and species data

We used a 1:10 000 scale vegetation map of the study area (Jiménez-Alfaro & Bueno 2008) to create a systematic sampling design in the patches identified as alpine grasslands (Fig. 1). To encompass much of the floristic diversity of the alpine grasslands, we designed nine latitudinal and longitudinal field transects covering the range of the vegetation in the study area. Fixed area plots with a radius of 5 m (size = 78 m2) were established at an approximate distance of 250 m from one another to avoid spatial autocorrelation. In the centre of each plot we recorded the geographical co-ordinates using a field GPS (Garmin e-trex, error = 4 m). Field sampling was conducted during the growing season (July–August) in 2008 and 2009 until we obtained a wide, normally distributed series of altitudinal and topographic ranges covering the whole study area. A total of 101 plots (Fig. 1) were sampled, for which we recorded the total number of vascular plants and estimated their cover using a seven point species cover–abundance scale (Braun-Blanquet 1964). In order to minimize bias in plant identification and cover estimates (Vittoz & Guisan 2007), at least two observers with a good knowledge of the local flora conducted sampling in all the plots. Since there is a predominance of long-lived perennials in the vegetation studied, significant floristic changes are not expected from year to year.

From the total species list, we identified the species whose distribution is restricted to high-mountain habitats (alpine sensu Körner et al. 2011), and classified them into the three predominant geographic groups represented in the study area: (1) alpine species (e.g. Saxifraga oppositifolia) mainly distributed (>95% of their known distribution) throughout the Alpine Biogeographic Region (Roekaerts 2002), and including arctic–alpine species present in the European mountain system and northern (arctic) latitudes; (2) mediterranean species (e.g. Festuca hystrix) mainly distributed throughout the Mediterranean mountains of the Iberian Peninsula, or in some cases with a wider distribution throughout the Mediterranean mountains (Italy, northern Africa, southeast France); (3) species endemic to the Cantabrian range (e.g. Saxifraga conifera) or the Cantabro-Pyrenaean mountains (e.g. Euphorbia pyrenaica) for which the study area represents an important refuge. Species with other distributions or not consistently associated with high-mountain vegetation were not analysed separately, but were included in the analyses of the whole community. Nomenclature and species distribution were reviewed according to Flora Iberica (www.floraiberica.es), Flora Europaea (Tutin et al. 1964) and Flora Alpina (Aeschimann et al. 2004).

Environmental variables

We used a geographic information system (ArcGIS 9.2, ESRI Inc., Redlands CA, US) to generate topographic variables commonly used for vegetation modelling (van Niel et al. 2004). A digital elevation model (DEM) at 15 m × 15 m (a scale fitted to the plot size and the GPS error) was derived from 1:10 000 digital cartography of the study area. Altitude (min. 1927 m; max. 2496 m; mean = 2178 m) and slope (min. 0°; max. 45°; mean = 21°) were derived from the DEM using the Spatial Analyst extension in ArcGIS. Solar radiation (WM2, min. 86.5; max. 179.0; mean = 145.0) was calculated from altitude, exposure and solar trajectory using mean annual global radiation and intermediate (0.5) values of light transmittance according to the Solar Analyst utility (Fu & Rich 2000). The topographic position index (TPI) was generated following ArcView (Topographic Position Index extension for ArcView 3.x, v. 1.3a), such that valleys <0, flats ~ 0 and ridge tops >1 (min. −5.3; max. 17.7; mean = 2.0). To match this index to the scale of sampling, we used a circular neighbourhood with a four-cell radius, meaning that the TPI value reflects the difference between the elevation of the target cell and the average elevation of all cells within 60 m from that cell. Finally, a topographic wetness index (TWI) was calculated using the algorithm of Beven & Kirkby (1979), TWI = ln(a/tanβ), which is based on the upslope area per unit contour length (a) and the local slope (tanβ). This equation assumes that topography controls the movement of water and thus the spatial pattern of soil moisture (Schmidt & Persson 2003), and provides an estimate of local water flows and snow accumulation (Essery & Pomeroy 2004). TWI (min. 0.9; max 5.3; mean = 2.5) was calculated with the TOPOCROP Arcview Extension (Schmidt & Persson 2003) using the neighbourhood statistics (mean filter) to even out major contrasts in values and to minimize small-scale heterogeneity. Although we assessed other topographic predictors, such as curvature and exposure (northernness, southernness), a preliminary pair-wise test (Pearson's r) revealed high (>0.8) correlations for these variables, and only those predictors with expected higher relevance and lower correlations were selected.

Data analysis

We used generalized linear models (GLM) and linear models (LM) to analyse the influence of environmental variables on species richness and plant cover in the whole data set and the subsets of mediterranean, alpine and endemic species. The most appropriate function for modelling the data was selected according to the lowest residual deviance and different data transformations. The number of species per plot was analysed as a Poisson-distributed random variable fitted with a logarithmic link function and GLM (McCullagh & Nelder 1989). The total percentage cover per plot was estimated by transformation of the ordinal cover of all species to a continuous scale from 0 to 100, assuming random overlap of species (Chytrý et al. 2010), as implemented in the JUICE software. Cover estimates per plot (ranging from 0.2 to 0.8) were logit-transformed to obtain normality (Warton & Hui 2011) and modelled using LM. For both species richness and cover measures, we used a forward step-wise model procedure to select the best predictors according to the Akaike information criterion (AIC), considering all variables and their interactions. The significance of predictors was calculated by means of a deviance test (Nicholls et al. 1991) and only those with a significant (α = 0.05) increase in residual deviance (or variance) were retained, determined by χ2 (in GLM) and F-tests (in LM). Lastly, we estimated the cumulative percentage deviance explained as a measure of the lack of fit of the data to the model. Analyses were conducted with R 2.14.1 (2003; R Foundation for Statistical Computing, Vienna, AT). The contribution of the significant variables was assessed using effect-plot displays in R (Fox 2003). Spatial autocorrelation of the response variables and the model residuals was tested using Moran's (I) statistic (Fortin & Dale 2005) and a Z score to test the null hypothesis that there is no spatial clustering, as implemented in ArcGIS 9.2.

