Influence of ungulates on the vegetation composition and diversity of mixed deciduous and coniferous mountain forest in Austria

  • Miriam Meier
  • Dieter Stöhr
  • Janette Walde
  • Erich Tasser
Original Article


The often highly elevated stocks of ungulates (red and roe deer and chamois) in the Alps shape the composition of the woody vegetation. The aim of this study was to investigate the influence of ungulates on the mixed deciduous and coniferous mountain forest in the district of Reutte, which boasts the highest density of ungulates in Tyrol (Austria), with a special focus on the effect of browsing by ungulates on plant diversity of the herb layer, different shrub layers. and the tree layer. Our results showed that within the fenced ungulate exclosures, (1) the composition of trees shifted towards fir (Abies alba) and various deciduous trees, whereas outside the fences, spruce became the dominant species; (2) the cover of dwarf shrubs and upper and lower shrub layers (1.3–5.0 and 0.5–1.3 m, respectively) increased significantly; (3) the cover of grasses decreased significantly and (4) the diversity decreased as an increase in the diversity of the tree and shrub layer was overcompensated by a significant decrease in the diversity of the undergrowth vegetation. Browsing by ungulates benefited grass species in the understory and altered the relative abundance of tree species in the lower layer which could, over time, result in compositional shifts in the canopy.


Game browsing Fenced ungulate exclosures Mixed mountain forest Plant diversity Understory vegetation Vegetation composition 


In most regions of the world, herbivory by ungulates is a natural process in the development of forests because for thousands of years, ungulates and forest species have co-evolved and adapted to one another as components of shared ecosystems (Reimoser and Reimoser 1998; Côté et al. 2004; Rooney et al. 2004; Kisanuki et al. 2012). In other parts of the world, however, mankind has introduced ungulate herbivores in the past centuries and therefore strongly changed natural processes and the community structure by causing a decrease of palatable plant species (Coomes et al. 2003; Martin et al. 2010; Mason et al. 2010; Peltzer et al. 2014). Even in the European Alps, the understory and young trees are an essential part of the diets of roe deer (Capreolus capreolus), red deer (Cervus elaphus) and chamois (Rupicapra rupicapra) (Reimoser and Reimoser 1998; Côté et al. 2004).

Nowadays, ungulate species frequently become overabundant (Office of the Tyrolean Government 2012; Côté et al. 2004; Apollonio et al. 2010). In addition to the widespread extinction of natural predators, high densities especially in the European Alps are a consequence of feeding during winter and providing intensive care (e.g. combating diseases or installation of game fences) which also prevent natural mechanisms of regulation (Fuller and Gill 2001; Apollonio et al. 2010). The physiology of ungulates is adapted to build fat reserves during the vegetation period to compensate for food shortages during the winter. With the additional winter feeding, the nutritional status of reproductive females increases which shortens the time up to sexual fertility and is likely one of the most significant reasons for the growth in the population density of deer (Rodriguez-Hidalgo et al. 2010; Milner et al. 2014). Mild winters in the last decades had similar effects: Due to better food accessibility, the deer body condition and winter survival increased which favoured population growth (Côté et al. 2004). However, other human activities, such as the fragmentation of forests and extensive recreational usage, constrict the habitat and disturb the animals’ browsing behaviour, increasing the pressure on biotopes (Reimoser and Reimoser 1998; Office of the Tyrolean Government 2012).

The effects of browsing by wild ungulates on the composition of tree species have been frequently investigated, with the conclusion that intensive, selective browsing can lead to a progressive loss of palatable woody species in forest stands (Côté et al. 2004; Martin et al. 2010; Perea et al. 2014). Fir (Abies alba) and deciduous trees, such as maple (Acer pseudoplatanus), beech (Fagus sylvatica) or mountain-ash (Sorbus aucuparia), are preferentially browsed and thereby spruce (Picea abies) becomes dominant (Office of the Tyrolean Government 2012; Katona et al. 2013; Dávalos et al. 2014; Holtmeier 2015). Particularly in mountainous regions, an adequate mixture of tree species and a multi-layered stand is required for the effective protection of the forest (Office of the Tyrolean Government 2012; Dorren et al. 2004). According to the so-called insurance hypothesis (Naeem 1998), resistance and resilience to disturbances also increase in mixed forests. This suggests that high diversity leads to more stable ecosystem functions under varying environmental conditions (i.e. land-use or climate change (Bengtsson et al. 2000; Katona et al. 2013).

The effects of ungulates on the composition of tree communities have been well analysed and also the effects of ungulates on the shrub layers and understory vegetation are basically known. These studies indicate that grasses often gain an advantage (Kirby 2001; Horsley et al. 2003; Rooney and Waller 2003; Côté et al. 2004; Collard et al. 2010). Furthermore, according to Baines et al. (1994) and Côté et al. (2004), browsing reduces the growth and height of the dwarf shrub species (e.g., Vaccinium myrtillus), and Martin et al. (2010) found that heavy ungulate browsing leads to a reduction in the shrub layer of forests to the benefit of understory species (Watkinson et al. 2001; Perea et al. 2014). However, regarding the effects of browsing on the diversity of plants, the results are not clear, and based on the literature, either a decrease or an increase in the diversity of species can occur under the influence of browsing by ungulates (Horsley et al. 2003; Côté et al. 2004; Hegland et al. 2013). According to Fuller and Gill (2001) and Côté et al. (2004), intensive grazing and browsing by deer reduce species richness. However, Holtmeier (2015) claims that increased browsing by red deer changes the vegetation structure and leads to an increase in the number of species. Some species, e.g. grasses resistant to browsing, plants with rosettes close to the ground and species avoided by deer can colonize subalpine meadows only because of browsing by red deer (Schütz et al. 2000). Thus, the effects of browsing on plant diversity are ambiguous. In line with the ‘intermediate disturbance hypothesis’ (Connell 1978; Hegland et al. 2013), we assume that the diversity in the underground vegetation increases due to ungulate browsing. Many species can particularly be found in a biocoenosis if the frequency or intensity of disturbances is neither too high nor too low.

In this study, the differences in the vegetation composition in forest areas with and without the influence of browsing ungulates are investigated, with particular focus on the response of understory vegetation to the exclusion of ungulates. We focus on forests at the montane vegetation belt (<1500 m a.s.l.) which do not grow at any climatically extreme or infertile sites, and therewith not on forests at the timberline (~2200 m a.s.l.). Specifically, we hypothesise that the exclusion of ungulates has the following consequences: (1) a shift towards deciduous trees, (2) an increase in cover of the shrub layer, (3) an increase in ground cover with dwarf shrubs, (4) a shift to higher cover of herbaceous plants compared with grasses and (5) a decrease in diversity.


