Dataset DOI: 10.5061/dryad.rjdfn2zs1

Description of the data and file structure

The data given here are the supplementary data for the manuscript Kinzner, S., Ramskogler, K., Castlunger, S., Hofmeister, F. & Tasser, E. (2026): Influence of geomorphic disturbance on phenotypic species plasticity and vegetation cover in high-elevated belts. Ecology and Evolution, 16 (3). https://doi.org/10.1002/ece3.73056, including the information on the weather stations used for the description of the study areas and the species used for the analysis. Furthermore, is the information on the Principal Component Analysis (PCA), the Generalised Additive Models (GAMs) and the mean values of the measured traits per species provided. The raw data were already stored in PANGEA (https://doi.org/10.1594/PANGAEA.991478; https://doi.org/10.1594/PANGAEA.988475).

Files and variables

File: Data_A.xlsx

Description: This file includes a table with the weather station used for describing the study area with elevation and date of access, where KA is the Kauner Valley, HO the Horlach Valley, and MA the Martell Valley.

Variables

• study area, station, elevation of station, date of access

File: Data_B.xlsx

Description: There are the species used for the analysis of the different functional plant traits as well as the plot pair in the three different study areas (KA = Kauner Valley, HO = Horlach Vallely, and MA = Martell Valley). The plot pair is two plots in the same area, at the same elevation, one disturbed and one undisturbed

Variables

• no. (= number), species, plot pair

File: Data_C.xlsx

Description: The table shows the standardised loadings with the proportion explained for the three rotated components (RC) and the loadings of each variable to the components. The Principal Component Analyis (PCA) was performed in RStudio (RStudio Team, 2023) utilising the package psych (Revelle, 2024). Before performing the PCA a Kaiser-Meyer-Olking-test was performed to evaluate the suitability of the variables for the PCA. Afterwards, an unrotated PCA was run using the number of variables used as number of components followed by a rotated PCA using the number of components from the unrotated PCA with a standardised loading larger than one (Kaiser’s criterion) and a varimax rotation.

The 10 environmental variables were reduced to three rotated components (RC1–RC3). The three components explained 81 % of the variance. RC1 accounted for 39 % of the variance, RC2 for 21 %, and RC3 for 21 %, respectively (Table S8a). RC1 included, among others, Landolt indicator value T (0.96), elevation (-0.87), and mean annual temperature (0.74; Table S8b). Therefore, RC1 summarised key elevation-related climate parameters, whereby increasing component values mean improved growth conditions. RC2 included primarily edaphic parameters such as nutrient availability (0.85), continentality (-0.76) and water availability in the soil (0.75; Table S8b). Increasing values of the components thus reflect improved edaphic conditions. RC3 was positive with pH (0.86) and dispersity (0.72) and negative with humus (-0.67; Table S8b). This component will be referred to as less acidic debris sites. The variables SPI, northness, eastness, slope and precipitation were separately used in the GAM as they did not load sufficiently in the components RC1 to RC3.

Table S8: a) The column "Measure" includes the names of the statistical measures Standardised loadings (= sum of squared loadings, variance explained by the component), the proportion variance (= fraction of total variance explained at this component), and the proportion explained (= total variance explained up to this component) for the three rotated components (RC) and b) the loadings of each variable to the components.

The abbreviations used here:

cwm = community weighted mean (calculated based on the cover of the single species and the cover sum of all species in a plot)

LIV = Landolt indicator value (Landolt et al., 2010)

Table S8 refers to the article: Kinzner, S., Ramskogler, K., Castlunger, S., Hofmeister, F. & Tasser, E. (2026): Influence of geomorphic disturbance on phenotypic species plasticity and vegetation cover in high-elevated belts. Ecology and Evolution, 16 (3). https://doi.org/10.1002/ece3.73056

Landolt, E., Bäumler, B., Ehrhardt, A., Hegg, O., Klötzli, F., Lämmer, W., Nobis, M., Rudmann-Maurer, K., Schweingruber, F., Theurillat, J.-P., Urmi, E., Vust, M., & Wohlgemuth, T. (2010). Flora indicativa: Ökologische Zeigerwerte und biologische Kennzeichen zur Flora der Schweiz und der Alpen (2., völlig neu bearbeitete und erweiterte Auflage der Ökologische Zeigerwerte zur Flora der Schweiz (1977)). Haupt. https://haupt.ch/flora-indicativa/2329783258074610

Revelle, W. (2024). psych: Procedures for Psychological, Psychometric, and Personality Reserach (Version 2.4.1.) [Computer software]. Northwestern University. https://CRAN.R-project.org/package=psych

RStudio Team. (2023). RStudio: Integrated Development Environment in R [Computer software]. RStudio, PBC.

Variables

• RC1 (rotated component 1), RC2 (rotated component 2), RC3 (rotated component 3), SS loadings (standardised loadings), proportion Var (proportion of variance explained), proportion explained and the loadings of the variables in the components 

File: Data_E.xlsx

Description: This table shows the means of the measured functional traits of the species grown on undisturbed (0) and disturbed (1) sites with giving the mean with the standard deviation, the number of measurements and the p-value of the Kruskal-Wallis-test. The different colours indicate the different lifeforms and the different intensities of green the different significant levels. Yellow indicates nearly significant measurements. In the second column (statistic) are given the number of measured traits (n), and the level of disturbance (0 = undisturbed, 1 = disturbed) and p for p-value for each species.