We used ordination methods to assess the variation in species composition along altitudinal and topographic gradients. An exploratory analysis of the data using detrended correspondence analysis (DCA; length of first axis = 2.58 SD units) suggested a linear response by the species (Lepš & Šmilauer 2005). We therefore used redundancy analysis (RDA) to test the null hypothesis that species composition and their distribution are independent of environmental variables (Legendre & Anderson 1999). The effect of each variable was first assessed in a separate RDA using a Monte Carlo permutation test with 999 permutations. The most significant predictors were selected using a forward selection procedure, adding new variables in order of their decreasing eigenvalues until the variables were non-significant (> 0.05). Biplots of significant predictors and species were generated. When only one predictor was considered, species fit was calculated solely for the first axis. The analyses were computed with CANOCO 4.5 (Biometris, Wageningen, NL) using square-root transformed cover values as abundances. Different RDAs were computed for the whole data set and for the different geographic species groups.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Author contributions
  10. References
  11. Supporting Information

We recorded a total number of 164 vascular plant species, of which 44 were alpine, 23 mediterranean and 31 endemic. The remaining 66 species were identified as widely distributed or not restricted to mountain habitats (see full list in Appendix S1). The number of species per plot varied between 11 and 39 (22.3 ± 6.7; mean ± SD). Of the three geographic species groups, the alpine group contributed the most species on average (7.9 ± 2.6), followed by the endemic group (5.5 ± 2.4) and the mediterranean group (3.2 ± 1.7). The vascular plant cover estimated per plot varied between 35% and 81% (59.6 ± 10.6), with higher cover values for alpine (40.8 ± 14.3) compared to endemic (27.9 ± 12.4) and mediterranean (20.2 ± 11.3) species.

As expected by the sampling procedure, neither the response variables nor the model residuals were spatially autocorrelated (Moran's I; > 0.05). GLM analysis revealed that total species richness is negatively associated with altitude and TWI (Fig. 2), explaining 14% and 11% of the deviance, respectively (Table 1). From the three species subsets, only the mediterranean group showed significant relationships between species richness and the predictor variables, with a negative effect of slope (explaining 8% of variance) and TWI (an additional 8% of explained variance) (Fig. 2). According to the LMs, the species cover of the whole species data set was negatively associated with altitude (Table 2), which explained 17.5% of variance. In contrast with the patterns observed for species richness, species cover was significantly associated with predictors in all the geographic groups. The lowest values of total explained variance were found in the alpine subset (9.9% explained, with a negative effect of altitude and a positive effect of slope) and the endemic subset (10.9%, negative effect of TPI and positive effect of slope). The cover of mediterranean species was better explained (23.2%) than in the other groups, with negative relationships for slope and TWI.

Table 1. Generalized linear models fitted for species richness. Only the parameters computed over the whole data set and the species geographic groups with significant relationships (mediterranean) are shown. Variables are ranked according to their importance as predictors. Exp Dev indicates the cumulative percentage of deviance explained by the variables
ParameterRes. dev χ 2 P Exp devAIC
All species richness
Null199.03    
1. Altitude171.4548.1<0.00114%671
2. TWI149.8421.2<0.00125%651
Mediterranean species richness
Null100.40    
1. Slope92.2614.80.0048%387
2. TWI84.307.80.00516%381
Table 2. Linear models fitted for species cover (logit transformed). Results show the parameters of different models computed over the whole data set and species groups with only alpine, mediterranean or endemic species. Variables are ranked according to their importance as predictors. Exp Var indicates the cumulative percentage of variance explained by the variables
ParameterRes varF P Exp varAIC
All species cover
Null79.73    
1. Altitude66.0521.36<0.00117.5%250
Alpine species cover
Null123.87    
1. Altitude118.376.200.0314.5%309
2. Slope111.615.920.0179.9%305
Mediterranean species cover
Null89.91    
1. Slope74.6521.49<0.00116.9%262
2. TWI69.557.180.00823.2%257
Endemic species cover
Null97.93    
1. TPI90.648.200.0057.4%282
2. Slope87.163.910.05010.9%280
image

Figure 2. Effect plots of the models computed for species richness (a) and logit-transformed cover (b). Plots represent the effect of each factor in the generalized linear (a) and linear (b) models computed over the whole data set (All species) and the subsets of alpine, mediterranean and endemic species. A 95% confidence interval (grey lines) is drawn around the estimated effect. Altitude in meters, Slope in degrees, Topographic wetness index (TWI) and position index (TPI) are explained in the text.