Study area

The Tyrolean district of Reutte (Tyrol, Austria; 47° 17′ to 47° 32′ N, 10° 29′ to 10° 56′ E) was selected as the study site because it has the highest density of ungulates (Fig. 1), the highest hunting activity and the highest browsing damages caused by ungulates in Tyrol (Office of the Tyrolean Government 2012). The ranking is based on studies regarding: (a) forest rejuvenation, (b) bark stripping, (c) damages caused by browsing, (d) stocks of wildlife and (e) food requirements of wildlife. In 2013, nine wild hoofed animals per square kilometre were shot in this district, more than twice the Tyrolean average of 3.9 animals per square kilometre (Statistik Austria 2013). According to estimations by the Office of the Tyrolean Government (2012), 46 red deers (mean in Tyrol = 23), 42 roe deers (mean in Tyrol = 20) and 3.6 chamois (mean in Tyrol = 3.7) live on 1 km2 winter habitat. Wild boars, however, do not inhabit in the region permanently. In summary, the game density is far higher than in many other temperate zones (Côté et al. 2004).
Fig. 1

The study area of Tyrolean district of Reutte (Austria) and the geographic distribution of the 40 fenced ungulate exclosures

The district of Reutte is situated in the transition zone between a moderately warm, constantly wet climate with warm summers and a wet, snowy climate with cool summers (Kottek et al. 2006). Annual average temperatures vary between −4.8 and 6.7 °C, and annual average precipitation varies between 1362 and 2003 mm. Most of the forests in the district consist of montane northern alpine spruce-fir-beech forest (Aposerido – Fagetum Oberd. ex Oberd. et al. 1967). In addition to the dominant spruce, beech, fir and maple also occur in the tree layer. Frequent species of the herb layer are Polygonatum verticilliatum, Adenostyles glabra, Mercurialis perennis and Oxalis acetosella (Mucina et al. 1993). The last three species are particularly abundant.

Experimental design

As the focus of our study was on the montane vegetation belt (see Fig. 2), the study sites were situated between 850 m and 1470 m a.s.l. We selected 40 fenced ungulate exclosures from a total of 174 fenced exclosures in the district of Reutte. Over the past 40 years, these 174 fenced exclosures were installed by local foresters across the entire forest area. The selection criteria of the foresters for the enclosures were (a) location in the montane vegetation belt, (b) coverage of the entire forest area in the district, (c) no wild feeding in the surrounding area and (d) no tourist disturbances nearby (e.g., hiking trails, ski resorts). Additionally, we selected our 40 fenced ungulate exclosures according to the following criteria (a) the age of the exclosures (>10 years), (b) the size of the exclosures (>30 m2), (c) the altitude (evenly distributed throughout the whole montane altitudinal range), (d) the slope angle (evenly distributed throughout the whole slope angle range), (e) no infertile sites and (f) the location in the forest area (preferably in the centre of the forest area, only one site is located close to the forest edge). On average, the fences were in place since 21.6 years (min = 10 years; max = 34 years) and the median size of the fenced area was 140 m2 (min = 30 m2; max = 20,100 m2). From June to August 2012, the plant composition was determined within and outside of the fenced ungulate exclosures. The composition of the tree species and the structure of the mature stand were evaluated according to Hotter et al. (2011) on a 5000 m2 plot (50 × 100 m) per site, with the fenced area in the centre. For each plot, the topographic characteristics (i.e. altitude, slope angle, slope inclination, annual sum of sunshine hours); the vertical structure; the composition and relative contributions of species to the mixture of trees were assessed. The vertical structure depicts the number of vertical layers of tree life forms present in the mature stand. There are five vertical structure classes: (1) single story, (2) two-story, (3) more-story, (4) continuous vertical structure and (5) none, which indicates that not enough trees are present to assess the structure. Furthermore, the total canopy cover was determined in order to describe the proportion of the forest floor covered by the vertical projection of the tree crowns. We used six classes: (1) single trees, (2) scattered (crown cover <30%), (3) very sparse (cover = 30–50%), (4) sparse (cover = 50–70%), (5) dense (cover = 70–90%) and (6) very dense (cover = 90–110%). For the single tree species, the relative crown cover (%) determining the species dominance within a canopy was classified on the basis of four classes (dominant = cover >50%; subdominant = cover 25–50%; blended = 5–25%; scattered = <5%).
Fig. 2

Study design and comparison of important topographic, climatic and soil characteristics partly resulting from indicator values after Ellenberg (E) of paired ungulate exclosure and control plots. \( \overset{-}{x} \) = mean; s.e. = standard error of the mean

Within the fenced area, the plot for the detailed diversity investigations with a size of 5 × 5 m was located on the lower half of the hillside, with the lowest border of the plot situated 1 m from the game fence (Fig. 2) to evaluate vegetation out of the reach of ungulates. The reference plot outside the fenced ungulate exclosure was located 5 m lower on the slope; this design was used to exclude areas with more intense browsing caused by an accumulation of ungulates close to the fence. The experimental design with paired ungulate exclosure and control plots was similar to Reimoser and Reimoser (1998)), Mason et al. (2010) or Peltzer et al. (2014).

For the topographical characterization of plots, the coordinates (UTM–grid, map date VGS 84); altitude; exposure; slope inclination and the annual sum of sunshine hours were recorded. The annual sum of sunshine hours was estimated at 0.8 m above the ground with a horizontoscope by means of hemispherical drawings developed by Tonne (1954). As shown in Fig. 2, no big differences were observed between inside and outside the fences for the most important topographic factors with the obvious exception of the altitude because the unfenced areas were situated on average 2 m downwards.