Significance levels: . = p < 0.10, * = p < 0.05, ** = p < 0.01, *** = p <0.001, n.s. = not significant

Variables

• species, number of traits measured (n), level of disturbance (0 = undisturbed, 1 = disturbed), mean and standard deviation of the plant traits (plant height [cm], leaf dry weight [mg], leaf area [mm²], and specific leaf area (SLA) [mm²mg^-1^]  as well as the significance level

File: Data_D.zip

Description: Data_D.docx shows the equations of the different models and the description for the model equations (where g()  is the link function and E the expectation, total cover and rel. cov. (relative cover) the transformed ratio of the cover values,  the overall intercept,   the normally distributed random intercept for species,   the intercept,  the smooth function, and GMD geomorphic disturbance. For the analyses of the transformed ratio of the cover values, the GAM model was fitted using a beta distribution and, for species richness, a Poisson distribution).

Furthermore, we show in the subfolder "model_diagnostics" the graphs with the model diagnostics for each model, with the name of the response variable in the name for the figure, including A the quantile plot, B residuals vs linear predictors, C distribution of the residuals, and D response vs linear values.

Additionally, the detailed results of the GAMs (in the subfolder model_results) analysing the effects of the environmental variables on total cover, species richness, the relative cover of the different strategy types, life forms, and temperature types are included as a table (Results_models.xlsx) . The SPI is the stream power index which was calculated following Florinsky ( 2017), GMD is geomorphic disturbance a factor with the two levels disturbed and stable, and RC are the rotate components. For the parametric coefficients are given the estimate (est.), standard error (SE), and the p-value and for the approximate (approx.) significance (sign.) of the smooth terms (s) the estimated degree of freedoms (edf), the Chi-square (Chi Sq), and the p-value. Furthermore, we show here the adjusted R-square (R.-sq. adj.) and the deviance explained (dev. expl %) for each model. In the column p-value the bold numbers indicate a nearly significant value or significant value (different levels). "-" indicates if appearing in the part with the parametric coefficients this variable was a smooth term or the other way round. Row 14 has only empty cells to better separate the rows for the parametric coefficients and the approximate significance of the smooth terms.

We also show in the subfolder "model_results" the figures for the significant results of the smoothed terms: Fig. S19: Significant results of the smooth terms of the generalised additive models;1A – total cover, 1B-D – species richness, 1 E and 2A/B – relative cover cryophilic species, 2C – relative cover thermophilic species, 2D – relative cover competitive species, 2E and 3A – relative cover ruderal species, 3B – relative cover stresstolerant species, 3C – relative cover indifferent species, 3D/E and 4A/B – relative cover bryophytes, 4C-E relative cover graminoids, 5A/B – relative cover herbs, 5C/D- relative cover dwarf shrubs, 5 E and 6A/B- relative cover trees. and Fig. S21: Significant results of the smooth terms of the generalised additive models with the traits in the rows (1 plant height, 2 leaf dry weight, 3 leaf area, and 4 SLA) and the response variables in the columns as well as in row 5 the significant difference in undisturbed (0) and disturbed (1) plots (GMD = geomorphic disturbance) for the leaf area and SLA. as well as the figure related to the significant results of the categorical variable geomorphic disturbance (GMD), Fig. S20: Significant results categorical variable GMD (geomorphic disturbance; 0 = undisturbed, 1 = disturbed) for A total cover, B relative cover of bryophytes, C relative cover of lichens, and D relative cover of trees. Furthermore, the figure related to the random effect for the single species is given, Fig. S22: Random effect estimates for the species for A plant height, B leaf dry weight, C leaf area, and D SLA. The numbers in the figure with the random effect for the single species correspond with the numbers given in table Data_B.

All the data (also Fig. S19 to Fig. S22) are related to Kinzner, S., Ramskogler, K., Castlunger, S., Hofmeister, F. & Tasser, E. (2026): Influence of geomorphic disturbance on phenotypic species plasticity and vegetation cover in high-elevated belts. Ecology and Evolution, 16 (3). https://doi.org/10.1002/ece3.73056.

Code/software

RStudio (RStudio Team 2023) was used for the analysis, the package psych for the PCA (Revelle, W. 2024) and the package mgcv (Woods, S. 2023).

Revelle, W. (2024). psych: Procedures for Psychological, Psychometric, and Personality Research (Version 2.4.1.) [Computer software]. Northwestern University. https://CRAN.R-project.org/package=psych

RStudio Team. (2023). RStudio: Integrated Development Environment in R [Computer software]. RStudio, PBC.

Wood, S. (2023). mgcv: Mixed GAM Computation Vehicle with Automatic Smoothness Estimation [R]. https://cran.r-project.org/package=mgcv

Access information

Other publicly accessible locations of the data:

• https://doi.org/10.1594/PANGAEA.991478, https://doi.org/10.1594/PANGAEA.988475

Data was derived from the following sources:

• N/A