Download figure to PowerPoint

Redundancy analysis (RDA) showed a significant response of species composition to environmental factors (Table 3). The total explained variance was higher for the whole data set than for the separate species groups, due in part to the higher number of explanatory variables that are included in the model. The species–environment correlation was 0.75 for the first axis and 0.68 for the second axis (76% of the total cumulative variance), and altitude was the most important of the four selected variables. However, the results of species–environment relationships differed for the three species groups. Slope was the only significant variable for the alpine subset (correlation = 0.52 for axis 1); altitude and solar radiation partially explained variation in the mediterranean subset (correlation = 0.58 for axis 1 and 0.36 for axis 2); and solar radiation, slope and altitude were the factors influencing the endemic subset (correlation = 0.47 for axis 1 and 0.58 for axis 2).

Table 3. Redundancy analysis (RDA) for altitude and topographic variables. Results show the statistics of partial RDAs computed over the whole data set (all species) and species groups with only alpine, mediterranean or endemic species. Only the significant variables selected by the forward selection procedure (α = 0.05) are shown. Lambda (λ) indicates the variance explained by each variable individually. Exp Var indicates the cumulative percentage of variance explained by the model
Parameterλ F-ratio P Exp var
All species
1. Altitude0.0808.590.0028.0%
2. Solar radiation0.0024.270.00211.8%
3. TWI0.0023.400.00214.8%
4. Slope0.0303.280.00217.6%
Alpine species
1. Slope0.0606.500.0026.2%
Mediterranean species
1. Altitude0.0505.590.0025.5%
2. Solar radiation0.0302.650.0268.1%
Endemic species
1. Solar radiation0.0504.990.0044.8%
2. Slope0.0302.920.0067.6%
3. Altitude0.0202.220.0309.7%

Species–environment plots from the RDA of the whole data set (Fig. 3) indicate that 13 species are associated with increasing altitude (e.g. Galium pyrenaicum) or TWI (e.g. Koeleria vallesiana) and decreasing solar radiation (e.g. Sesleria albicans). In the geographic species groups, the number of species fitting with the explained variance was lower. Main species indicators were negatively associated to slope in the alpine subset (e.g. Pritzelago alpina) and to altitude in the mediterranean subset (e.g. Festuca hystrix). The species composition of endemic species showed a more complex relationship with environmental variables, with solar radiation and slope partially explaining the distribution of up to five species (e.g. Armeria cantabrica).

image

Figure 3. Species–environment plots from the RDA. The plots include the significant variables selected by forward selection (α = 0.05) in the whole data set (All) and the subsets of alpine, mediterranean and endemic species. Only the species with a fit range higher than the percentage variation explained are plotted. Review alphabetic order.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Author contributions
  10. References
  11. Supporting Information

Environmental filtering of geographic species groups

This study demonstrates that biogeographic deconstruction produces species groups with predictable responses to environmental filters in alpine communities. In particular, our separation of geographic species groups provides information about the species' relationships with topographic gradients that may be neglected when using data for the whole community. Although we cannot say whether this pattern could be repeated in other habitats, biogeographic deconstruction as performed here seems to be an informative method for analysing the diversity of alpine communities with mixed floristic composition. This procedure is generally used on large scales (Real et al. 2008; Fløjgaard et al. 2011) but we show that it is also feasible to link local ecological patterns with the biogeographical context of the species pool, following the expectations of the community assembly theory (Ricklefs 2004). Moreover, our results support the general expectation of the species pool hypothesis regarding links between the abundance of species in one habitat and their habitat preferences throughout evolutionary history (Zobel et al. 2011). Biogeographic deconstruction also agrees with similar attempts to reconcile geographic and local ecological factors in order to interpret community assembly across multiple levels of diversity (Hardy et al. 2012).

Since our study system is rarely grazed by domestic stock or impacted by other human activities, we assume that niche-based patterns are mainly driven by abiotic factors. Although the data are restricted to one region and do not consider other assembly processes (Spasojevic & Suding 2012), the observed pattern suggests a deterministic selection from the regional species pool (phylogeographic assembly) in local communities (ecological assembly). For example, the combined influence of non-slope surfaces and TWI on mediterranean species richness may be related to the inhibitory effect of cryoturbation (repetition of freezing and thawing processes in periglacial zones) on the development of alpine soils (Amico & Previtali 2012), indicating micro-habitat refuges for these species in the context of a non-Mediterranean mountain. Moreover, the fact that alpine species cover was negatively influenced by altitude but positively by slope may be explained by their preference for snow patches, which are more commonly found at low or medium altitudes in the study area. Thus, mediterranean species seem to be restricted to xeric and sunnier micro-habitats, similar to those described in the calcareous grasslands of the nearby Pyrenees (Sebastiá 2004); while alpine species are more closely related to steep topographies with high snow accumulation, more commonly found in temperate mountains (Nagy & Grabherr 2009). Similarly, the influence of TPI and slope on the cover of endemic species suggests a preference for micro-valleys with high snow cover. Nevertheless, endemic species showed different environmental relationships to the other geographic groups, in agreement with studies indicating that mountain endemics from calcareous bedrock have distinct ecological patterns (Essl et al. 2009).