Within the plots, phytosociological surveys according to Wilmanns (1998; modification of the Braun-Blanquet method) were conducted for the different layers: (1) tree layer (>5 m), (2) upper shrub layer (1.3–5 m), (3) lower shrub layer (0.5–1 m), (4) herb layer (up to 0.5 m) and (5) moss layer. Herein, the total cover and the cover of all vascular plant species were determined. Furthermore, within the herb layer, the total cover by mosses, trees and shrub seedlings, dwarf shrubs, grasses, ferns and herbs were estimated. The herbivory by ungulates were assessed by surveying the browsing damages of all individuals of the occurring tree species within the plots. The browsing classes were classified according to Reimoser and Reimoser (1998): (0) no damages, (1) low browsing: no damage on the leading shoot and less than 33% of the side shoots are browsed, (2) middle browsing: leading shoot is browsed once and between 33 and 66% of the side shoots are browsed and (3) strong browsing: leading shoot is browsed several times and more than 66% of the side shoots are browsed. Furthermore, grazing intensity was divided according a simplified classification: (1) no or only few droppings present, (2) many faecal piles and few trampling damages noticeable and (3) many faecal piles and many trampling damages as well as some resting places visible.

To control for potential site-specific differences inside and outside the fenced ungulate exclosures, we calculated the indicator values of Ellenberg (Ellenberg et al. 1992; Schaffers and Sýkora 2000; Warmelink et al. 2005) for moisture (F), nitrogen or productivity (N), soil reaction (R), light (L) and temperature (T).

Data analyses

Phytosociological data was entered in the phytosociological database program TURBOVEG for Windows Hennekens and Schaminee 2001). The nomenclature of vascular plants was according to the reference list German SL (Jansen and Dengler 2008). The phytosociological classifications of the forest communities were determined according to Mucina et al. (1993) and Fischer et al. (2008).

To assess the plant diversity inside and outside the fences, the Shannon–Wiener index was calculated. The value of this index increases with an increasing number of species and with an increasingly uniform distribution of relative numbers of individuals of each species (Smith and Smith 2015). To determine whether the value of the Shannon–Wiener index was caused by a high number of species or an equal distribution of a low number of species, the evenness index (E), which measures the uniformity of distribution of the individuals within the species, was also calculated (Mulder et al. 2004; Smith and Smith 2015).

Generalized linear models (GLMs; McCullagh and Nelder 1989) were used to fit the relationships between the single plant cover and diversity variables (as dependent variables) and the topographic characteristics, the forest vertical structure and composition properties and the intensity of browsing and grazing at the study areas (all independent variables):
$$ E\left({y}_i\right)={\mu}_i\ \mathrm{and}\ g\left({\mu}_i\right)={x}_i^T\beta $$

where yi is the observed value of the dependent variable of observation unit i (i = 1, ... N) following an exponential family distribution with the expectation value μi; g(μi)is the appropriate link function; xi is the , vector of independent variables, and β is the vector of regression coefficients which was estimated via maximum likelihood methods.

Our sample (sample siz e N) covers all 80 study plots. As independence of the plots could not be assumed due to the study design the variance-covariance matrix was corrected by computing cluster-robust standard errors. These cluster-robust standard errors were proposed by White (1984) for ordinary least squares (OLS) and by Liang and Zeger (1986) for linear and nonlinear models. The variable indicating each pair of control and exclosure was employed as cluster ID variable.

The measurement level of the dependent variables indicated appropriate link functions. Using information criteria and residual diagnostics, the most appropriate model was chosen and the corresponding settings are given in Appendix S1, 2 and 3.

With the GLMs, we aimed to explore the main effects of ungulate herbivory and all other independent variables on plant cover and diversity (e.g. Mason et al. 2010; Peltzer et al. 2014). As model quality criterion, we provide the squared correlation coefficient between the observed values of the dependent variable and its estimates for each model, i.e. cor2(y, \( \widehat{y} \) ), which is the coefficient of determination for an OLS model and therefore denoted as pseudo R2.

The vector of independent variables had to be adjusted with regard to the dependent variable. For modelling ‘tree layer’, the independent variables ‘horizon heightening’, ‘light value (Ellenberg)’, and ‘tree cover’ were not employed in the model as (a) the variable horizon heightening was measured at a height of approx. 150 cm and was therefore not relevant for the tree layer, (b) the light value (Ellenberg) was derived from the species composition of the herb layer and therefore was not relevant for the corresponding layer and (c) tree cover correlated perfectly with the dependent. For the same reasoning, the independent variable light value (Ellenberg) was also excluded from the models for upper and lower shrub layer. This circumstance is denoted in the Tables in the Appendix. Additionally, due to missings, ‘tree crown: sparse’ was dropped for Hs and E for the lower shrub layer (indicated in the Tables in the Appendix appropriately). All analyses were performed in Stata Statistical Software: Release 13, and R (R Core Team (2013).


Plant cover and diversity inside and outside the fences

Plant cover for the upper and lower shrub layers and average dwarf shrubs cover were higher in the fenced ungulate exclosures than in the unfenced areas (Table 1). On the other hand, the cover of grasses increased under the influence of ungulates. For the fenced and unfenced areas, no differences (p > 0.05, cf. Appendix 1) were observed for the overall cover of the tree, herb and moss layers or for the cover of the trees and shrubs, ferns and herbs within the herb layer.
Table 1

Comparison of the mean cover for different vegetation layers, different life forms in the herb layer and the Shannon–Wiener Index (Hs) and Evenness (E) for different vegetation layers inside and outside the fenced ungulate exclosures in the district of Reutte (Tyrol, Austria) from June to August 2012. \( \overline{x} \) denotes the mean value and s.e. the standard error of the mean

Total cover by layer



(\( \overline{x} \) ± s.e.)


(\( \overline{x} \) ± s.e.)

Tree layer (%)


40.0 ± 5.6

39.0 ± 5.2

Upper shrub layer (%)


14.5 ± 3.4

41.5 ± 4.56

Lower shrub layer (%)


19.8 ± 3.2

29.3 ± 3.5

Herb layer (%)


81.6 ± 2.8

73.9 ± 4.3

Moss layer (%)


27.1 ± 3.9

27.4 ± 4.5

Total cover of different life forms in the herb layer

Trees and shrubs (%)


14.5 ± 1.5

17.3 ± 2.3

Dwarf shrubs (%)


11.0 ± 2.4

20.5 ± 3.3

Grasses (%)


40.9 ± 3.0

29.4 ± 3.7

Herbs (%)


29.2 ± 2.4

29.1 ± 3.3

Ferns (%)


4.5 ± 1.1

3.64 ± 0.9

Tree seedlings (n m−2)


6.33 ± 3.7

8.15 ± 5.0

Shannon–Wiener index (Hs)