General influence of altitude and topography

When analysing the communities as a whole, we found that altitude is the main abiotic filter influencing species richness, cover and composition. The decrease in species richness with increasing altitude is in agreement with the pattern expected in narrow elevational ranges (Nogués-Bravo et al. 2008), and is widely attributed to the temperature decrease in similar habitats (Bruun et al. 2006; Qiong et al. 2010). However, we found that the effect of altitude is reduced and the effect of topography is increased when mountain species are analysed separately. This may be explained by the exclusion of generalist species that are not restricted to high mountains (40%) and that can be expected to be more influenced by the elevational gradient (Bruun et al. 2006). The relative abundance of generalist species in the study area may also reflect former shifts from low altitudes in response to past or recent climate changes, as has been indicated in other regions (Choler et al. 2001). In addition, our results suggest that shifts of high-mountain species can be driven by topographic rather than elevation gradients, and those factors should also be considered to assess changes in species richness and composition in alpine communities.

Total species richness was also influenced by TWI, which is consistent with findings in dry alpine (Dirnböck et al. 2003ab) and arctic (Ostendorf & Reynolds 1998) communities. The fact that TWI had a negative effect on species richness but not on total plant cover suggests that it may determine special conditions that limit the presence of some plant species. Topographic wetness index has been used to explain variation in species diversity in highly contrasting habitats, as a surrogate for water flow accumulation (Vittoz et al. 2009), soil saturation (Gessler et al. 1995) and periglacial processes such as solifluction sheets (Hjort et al. 2007) or patterned ground features (Feuillet 2011). In our study, the absence of surface water flows and the periglacial structure of the landscape suggest that the decrease in species richness with high values of TWI may be better explained by processes such as cryoturbation that limit soil formation and nutrient availability in cold environments (Celi et al. 2010).

Altitude and topography were also good predictors of species composition in the whole data set. The relatively low number of species with a good fit to the constrained (RDA) axes suggests that the pattern of species composition is governed by multiple gradients, and therefore few species can be considered indicators of only one predictor. Among these species we found arctic–alpine species (e.g. Saxifraga oppositifolia) and local endemics (e.g. Jasione cavallinesi) to be good indicators of high altitudes, and characteristic species of alpine European grasslands (e.g. Carex sempervirens) to be indicators of low solar radiation. Moreover, one mediterranean species that is closely associated with TWI (Koeleria vallesiana) has elsewhere been associated with extremely low productivity soils (Martinez-Duro et al. 2011), supporting the idea that this variable indicates poorly developed soils. These results agree with the general patterns detected for species richness and cover, and provide examples of the correlation between abiotic factors and the geographic species groups. Furthermore, the consistent effect of topographic variables across different groups supports the importance of these factors in the study system. Surprisingly, we found minimal or no effect of TPI, despite this variable being directly related to ridge top transitions along topographic gradients (Billings 1974). In contrast, our results suggest a strong influence of TWI, although more detailed studies are needed to understand the relationship between TWI values (Kopecký & Čížková 2010) and related factors (e.g. temperature or soil development) in alpine communities.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Author contributions
  10. References
  11. Supporting Information

Overall, our results demonstrate that the diversity patterns of alpine communities across local environmental gradients may be partially determined by the ecological similarity of mountain species with shared biogeographic history. Explicitly accounting for biogeography may assist in assessing the responses of plant communities to climate changes along temperature-related gradients, and more studies are required to test similar patterns in different regions. In our study, high-mountain species seem to be more influenced by the topographic factors that determine thermal differences in alpine landscapes (Scherrer & Körner 2011) than generalist species. Analysing similar patterns in other regions could at least partially explain the decline of cold-adapted species and the increase of warm-adapted species detected in the European mountains (Gottfried et al. 2012), especially in those regions with a complex biogeographic history. For example, arctic–alpine species are expected to find topographic refuges in southern regions where they have become rare (Jiménez-Alfaro et al. 2012), while mediterranean species are subjected to local extinction in alpine environments (Pauli et al. 2012). Given the recent interest in measuring the effects of climate change along topographic gradients (Fridley 2009; Scherrer et al. 2010), we conclude that biogeographic deconstruction may provide a useful tool for the separate assessment of the performance of high-mountain species in alpine communities.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Author contributions
  10. References
  11. Supporting Information

This study was funded by the Fundación Biodiversidad and the Spanish Ministry of Science (BIOALPI, CGL2008-00901). We would like to thank the Picos de Europa National Park for technical support and INDUROT in Oviedo University for the project management. BJA was partially supported by a post-doctoral grant from the REMEDINAL2 network project (Madrid Autonomous region, S-0505-AMB1783). CM has benefited from a research stay from the Universitá degli Studi di Palermo. We thank David Zelený for reviewing the manuscript and two anonymous reviewers for their helpful comments.