1.07 ± 0.03

1.01 ± 0.03

Tree layer


0.05 ± 0.02

0.12 ± 0.04

Upper shrub layer


0.13 ± 0.04

0.34 ± 0.05

Lower shrub layer


0.33 ± 0.04

0.45 ± 0.04

Herb layer


1.08 ± 0.03

0.98 ± 0.04

Evenness (E)



0.74 ± 0.02

0.74 ± 0.01

Tree layer


0.15 ± 0.01

0.35 ± 0.01

Upper shrub layer


0.32 ± 0.10

0.69 ± 0.08

Lower shrub layer


0.62 ± 0.07

0.75 ± 0.04

Herb layer


0.75 ± 0.02

0.74 ± 0.02

These changes in cover were the result of shifts in the cover of single species (see Appendix 4). Some tree species such as A. pseudoplatanus, F. sylvatica and S. aucuparia benefited from ungulate exclusion, as well as the dwarf shrub species Erica carnea and V. myrtillus. Thereby, the decrease in tree species cover was strongly influenced by ungulate herbivory (Fig. 3). Particularly affected were young trees in the herb and the lower shrub layer. Unless they outgrew of the immediate browsing zone, browsing damages clearly increased outside the exclosures. Some tree species such as fir and Scots pine (Pinus sylvestris) did not or only sporadically survive. These species were more or less absent in the upper shrub layer. On the other hand, many grass species such as Calamagrostis varia, Carex alba, Carex sempervirens, Sesleria albicans, Melica nutans and Deschampsia cespitosa profit from ungulate herbivory. Among the beneficiaries were also some herbs like A. glabra, Melampyrum sylvaticum and Petasites albus.
Fig. 3

Browsing damage by ungulate herbivory for various tree species and in different layers outside the fences at our study plots (n = 40). The discrepancy to 100% corresponds to the fraction of the plot where the corresponding species could not be detected

The overall species richness increased in the areas influenced by ungulates (Fig. 4). However, different developments occurred within the individual layers. In the tree layer, no differences in species richness were observed. On the other hand, in both shrub layers, diversity increased with ungulate exclusion. The opposite trend was observed in the herbaceous layer: the species richness increased by ungulate browsing.
Fig. 4

Comparison of the mean number of species (± s.e.) per vegetation layer inside and outside the fenced ungulate exclosures in the district of Reutte (Tyrol, Austria). The direction of the arrows indicates an increase in species richness; horizontal arrows indicate no differences and their statistical significance are shown in Appendix 2

The Shannon diversity index (Hs) also indicated a difference in diversity varying between areas with and without the influence of ungulates (Table 1). Within the tree layer and in the upper and lower shrub layers, diversity was reduced whereas diversity increased within the herb layer. The evenness index showed that the cover of particular single species increased from the lower to the upper shrub layer, whereas the proportions of species cover within the herb layer were evenly distributed inside and outside the fences (Table 1).

Effects of site variables ungulate browsing

Importance of site variables and browsing were determined using GLMs for total plant cover by layer, total cover of different life forms in the herb layer, as well as for tree seedling density (Appendix 1). Similar analyses were conducted for species number (Appendix 2) and plant diversity in the different horizontal layers (Appendix 3). The results of the models for plant cover changes and species richness regarding all forest layers confirm as commonly statistically significant predictors soil characteristics (moisture value, reaction value, nitrogen/productivity value); the light condition (light value); and the influence of ungulates. Most frequently (65.4% of all models), the browsing and grazing intensity by ungulates or the age of ungulate exclosure were identified as important predictors. In the following, we focus mainly on differences resulting from the influence of ungulates. The total tree cover was not statistically affected by any of the employed variables (cf. Appendix 1), but the cover of the upper shrub layer was significantly reduced by the browsing. Furthermore, both the upper and lower shrub layers were positively influenced by higher soil fertility. On the contrary, the herb cover was controlled by the light conditions in the underground of the trees. If more light was present, the herb cover increased significantly. However, the wild influence reduced the cover slightly, but significantly, whereby the dwarf shrub cover significantly decreased and the grass cover, however, significantly increased. In addition, dwarf shrubs were increasingly growing at more humid and acidic soils, while the grass cover was higher at alkaline sites. The tree seedlings density was positively influenced by the low light availability under dense tree canopies, but significantly negatively by intensive grazing. Even more considerable was that grazing and browsing affected biodiversity in the individual layers (Appendix 2 and 3). Basically, biodiversity (species richness, Hs, E) decreased significantly due to the influence of the game in the upper and lower shrub layer, whereas it increased in the herb layer. The impact of topographic, climatic, and soil variables were much less important.


In the light of the selection criteria of the study plots (described in ‘Experimental design’) the generalisability of our results is limited to forest areas in the montane belt and may not be valid for infertile sites and higher or lower altitudes. For areas with wild feeding nearby and tourist disturbances different findings may also be found.

Influence of ungulates on tree composition

Our first hypothesis is confirmed as the abundance and dominance shifted to fir and various deciduous tree species within the exclosures where tree establishment and growth are not affected by ungulates. Besides the species preferred by ungulates such as maple and rowan, the cover of beech increases in particular within the upper shrub layer. Beech is not the preferred food of ungulates but is more palatable than spruce and Scots pine (Dávalos et al. 2014; Holtmeier 2015). Therefore, the dominance of spruce increases in areas browsed by ungulates. Although the habitat requirements for spruce are broad and the species reaches dominance at high mountain sites (Otto 1994), spruce is suppressed by deciduous tree species when ungulates are excluded in montane areas.

Influence of game on vertical layers

Secondly, we hypothesise that the cover of the shrub layer increases with the exclusion of ungulates which is confirmed by the analysis of the total cover of the individual vertical layers. Particularly in the upper shrub layer (1.3–5 m), the uninhibited growth of species typically preferred by ungulates, specifically the early growth stage-fast growing species, led to an increase in the cover of fir. In the areas affected by ungulates, fir is absent from the higher vegetation layers (lower shrub, higher shrub, and tree layers), and the growth of maple and rowan is severely limited. The disappearance of fir may be caused by the lower regeneration capacity following browsing damage of fir compared to maple (Vilhar et al. 2015). According to Baumann et al. (2010), the trees remain small because of repeated herbivory by ungulates, which also increases the availability of these trees to ungulates for a longer period. In our study, only spruce and beech are damaged by ungulates within the upper shrub layer. Severe damage to evergreen spruce within the upper vegetation layer is associated with the seasonal availability of food. During the deep and long-lasting snow cover in winter, it is difficult for ungulates to reach small trees. However, tall individuals are above the snow cover and are therefore exposed to longer and more intensive pressure by ungulate herbivory (Heinze et al. 2011). The number of seedlings is lower outside than within the fenced ungulate exclosures (see Table 1). In our analyses (Appendix 1), this can be statistically linked to intensive ungulate browsing. Tree seedlings are an attractive and substantial food for ungulates (Gill and Beardall 2001; Reimoser 2003). According to our results, the number of tree seedlings is generally higher under more acid soil conditions in dense and darker forests. This is well understood, as all main tree species (spruce, fir, beech) germinate preferably in the shade or intermediate shade (Stancioiu and O’Hara 2006).

Influence of ungulates on understory vegetation

Dwarf shrubs

Hypothesis 3 is also confirmed. The overall cover of dwarf shrubs is significantly larger within the fenced ungulate exclosures, in particular, the cover of V. myrtillus. Dwarf shrubs are an important winter food resource (Hofmann et al. 2008; Perea et al. 2014) and are severely damaged when ungulates have little access to fodder, particularly in winter (Reimoser and Reimoser 1998; Suter et al. 2004; Baumann et al. 2010; Heinze et al. 2011). In addition to soft food such as grasses, herbs and lichens, hoofed animals require a high proportion of freshly lignified biomass (hard food), which provides the moisture necessary to maintain the activity of food processing bacteria in the rumen (Hofmann et al. 2008). Because of the morphology of their digestive tract, roe deer in particular concentrate on selecting food rich in easily digestible cellular contents, such as buds, needles, leaves or herbs (Tixier et al. 1997). The buds of dwarf shrubs are 25 to 50 cm aboveground and are reached easily by ungulates in particular at the end of the winter season. The severe browsing of the buds damages rejuvenation of dwarf shrubs. The rejuvenation of V. myrtillus is affected at even moderate browsing intensity (Hegland et al. 2010). The significant decrease in the cover of dwarf shrubs outside the fenced ungulate exclosures may have negative effects on other components of the forest ecosystem. For example, V. myrtillus is important as food and refuge in the habitat of the capercaillie (Tetrao urogallus; Rolstad 1988; Storch 1993). Extensive cover of V. myrtillus may also have a beneficial effect on the germination of trees. For example, spruce seeds reach the ground despite the cover of dwarf shrubs which fosters germination. Thus, dwarf shrubs are more favourable for the germination of trees than dense grass vegetation (Bayer 2006). However, the significant increase in competing bilberry in areas without ungulates might reduce the availability of water for trees, which could impede the germination of beech seeds that prefer moist habitats (Otto 1994). Competing species might also have a negative effect on seedling establishment and growth as a consequence of the reduction of light availability as mentioned by Kisanuki et al. (2012) for dwarf bamboo (Sasa nipponica) in a subalpine forest in central Japan.

Grasses and herbs

Our fourth hypothesis proposes that herbs would become dominant over grasses with ungulate exclosure. This hypothesis has not been confirmed. The cover of herb layer within the fences is reduced due to the heavy shading caused by the tree and shrub layers. Generally, it correlates positively with the light value (see light value, Fig. 2, Appendix 1). Outside the fences, however, grasses are displacing the herbs. The grass cover in areas accessible to ungulates increased significantly, which is likely because woody species regenerated easily in the absence of browsing. According to Smith and Smith (2015), the productivity of many grass species increases with moderate browsing. Moreover, the meristems of grasses are located at the base of their leaves to avoid consumption by grazers and allow fast regeneration. In contrast, the sensitive meristems of many herbaceous plants do not tolerate heavy grazing (Côté et al. 2004; Smith and Smith 2015). Their dense root system also contributes to the expansion of grasses. The rooting systems of herbaceous and woody plants are typically loose, horizontal and deep and usually cannot compete with those of grasses (Tasser and Tappeiner 2005). Van Auken (2000) reported that in stable, dense grass canopies (e.g. nard grass communities) the germination of tree and shrub seeds is impeded by the high light and root competition. In this study the cover of two grass species, C. alba and S. albicans, is particularly high under the influence of ungulates. C. alba forms long underground stolons that reach deep (Polomski and Kuhn 1998) and allow vegetative propagation despite browsing by ungulates, leading to a growth advantage. In addition to a dense root system near the surface, S. albicans has vertical, fine-spun, bunched roots that reach down to the solid bedrock. With the advantage gained by the exploitation of water and nutrients from lower soil layers, the blue moor grass is particularly competitive on steep, dry slopes (Polomski and Kuhn 1998). Furthermore, the open areas in the forest created by extensive browsing on competing species also promote light demanding grass species, including blue moor grass. The increase in grass cover in areas accessible to ungulates in this study is consistent with several earlier investigations. According to Putman et al. (1989), Kirby (2001) and Côté et al. (2004), the decrease in tree and shrub cover under the influence of herbivores is associated with an increase of Poales. Mosandl (1991) obtained a similar result in an approximately 10-year-old deforested area, with 51.5% of the biomass contributed by grass species outside the ungulates fences compared with 19.3% inside.

In our study, P. albus and M. sylvaticum are dominant in the unfenced areas. M. sylvaticum contains the weakly poisonous aucubin (Munteanu et al. 2010), and P. albus contains the alkaloid senkirkin (Zhang et al. 2015); therefore, these species are avoided by hoofed animals. The cover of A. glabra also increased in unfenced areas. This species rarely occurs at lower than 20% relative solar radiation (light value 6). Thus, the ungulate exclusion areas provided too much shade.

Ferns and mosses

The cover of moss layer and fern are similar inside and outside the fenced ungulate exclosures, most likely because moss and fern species are not a significant part of the diet of ungulates (Hegland et al. 2013). Especially some fern species follow characteristic survival strategies. For example, Pteridium aquilinum, contains carcinogenic substances and mycotoxins to deter herbivory (Hirono and Grasso 1981).