Author contributions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Author contributions
  10. References
  11. Supporting Information

The idea was conceived by BJA, who collected and analysed the data; CM collected data and compiled the database; AB collected data; RG and JRO participated in the experimental design and commented on the manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Author contributions
  10. References
  11. Supporting Information
  • Aeschimann, D., Lauber, K., Moser, D. & Theurillat, J.-P. 2004. Flora Alpina. vols I-III, Haupt, Bern, CH.
  • Amico, M.E.D. & Previtali, F. 2012. Edaphic influences of ophiolitic substrates on vegetation in the Western Italian Alps. Plant and Soil 351: 7395.
  • Anderson, T.M., Shaw, J. & Olff, H. 2011. Ecology's cruel dilemma, phylogenetic trait evolution and the assembly of Serengeti plant communities. Journal of Ecology 99: 797806.
  • Azeria, E.T., Bouchard, M., Pothier, D., Fortin, D. & Hébert, C. 2011. Using biodiversity deconstruction to disentangle assembly and diversity dynamics of understory plants along post-fire succession in boreal forest. Global Ecology and Biogeography 20: 119133.
  • Barrio, G., Alvera, B., Puigdefabregas, J. & Diez, C. 1997. Response of high mountain landscape to topographic variables: central pyrenees. Landscape Ecology 12: 95115.
  • de Bello, F., Lavorel, S., Lavergne, S., Albert, C.H., Boulangeat, I., Mazel, F. & Thuiller, W. 2012a. Hierarchical effects of environmental filters on the functional structure of plant communities: a case study in the French Alps. Ecography 35: 110.
  • de Bello, F., Price, J.N., Münkemüller, T., Liira, J., Zobel, M., Thuiller, W., Gerhold, P., Götzenberger, L., Lavergne, S., Lepš, J., Zobel, K. & Pärtel, M. 2012b. Functional species pool framework to test for biotic effects on community assembly. Ecology 93: 22632273.
  • Beven, K. & Kirkby, M. 1979. A physically based, variable contributing area model of basin hydrology. Hydrological Sciences Journal 24: 4369.
  • Billings, W.D. 1974. Adaptations and origins of alpine plants. Arctic and Alpine Research 6: 129142.
  • Blamires, D., de Oliveira, G., de Souza Barreto, B. & Diniz-Filho, J.A.F. 2008. Habitat use and deconstruction of richness patterns in Cerrado birds. Acta Oecologica 33: 97104.
  • Braun-Blanquet, J. 1964. Pflanzensoziologie. Springer, Berlin, DE.
  • Bruun, H.H., Moen, J., Virtanen, R., Grytnes, J.-A., Oksanen, J. & Angerbjörn, A. 2006. Effects of altitude and topography on species richness of vascular plants, bryophytes and lichens in alpine communities. Journal of Vegetation Science 17: 3746.
  • Callaway, R.M., Brooker, R.W., Choler, P., Kikvidze, Z., Lortie, C.J., Michalet, R., Paolini, L., Pugnaire, F.I., Newingham, B., Aschehoug, E.T., Armas, C., Kikodze, D. & Cook, B.J. 2002. Positive interactions among alpine plants increase with stress. Nature 417: 844848.
  • Cannone, N., Sgorbati, S. & Guglielmin, M. 2007. Unexpected impacts of climate change on alpine vegetation. Frontiers in Ecology and the Environment 5: 360364.
  • Celi, L., Rosso, F., Freppaz, M., Agnelli, A. & Zanini, E. 2010. Soil organic matter characteristics in sporadic permafrost-affected environment Creux du Van, Switzerland. Arctic, Antarctic, and Alpine Research 42: 18.
  • Chase, J.M. & Myers, J. 2011. Disentangling the importance of ecological niches from stochastic processes across scales. Philosophical Transactions of the Royal Society of London. Series B, Biological sciences 366: 23512363.
  • Choler, P. 2005. Consistent shifts in alpine plant traits along a mesotopographical gradient. Arctic, Antarctic, and Alpine Research 37: 444453.
  • Choler, P., Michalet, R. & Callaway, M. 2001. Facilitation and competition on gradients in alpine plant communities. Ecology 82: 32953308.
  • Chytrý, M., Danihelka, J., Axmanová, I., Božková, J., Hettenbergerová, E., Li, C.-F., Rozbrojová, Z., Sekulová, L., Tichý, L., Vymazalová, M. & Zelený, D. 2010. Floristic diversity of an eastern Mediterranean dwarf shrubland: the importance of soil pH. Journal of Vegetation Science 21: 11251137.
  • Cingolani, A.M., Cabido, M., Gurvich, D.E., Renison, D. & Díaz, S. 2007. Filtering processes in the assembly of plant communities: are species presence and abundance driven by the same traits? Journal of Vegetation Science 18: 911920.
  • Dirnböck, T., Dullinger, S., Gottfried, M., Ginzler, C. & Grabherr, G. 2003a. Mapping alpine vegetation based on image analysis, topographic variables and canonical correspondence analysis. Applied Vegetation Science 6: 8596.
  • Dirnböck, T., Dullinger, S. & Grabherr, G. 2003b. A regional impact assessment of climate and land-use change on alpine vegetation. Journal of Biogeography 30: 401417.
  • Essery, R. & Pomeroy, J. 2004. Vegetation and topographic control of wind-blown snow distributions in distributed and aggregated simulations for an Arctic tundra basin. Journal of Hydrometeorology 5: 735744.
  • Essl, F., Staudinger, M., Stöhr, O., Schratt-Ehrendorfer, L., Rabitsch, W. & Niklfeld, H. 2009. Distribution patterns, range size and niche breadth of Austrian endemic plants. Biological Conservation 142: 25472558.
  • Ferrer-Castán, D. & Vetaas, O.R. 2003. Floristic variation, chorological types and diversity: do they correspond at broad and local scales? Diversity and Distributions 9: 221235.
  • Feuillet, T. 2011. Statistical analyses of active patterned ground occurrence in the taillon massif pyrénées, France/Spain. Permafrost and Periglacial Processes 22: 228238.
  • Fløjgaard, C., Normand, S., Skov, F. & Svenning, J.-C. 2011. Deconstructing the mammal species richness pattern in Europe – towards an understanding of the relative importance of climate, biogeographic history, habitat heterogeneity and humans. Global Ecology and Biogeography 20: 218230.
  • Fortin, M.-J. & Dale, M.R.T. 2005. Spatial analysis: a guide for ecologists. Cambridge University Press, Cambridge, UK.
  • Fox, J. 2003. Effect displays in R for generalised linear models. Journal of Statistical Software 8: 127.
  • Freschet, G.T., Dias, A.T.C., Ackerly, D.D., Aerts, R., van Bodegom, P.M., Cornwell, W.K., Dong, M., Kurokawa, H., Liu, G., Onipchenko, V.G., Ordoñez, J.C., Peltzer, D.A., Richardson, S.J., Shidakov, I.I., Soudzilovskaia, N.A., Tao, J. & Cornelissen, J.H.C. 2011. Global to community scale differences in the prevalence of convergent over divergent leaf trait distributions in plant assemblages. Global Ecology and Biogeography 20: 755765.
  • Fridley, J.D. 2009. downscaling climate over complex Terrain: high finescale < 1000 m spatial variation of near-ground temperatures in a Montane Forested landscape Great Smoky Mountains. Journal of Applied Meteorology and Climatology 48: 10331049.
  • Fu, P. & Rich, P. 2000. Manual Solar Analyst 1.0. Helios Environmental Modelling Institute (HEMI), KS, US.
  • Gessler, P.E., Moore, I.D., McKenzie, N.J. & Ryan, P.J. 1995. Soil–landscape modelling and spatial prediction of soil attributes. International Journal of Geographical Information Science 9: 421432.
  • Gonzalez Trueba, J.J., Moreno, R.M., Martinez de Pison, E. & Serrano, E. 2008. ‘Little Ice Age’ glaciation and current glaciers in the Iberian Peninsula. The Holocene 18: 551568.
  • Gottfried, M., Pauli, H. & Grabherr, G. 1998. Prediction of vegetation patterns at the limits of plant life: a new view of the Alpine–Nival Ecotone. Arctic and Alpine Research 30: 207221.
  • Gottfried, M., Pauli, H., Futschik, A., Akhalkatsi, M., Barančok, P., Benito Alonso, J.L., Coldea, G., Dick, J., Erschbamer, B., Fernández Calzado, M.R., Kazakis, G., Krajči, J., Larsson, P., Mallaun, M., Michelsen, O., Moiseev, D., Moiseev, P., Molau, U., Merzouki, A., Nagy, L., Nakhutsrishvili, G., Pedersen, B., Pelino, G., Puscas, M., Rossi, G., Stanisci, A., Theurillat, J.-P., Tomaselli, M., Villar, L., Vittoz, P., Vogiatzakis, I. & Grabherr, G. 2012. Continent-wide response of mountain vegetation to climate change. Nature Climate Change 2: 111115.
  • Grime, P. 2002. Plant Strategies, Vegetation Processes, and Ecosystem Properties. 2nd ed. Wiley, Chichester, UK.
  • Gutiérrez-Girón, A. & Gavilán, R.G. 2010. Spatial patterns and interspecific relations analysis help to better understand species distribution patterns in a Mediterranean high mountain grassland. Plant Ecology 210: 137151.
  • Hardy, O.J., Couteron, P., Munoz, F., Ramesh, B.R. & Pélissier, R. 2012. Phylogenetic turnover in tropical tree communities: impact of environmental filtering, biogeography and mesoclimatic niche conservatism. Global Ecology and Biogeography 21: 10071016.
  • Hjort, J., Luoto, M. & Seppälä, M. 2007. Landscape scale determinants of periglacial features in subarctic Finland: a grid-based modelling approach. Permafrost and Periglacial Processes 18: 115127.
  • Jiménez-Alfaro, B. & Bueno, Á. 2008. Cartografia de Vegetación del Parque Nacional Picos de Europa (Hoja 55-3-4). Organismo Autónomo Parques Nacionales – Jardín Botánico Atlántico, ES.
  • Jiménez-Alfaro, B., Draper, D. & Nogués-Bravo, D. 2012. Modeling the potential area of occupancy at fine resolution may reduce uncertainty in species range estimates. Biological Conservation 147: 190196.
  • Kammer, P. & Möhl, A. 2002. Factors controlling species richness in alpine plant communities: an assessment of the importance of stress and disturbance. Arctic, Antarctic, and Alpine Research. 34: 398407.
  • Keddy, P. 1992. Assembly and response rules: two goals for predictive community ecology. Journal of Vegetation Science 3: 157164.
  • Kikvidze, Z., Pugnaire, F.I., Brooker, R.W., Choler, P., Lortie, C.J., Michalet, R. & Callaway, R.M. 2005. Linking patterns and processes in alpine plant communities: a global study. Ecology 86: 13951400.