Influence of ungulates on diversity

Diversity within the herb layer decreased inside the fenced ungulate exclosures and with the age of the exclosure (Appendix 2 and 3), which confirmed hypothesis 5. The level of ungulate browsing outside the fences is comparable to extensive grazing by livestock, which often leads to higher diversity than on intensively grazed or abandoned systems (Tasser and Tappeiner 2002). Extensive grazing is an intermediate type of disturbance to vegetation, and according to the intermediate disturbance hypothesis, many species can then be found in a biocoenosis if the frequency or intensity of disturbances is neither too high nor too low (Connell 1978; Côté et al. 2004; Rooney et al. 2004; Hegland et al. 2013). With very frequent and intensive disturbances, only a few pioneering species are expected to establish themselves, but with very rare disturbances, most biocoenosis will be dominated by strong and competitive species (Hegland et al. 2013), such as V. myrtillus and various deciduous tree species in this study. On the other hand, the diversity in the shrub and tree layer inside the fences increase as tree and shrub species typically preferred by ungulates are protected (see also Baumann et al. 2010; Heinze et al. 2011; Vilhar et al. 2015).


The negative influence of ungulates on the rejuvenation of tree species was confirmed for the district of Reutte. Outside the fenced ungulate exclosures, the abundance of spruce increased at the expense of other tree species. Although the development of mixed species forest is consistent with the location and is important for the protective function of forest, a mixed forest is only possible with a limited number of ungulates. The increase of spruce reduces the resilience of forests in response to climate change in the long-term because of the apparent sensitivity of spruce to drought. Moreover, the dominance of spruce for longer periods triggered by ungulate browsing might lead to mass outbreaks of pests, in particular, bark beetles, which will reduce the forest’s protective function. Additionally, the browsing of the understory vegetation benefits grass species, which delays the rejuvenation of trees. The cover of dwarf shrubs was also reduced. While negative effects of browsing on the woody species were demonstrated in this study, the positive influence of ungulates on the diversity of herbaceous plants in the herb layer was also shown.

Supplementary material

10344_2017_1087_MOESM1_ESM.docx (87 kb)
ESM 1(DOCX 87 kb)