  • Kopecký, M. & Čížková, Š. 2010. Using topographic wetness index in vegetation ecology: does the algorithm matter? Applied Vegetation Science 13: 450459.
  • Körner, C., Paulsen, J. & Spehn, E.M. 2011. A definition of mountains and their bioclimatic belts for global comparisons of biodiversity data. Alpine Botany 121: 7378.
  • Legendre, P. & Anderson, M.J. 1999. Distance-Based redundancy analysis: testing multispecies responses in multifactorial ecological experiments. Ecological Monographs 69: 124.
  • Lepš, J. & Šmilauer, P. 2005. Multivariate Analysis of Ecological Data using CANOCO. Bulletin of the Ecological Society of America 86: 66.
  • Lessard, J.-P., Belmaker, J., Myers, J.A., Chase, J.M. & Rahbek, C. 2012. Inferring local ecological processes amid species pool influences. Trends in Ecology & Evolution, 27: 600607.
  • Marquet, P., Fernández, M., Navarrete, N. & Valdovinos, C. 2004. Diversity emerging: towards a deconstruction of biodiversity patterns. In: Lomolino, M. & Heaney, L. (eds.). Frontiers of Biogeography: New directions in the Geography of Nature, pp. 191209. Sinauer, Sunderland, MA, US.
  • Martinez-Duro, E., Luzuriaga, A.L., Ferrandis, P., Escudero, A. & Herranz, J.M. 2011. Does aboveground vegetation composition resemble soil seed bank during succession in specialized vegetation on gypsum soil? Ecological Research 27: 4351.
  • McCullagh, P. & Nelder, J. 1989. Generalized linear models. 2nd ed. Chapman & Hall/CRC, Boca Raton, FL, US.
  • Moreno, A., Valero-Garcés, B.L., Jiménez-Sánchez, M., Domínguez-Cuesta, M.J., Mata, M.P., Navas, A., González-Sampériz, P., Stoll, H., Farias, P., Morellón, M., Corella, J.P. & Rico, M. 2010. The last deglaciation in the Picos de Europa National Park Cantabrian Mountains, northern Spain. Journal of Quaternary Science 25: 10761091.
  • Nagy, L. 2006. European high mountain alpine. vegetation and its suitability for indicating climate change impacts. Biology and Environment: Proceedings of the Royal Irish Academy 106: 335341.
  • Nagy, L. & Grabherr, G. 2009. The Biology of Alpine Habitats. Oxford University Press, Oxford, UK.
  • Nava, H.S. 1988. Flora y vegetación orófila de los Picos de Europa. Ruizia 6: 1243.
  • Nicholls, A.O., Margules, C.R. & Austin, M.P. 1991. An Introduction to Statistical Modelling using GLIM. In: Margules, C.R. & Austin, M.P. (eds.) Nature Conservation: Cost Effective Biological Surveys and Data Analysis, pp. 191201. CSIRO, Canberra, AU.
  • van Niel, K.P., Laffan, S.W. & Lees, B.G. 2004. Effect of error in the DEM on environmental variables for predictive vegetation modelling. Journal of Vegetation Science 15: 747756.
  • Nogués-Bravo, D., Araújo, M.B., Romdal, T. & Rahbek, C. 2008. Scale effects and human impact on the elevational species richness gradients. Nature 453: 216219.
  • Ostendorf, B. & Reynolds, J.F. 1998. A model of arctic tundra vegetation derived from topographic gradients. Landscape Ecology 13: 187201.
  • Pauli, H., Gottfried, M., Reiter, K., Klettner, C. & Grabherr, G. 2007. Signals of range expansions and contractions of vascular plants in the high Alps: observations 1994–2004 at the GLORIA master site Schrankogel, Tyrol, Austria. Global Change Biology 13: 147156.
  • Pauli, H., Gottfried, M., Dullinger, S., Abdaladze, O., Akhalkatsi, M., Benito Alonso, J.L., Coldea, G., Dick, J., Erschbamer, B., Fernández Calzado, R., Ghosn, D., Holten, J.I., Kanka, R., Kazakis, G., Kollár, J., Larsson, P., Moiseev, P., Moiseev, D., Molau, U., Molero Mesa, J., Nagy, L., Pelino, G., Puşcaş, M., Rossi, G., Stanisci, A., Syverhuset, A.O., Theurillat, J.P., Tomaselli, M., Unterluggauer, P., Villar, L., Vittoz, P. & Grabherr, G. 2012. Recent plant diversity changes on Europe's mountain summits. Science 336: 353355.
  • Peredo, E.L., Revilla, M., Jiménez-Alfaro, B., Bueno, Á., Prieto, J. & Abbott, R.J. 2009. Historical biogeography of a disjunctly distributed, Spanish alpine plant, Senecio boissieri (Asteraceae). Taxon 58: 883892.
  • Peyre, G. & Font, X. 2011. Syntaxonomic revision and floristic characterization of the phytosociological alliances corresponding to subalpine and alpine grasslands of the Pyrenees and Cantabrian Mountains classes Caricetea curvulae, Carici-Kobresietea, and Festuco-Seslerietea. Plant Biosystems 145: 220232.
  • Pfeffer, K., Pebesma, E.J. & Burrough, P.A. 2003. Mapping alpine vegetation using vegetation observations and topographic attributes. Landscape Ecology, 75: 9776.
  • Pickering, C.M. & Green, K. 2009. Vascular plant distribution in relation to topography, soils and micro-climate at five GLORIA sites in the Snowy Mountains, Australia. Australian Journal of Botany 57: 189199.
  • Pottier, J., Dubuis, A., Pellissier, L., Maiorano, L., Rossier, L., Randin, C.F., Vittoz, P. & Guisan, A. 2012. The accuracy of plant assemblage prediction from species distribution models varies along environmental gradients. Global Ecology and Biogeography 22: 5263.