  1. Apollonio M, Andersen R, Putman R (2010) European ungulates and their management in the twenty-first century. Cambridge University Press, CambridgeGoogle Scholar
  2. Baines D, Sage RB, Baines MM (1994) The implications of red deer grazing to ground vegetation and invertebrate community structure of Scottish native pinewoods. J of Appl Ecol 31:776–783. doi:10.2307/2404167 CrossRefGoogle Scholar
  3. Baumann M, Brang P, Burger T, Eyholzer R, Herzog S, Imesch N, Kupferschmid A, Rüegg D, Wehrli A (2010) Wald und Wild – Grundlagen für die Praxis. Wissenschaftliche und methodische Grundlagen zum integralen Management von Reh, Gämse und Rothirsch in ihrem Lebensraum. Bundesamt für Umwelt BAFU Bern. Umwelt- Wissen 1013:1–232Google Scholar
  4. Bayer R (2006) Root development, nutrition, mycorrhization and “positive microsites” of Norway spruce (Picea abies [L.] Karst.) regeneration on protective forest sites in the Bavarian Limestone Alps. PhD Thesis, School of Life Sciences Weihenstephan, TUMGoogle Scholar
  5. Bengtsson J, Nilsson SG, Franc A, Menozzi P (2000) Biodiversity, disturbances, ecosystem function and management of European forests. For Ecol Manag 132:39–50. doi:10.1016/S0378-1127(00)00378-9 CrossRefGoogle Scholar
  6. Collard A, Lapointe L, Ouellet JP, Crete M, Lussier A, Daigle C, Cote SD (2010) Slow responses of understory plants of maple-dominated forests to white-tailed deer experimental exclusion. For Ecol Manag 260:649–662. doi:10.1016/j.foreco.2010.05.021 CrossRefGoogle Scholar
  7. Connell JH (1978) Diversity in tropical rain forests and coral reefs. High diversity of trees and corals is maintained only in a nonequilibrium state. Sci. New Series 199:1302–1310Google Scholar
  8. Coomes DA, Allen RB, Forsyth DM, Lee WG (2003) Factors preventing the recovery of New Zealand forests following control of invasive deer. Cons Biol 17:450–459. doi:10.1046/j.1523-1739.2003.15099.x CrossRefGoogle Scholar
  9. Côté SD, Rooney TP, Tremblay J-P, Dussault C, Waller DM (2004) Ecological impacts of deer overabundance. Annu Rev Ecol Evol Syst 35:113–147. doi:10.1146/annurev.ecolsys.35.021103.105725 CrossRefGoogle Scholar
  10. Dávalos A, Nuzzo V, Blossey B, Wardle D (2014) Demographic responses of rare forest plants to multiple stressors: the role of deer, invasive species and nutrients. J Ecol 102(5):1222–1233Google Scholar
  11. Dorren LKA, Berger F, Imeson AC, Maier B, Rey F (2004) Integrity, stability and management of protection forests in the European Alps. For Ecol Manag 195:165–176. doi:10.1016/j.foreco.2004.02.057 CrossRefGoogle Scholar
  12. Ellenberg H, Weber HE, Düll R, Wirth V, Werner W, Paulißen D (1992) Indicator values of plants in Central Europe. Scripta Geobotanica 18. Erich Goltze Kg, GöttingenGoogle Scholar
  13. Fischer A, Oswald K, Adler W (2008) Exkursionsflora für Österreich, Liechtenstein und Südtirol. Biologiezentrum der oberösterreichischen LandesmuseenGoogle Scholar
  14. Fuller R, Gill R (2001) Ecological impacts of increasing numbers of deer in British wood-lands. Institute of Chartered Foresters Forestry 74:193–199. doi:10.1093/forestry/74.3.193 Google Scholar
  15. Gill RMA, Beardall V (2001) The impact of deer on woodlands: the effects of browsing and seed dispersal on vegetation structure and composition. Forestry 74:209–218. doi:10.1093/forestry/74.3.209 CrossRefGoogle Scholar
  16. Hegland SJ, Jongejans E, Rydgren K (2010) Investigating the interaction between ungulate grazing and resource effects on Vaccinium myrtillus populations with integral projection models. Oecologia 163:695–706. doi:10.1007/s00442-010-1616-2 CrossRefPubMedGoogle Scholar
  17. Hegland SJ, Lilleeng MS, Moe SR (2013) Old-growth forest floor richness increases with red deer herbivory intensity. For Ecol Manag 310:267–274. doi:10.1016/j.foreco.s2013.08.031 CrossRefGoogle Scholar
  18. Heinze E, Boch S, Fischer M, Hessenmöller D, Klenk B, Müller J, Prati D, Schulze ED, Seele C, Socher S, Halle S (2011) Habitat use of large ungulates in northeastern Germany in relation to forest management. For Ecol Manag 261:288–296. doi:10.1016/j.foreco.2010.10.022 CrossRefGoogle Scholar
  19. Hennekens SM, Schaminee JHJ (2001) Turboveg, a comprehensive database management system for vegetation data. J Veg Sci 12:589–591. doi:10.2307/3237010 CrossRefGoogle Scholar
  20. Hirono I, Grasso P (1981) Natural carcinogenic products of plant origin. CRC Cr Rev Toxicol 8:235–277. doi:10.3109/10408448109109659 CrossRefGoogle Scholar
  21. Hofmann G, Pommer U, Jenssen M (2008) Wildökologische Lebensraumbewertung für die Bewirtschaftung des wiederkäuenden Schalenwildes im nordostdeutschen Tiefland. Ministerium für Ländliche Entwicklung, Umwelt und Verbraucherschutz (MLUV) des Landes Brandenburg, EberswaldeGoogle Scholar
  22. Holtmeier F-K (2015) Animals’ influence on the landscape and ecological importance. Natives, newcomers, homecomers. Springer, New YorkGoogle Scholar
  23. Horsley SB, Stout SL, DeCalesta DS (2003) White-tailed deer impact on the vegetation dynamics of a northern hardwood forest. Ecol Appl 13:98–118. doi:10.1656/045.020.0301 CrossRefGoogle Scholar
  24. Hotter M, Klosterhuber R, Wallner M, Maynollo H, Simon A (2011) Standorterkundung Zwischenalpen. Office of the Tyrolean government. Abteilung Forst, InnsbruckGoogle Scholar
  25. Jansen F, Dengler J (2008) German SL- Eine universelle taxonomische Referenzliste für Vegetationsdatenbanken in Deutschland. Tuexenia 28:239–253Google Scholar
  26. Katona K, Kiss M, Bleier N, Székely J, Nyeste M, Kovács V, Terhes A, Fodor A, Olajos T, Rasztovits E, Szemethy L (2013) Ungulate browsing shapes climate change impacts on forest biodiversity in Hungary. Biodivers Conserv 22:1167–1180. doi:10.1007/s10531-013-0490-8 CrossRefGoogle Scholar
  27. Kirby KJ (2001) The impact of deer on the ground flora of British broadleaved woodland. Forestry 74(3):219–229. doi:10.1093/forestry/74.3.219 CrossRefGoogle Scholar
  28. Kisanuki H, Kudo T, Nakai A (2012) Removing aboveground vegetation facilitates survival but slows height growth of spruce saplings in a fenced, degraded sub-alpine forest in Central Japan. J For Res 17:110–115. doi:10.1007/s10310-011-0252-y CrossRefGoogle Scholar
  29. Kottek M, Grieser J, Beck C, Rudolf B, Rubel F (2006) World map of the Köppen-Geiger climate classification updated. Meteorol Z 15:259–263. doi:10.1127/0941-2948/2006/0130 CrossRefGoogle Scholar
  30. Liang K-Y, Zeger SL (1986) Longitudinal data analysis using generalized linear models. Biometrika 73:13–22CrossRefGoogle Scholar
  31. Martin J-L, Stockton SA, Allombert S, Gaston AJ (2010) Top-down and bottom-up consequences of unchecked ungulate browsing on plant and animal diversity in temperate forests: lessons from a deer introduction. Biol Invasions 12:353–371. doi:10.1007/s10530-009-9628-8 CrossRefGoogle Scholar
  32. Mason NWH, Peltzer DA, Richardson SJ, Bellingham PJ, Allen RB (2010) Stand development moderates effects of ungulate exclusion on foliar traits in the forests of New Zealand. J Ecol 98:1422–1433. doi:10.1111/j.1365-2745.2010.01714.x CrossRefGoogle Scholar
  33. McCullagh P, Nelder JA (1989) Generalized linear models, Second edn. Chapman and Hall, LondonCrossRefGoogle Scholar
  34. Milner JM, Van Beest FM, Schmidt KT, Brook RK, Storaas T (2014) To feed or not to feed? Evidence of the intended and unintended effects of feeding wild ungulates. J Wildl Manag 78:1322–1334. doi:10.1002/jwmg.798 CrossRefGoogle Scholar
  35. Mosandl R (1991) Die Steuerung von Waldökosystemen mit waldbaulichen Mitteln-dargestellt am Beispiel des Bergmischwaldes. Mitteilungen aus der Staatsforstverwaltung Bayerns 46:1–246Google Scholar
  36. Mucina L, Grabherr G, Wallnöfer S (1993) Die Pflanzengesellschaften Österreichs. Teil III. Wälder und Gebüsche, Gustav Fischer Verlag, JenaGoogle Scholar
  37. Mulder CPH, Bazeley-White E, Dimitrakopoulos PG, Hector A, Scherer-Lorenzen M, Schmid B (2004) Species evenness and productivity in experimental plant communities. Oikos 107:50–63. doi:10.1111/j.0030-1299.2004.13110.x CrossRefGoogle Scholar
  38. Munteanu M, Dehelean CA, Ionescu D, Andoni M, Butnariu M (2010) Investigation of the use of Melampyrum sp. extract samples to assess metals contamination. J Agroaliment Proc Technol 16:382–386. doi:10.1186/2193-1801-2-676 Google Scholar
  39. Naeem S (1998) Species redundancy and ecosystem reliability. Conserv Biol 12:39–45. doi:10.1111/j.1523-1739.1998.96379.x CrossRefGoogle Scholar
  40. Office of the Tyrolean Government (2012) Tiroler Waldbericht 2011. Gruppe Forst, InnsbruckGoogle Scholar
  41. Otto H (1994) Waldökologie. Ulmer, StuttgartGoogle Scholar
  42. Peltzer DA, Allen RB, Bellingham PJ, Richardson SJ, Wright EF, Knightbridge PI, Mason NWH (2014) Disentangling drivers of tree population size distributions. For Ecol Manag 331:165–179. doi:10.1016/j.foreco.2014.06.037 CrossRefGoogle Scholar
  43. Perea R, Girardello M, San Miguel A (2014) Big game or big loss? High deer densities are threatening woody plant diversity and vegetation dynamics. Biodivers Conserv 23:1303–1318. doi:10.1007/s10531-014-0666-x CrossRefGoogle Scholar
  44. Polomski J, Kuhn N (1998) Wurzelsysteme. Birmendorf, Eidgenössische Forschungsanstalt für Wald, Schnee und Landschaft. Haupt, BernGoogle Scholar
  45. Putman RJ, Edwards J, JEE M, How RC, Hill SD (1989) Vegetational and faunal change in an area of heavily grazed woodland following relief of grazing. Biol Conserv 47:13–32. doi:10.1016/0006-3207(89)90017-7 CrossRefGoogle Scholar
  46. R Core Team (2013) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna URL Google Scholar
  47. Reimoser F (2003) Steering the impacts of ungulates on temperate forests. J Nat Conserv 10:243–252. doi:10.1078/1617-1381-00024 CrossRefGoogle Scholar
  48. Reimoser F, Reimoser S (1998) Richtiges Erkennen von Wildschäden am Wald. Zentralstelle Österreichische Landesjagdverbände, WienGoogle Scholar
  49. Rodriguez-Hidalgo P, Gortazar C, Tortosa FS, Rodriguez-Vigal C, Fierro Y, Vicente J (2010) Effects of density, climate, and supplementary forage on body mass and pregnancy rates of female red deer in Spain. Oecologia 164:389–398. doi:10.1007/s00442-010-1663-8 CrossRefPubMedGoogle Scholar
  50. Rolstad J (1988) Autumn habitat of capercaillie in southeastern Norway. J Wildl Manag 52:747–753. doi:10.2307/3800941 CrossRefGoogle Scholar
  51. Rooney TP, Waller DM (2003) Direct and indirect effects of white-tailed deer in forest ecosystems. For Ecol Manag 181:165–176. doi:10.1016/S0378-1127(03)00130-0 CrossRefGoogle Scholar
  52. Rooney TP, Wiegmann SM, Rogers DA, Waller DM (2004) Biotic impoverishment and homogenization in unfragmented forest understory communities. Cons Biol 18:787–798. doi:10.1111/j.1523-1739.2004.00515.x CrossRefGoogle Scholar
  53. Schaffers AP, Sýkora KV (2000) Reliability of Ellenberg indicator values for moisture, nitrogen and soil reaction: a comparison with field measurements. J Veg Sci 11:225–244. doi:10.2307/3236802 CrossRefGoogle Scholar
  54. Schütz M, Wildi O, Achermann G, Krüsi BO, Nievergelt B (2000) Predicting the development of subalpine grassland in the Swiss National Park: how to build a succession model based on data from long-term permanent plots. Natl park-Forsch Schweiz 89:207–235Google Scholar
  55. Smith RL, Smith TM (2015) Elements of ecology, 9/E. Benjamin Cummings, Menlo Park, CAGoogle Scholar
  56. Stancioiu PT, O’Hara KL (2006) Regeneration growth in different light environments of mixed species, multiaged, mountainous forests of Romania. Eur J Forest Res 125:151–162. doi:10.1007/s10342-005-0069-3 CrossRefGoogle Scholar
  57. Storch I (1993) Habitat selection by capercaillie in summer and autumn: is bilberry important? Oecol 95:257–265. doi:10.1007/BF00323498 CrossRefGoogle Scholar
  58. Suter W, Suter U, Krüsi B, Schütz M (2004) Spatial variation in summer diet of red deer Cervus elaphus in the eastern Swiss Alps. Wildlife Biol 10:43–50Google Scholar
  59. Tasser E, Tappeiner U (2002) Impact of land use changes on mountain vegetation. Appl Veg Sci 5:173–184. doi:10.1111/j.1654-109X.2002.tb00547.x CrossRefGoogle Scholar
  60. Tasser E, Tappeiner U (2005) New model to predict rooting in diverse plant community compositions. Ecol Model 185:195–211. doi:10.1016/j.ecolmodel.2004.11.024 CrossRefGoogle Scholar
  61. Tixier H, Duncan P, Scehovic J, Yani A, Gleizes M, Lila M (1997) Food selection by European roe deer (Capreolus capreolus): effects of plant chemistry, and consequences for the nutritional value of their diets. J Zool 242:229–245. doi:10.1111/j.1469-7998.1997.tb05799.x CrossRefGoogle Scholar
  62. Tonne F (1954) Better building by insolation and daylight studies. Karl Hofmann, StuttgartGoogle Scholar
  63. Van Auken OW (2000) Shrub invasions of north American semiarid grasslands. Annu Rev Ecol Syst 31:197–215. doi:10.1146/annurev.ecolsys.31.1.197 CrossRefGoogle Scholar
  64. Vilhar U, Roženbergar D, Simončič P, Diaci J (2015) Variation in irradiance, soil features and regeneration patterns in experimental forest canopy gaps. Ann For Sci 72:253–266. doi:10.1007/s13595-014-0424-y CrossRefGoogle Scholar
  65. Warmelink GWW, Goedhart PW, van Dobben HF, Berendse F (2005) Plant species as predictors of soil pH: replacing expert judgement with measurements. J Veg Sci 16:461–470. doi:10.1111/j.1654-1103.2005.tb02386.x CrossRefGoogle Scholar
  66. Watkinson AR, Riding AE, Cowie NR (2001) A community and population perspective of the possible role of grazing in determining the ground flora of ancient woodlands. Forestry 74:231–239. doi:10.1093/forestry/74.3.231 CrossRefGoogle Scholar
  67. White H (1984) Asymptotic theory for econometricians. Academic Press, San DiegoGoogle Scholar
  68. Wilmanns O (1998) Ökologische Pflanzensoziologie: Eine Einführung in die Vegetation Mitteleuropas. Quelle & Meyer, HeidelbergGoogle Scholar
  69. Zhang ZZ, Zhi HJ, Qin XM, Li ZY (2015) Chemical comparison of different Farfarae Flos by NMR-based metabolomic approaches. Acta Pharm Sin 50:599–604Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2017

Authors and Affiliations

  • Miriam Meier
    • 1
  • Dieter Stöhr
    • 2
  • Janette Walde
    • 3
  • Erich Tasser
    • 4
  1. 1.Institute of EcologyUniversity of InnsbruckInnsbruckAustria
  2. 2.Tyrolean Forest ServiceProvince of TyrolInnsbruckAustria
  3. 3.Department of Statistics, Faculty of Economics and StatisticsUniversity of InnsbruckInnsbruckAustria
  4. 4.Institute for Alpine EnvironmentEuropean Academy Bozen/BolzanoBolzanoItaly

Personalised recommendations