  • Qiong, L., Grytnes, J.-A. & Birks, H.J.B. 2010. Alpine vegetation and species-richness patterns along two altitudinal gradients in the Gyama Valley, south-central Tibet, China. Plant Ecology & Diversity 3: 235247.
  • Real, R., Olivero, J. & Vargas, J.M. 2008. Using chorotypes to deconstruct biogeographical and biodiversity patterns: the case of breeding waterbirds in Europe. Global Ecology and Biogeography 17: 735746.
  • Ricklefs, R.E. 2004. A comprehensive framework for global patterns in biodiversity. Ecology Letters 7: 115.
  • Roekaerts, M. 2002. The Biogeographical Regions Map of Europe. Basic principles of its creation and overview of its development. European Environmental Agency, Brussels, BE.
  • Scherrer, D. & Körner, C. 2011. Topographically controlled thermal-habitat differentiation buffers alpine plant diversity against climate warming. Journal of Biogeography 38: 406416.
  • Scherrer, D., Schmid, S. & Körner, C. 2010. Elevational species shifts in a warmer climate are overestimated when based on weather station data. International Journal of Biometeorology 55: 645654.
  • Schmidt, F. & Persson, A. 2003. Comparison of DEM Data Capture and Topographic Wetness Indices. Precision Agriculture 4: 179192.
  • Schöb, C., Kammer, P.M., Choler, P. & Veit, H. 2008. Small-scale plant species distribution in snowbeds and its sensitivity to climate change. Plant Ecology 200: 91104.
  • Sebastiá, M. 2004. Role of topography and soils in grassland structuring at the landscape and community scales. Basic and Applied Ecology 5: 331346.
  • Spasojevic, M.J. & Suding, K.N. 2012. Inferring community assembly mechanisms from functional diversity patterns: the importance of multiple assembly processes. Journal of Ecology 100: 652661.
  • Swenson, N.G. 2011. The role of evolutionary processes in producing biodiversity patterns, and the interrelationships between taxonomic, functional and phylogenetic biodiversity. American Journal of Botany 98: 472480.
  • Taberlet, P., Fumagalli, L., Wust-Saucy, A.-G. & Cosson, J.-F. 1998. Comparative phylogeography and postglacial colonization routes in Europe. Molecular Ecology 7: 453464.
  • Tutin, T., Burges, N., Chater, A., Edmonson, J., Heywood, V., Moore, D., Valentine, D., Walters, S. & Webb, D. 1964. Flora Europaea I-V. Cambridge University Press, Cambridge, UK.
  • Vellend, M. 2010. Conceptual synthesis in community ecology. The Quarterly Review of Biology 85: 183206.
  • Virtanen, R., Dirnböck, T., Dullinger, S., Pauli, H., Staudinger, M. & Grabherr, G. 2002. Multi-scale patterns in plant species richness of European high mountain vegetation. In: Körner, C. & Spehn, E.M. (eds.) Mountain Biodiversity – A Global Assessment, pp. 91111. Parthenon Publishing, London, UK.
  • Vittoz, P. & Guisan, A. 2007. How reliable is the monitoring of permanent vegetation plots? A test with multiple observers Journal of Vegetation Science 18: 413422.
  • Vittoz, P., Randin, C.F., Dutoit, A., Bonnet, F. & Hegg, O. 2009. Low impact of climate change on subalpine grasslands in the Swiss Northern Alps. Global Change Biology 15: 209220.
  • Vonlanthen, C.M., Kammer, P.M., Eugster, W., Bühler, A. & Veit, H. 2006. Alpine vascular plant species richness: the importance of daily maximum temperature and pH. Plant Ecology 184: 1325.
  • Walker, D., Molenaar, J. & Billings, W. 2001. Snow–Vegetation Interactions in Tundra Environments. In: Jones, H., Pomeroy, J., Walker, D. & Wharton, W. (eds.) Snow ecology: an Interdisciplinary Examination of Snow-Covered Ecosystems, pp. 264322. Cambridge University Press, Cambridge, UK.
  • Warton, D. & Hui, F. 2011. The arcsine is asinine: the analysis of proportions in ecology. Ecology 92: 310.
  • Weiher, E., Freund, D., Bunton, T., Stefanski, A., Lee, T. & Bentivenga, S. 2011. Advances, challenges and a developing synthesis of ecological community assembly theory. Philosophical Transactions of the Royal Society of London. Series B, Biological sciences 366: 24032413.
  • Zobel, M., Otto, R., Laanisto, L., Naranjo-Cigala, A., Pärtel, M. & Fernández-Palacios, J.M. 2011. The formation of species pools: historical habitat abundance affects current local diversity. Global Ecology and Biogeography 20: 251259.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Author contributions
  10. References
  11. Supporting Information
FilenameFormatSizeDescription
jvs12060-sup-0001-AppendixS1.docxWord document18KAppendix S1. Full list of species according to their biogeographic classification. “Alpine species” include arctic-alpine taxa mainly distributed in the European mountain system. “Mediterranean species” are distributed throughout the mountains of the Iberian Peninsula or the Mediterranean basin. “Endemic species” are restricted to the Pyreneo-Cantabrian mountains. “Other species” includes taxa not restricted to high-mountain habitats.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.