Open access peer-reviewed chapter - ONLINE FIRST

Assessment of the Impact of Ungulate Browsing on Tree Regeneration

Written By

Andrea Doris Kupferschmid, Raphaël Greilsamer, Peter Brang and Harald Bugmann

Submitted: 18 September 2022 Reviewed: 19 October 2022 Published: 16 November 2022

DOI: 10.5772/intechopen.108667

Animal Nutrition - Annual Volume 2024 IntechOpen
Animal Nutrition - Annual Volume 2024 Authored by Manuel Gonzalez Ronquillo

From the Annual Volume

Animal Nutrition - Annual Volume 2024 [Working Title]

Dr. Manuel Gonzalez Ronquillo

Chapter metrics overview

148 Chapter Downloads

View Full Metrics

Abstract

Ungulates browse on plants and may have an influence on tree regeneration. Browsing percentage (the relative number of browsed terminal shoots) captures little of the effective browsing impact on tree regeneration, such as reduction in stem number or loss of tree species in the future stand. The main objective is to present the most important factors that can influence the impact of browsing and how these factors can be measured objectively. We conducted a literature review of influencing factors, tested these in different areas of the Swiss forest and performed studies to objectively survey them in inventories. Apart from browsing percentage, the following factors are important for estimating the long-term browsing impact: (i) the spatial distribution and density of tree saplings; (ii) the within-tree browsing intensity; (iii) the site-specific height growth of the tree saplings, and thus the time needed to grow out of the reach of browsers and the possible changes in growth rate ranking between the different selected species; (iv) the possible delay in the tree response after browsing; and (v) the tree mortality induced by browsing. The first four of these factors can be assessed easily and should thus be included in future inventories of browsing impact.

Keywords

  • growth rate
  • herbivory
  • tree regeneration
  • ungulate browsing

1. Introduction

Wild ungulates, such as roe deer (Capreolus capreolus L.), red deer (Cervus elaphus L.), and chamois (Rupicapra rupicapra L.), depend on plants as food. The leaves, shoots, and bark of tree saplings are part of their normal diet [1, 2]. The browsing of terminal shoots (also called leading shoots) can cause a loss of tree height and, depending on the situation, a loss of increment in the following years or even increased mortality [3]. Ungulates thereby selectively browse on a single tree species [4] or even certain individuals [5]. Not all tree species are able to respond quickly and effectively to browsing (see review by [6]). For example, silver fir (Abies alba) from high-elevation provenances showed a delayed response to shoot pruning (simulated browsing), whereas Norway spruce (Picea abies) did not [7]. Wild ungulates, therefore, influence the growth of tree saplings differently, depending on the tree species, which can alter their relative abundance and thus their establishment success [8]. Observations, experiments, and model simulations in many temperate and boreal forests have shown that selective browsing by wild ungulates can affect the development of a forest stand and lead to substantial changes in the composition and structure of plant communities [3]; for example, fir may decline [9]. According to the recent enforcement aid on biodiversity in Swiss forests, the degradation of ecosystems as a result of the loss in tree species diversity due to ungulate browsing is now one of the main challenges in mountain forests in Switzerland [10].

In large areas of Europe, the density of wild ungulates has increased drastically in the last decades [11]. With higher ungulate densities, the proportion of tree saplings that are browsed increases [9], as does the amount of damage [12], although not linearly [13]. In Switzerland, the browsing percentages for the most selected tree species (silver fir and oak species) have increased considerably in the recent years [14]. However, the long-term influence of browsing on forest development cannot be derived directly from browsing percentage. Here, we present research results suggesting why this is the case.

In this chapter, we first define browsing percentage and discuss the methods that are frequently used to measure it. We then describe the main factors that determine the influence of browsing on tree regeneration (Figure 1). We do this by explaining each factor in turn and suggesting objective measurement possibilities. The factor “tree response after browsing” has no separate section, as it is explained in the sections about within-tree browsing intensity and height loss due to browsing. We conclude our review by suggesting that information on these important factors is necessary to estimate the effects of browsing on tree regeneration and should therefore be considered in future surveys of browsing.

Figure 1.

The main factors that affect the influence of ungulate browsing on tree regeneration. Adapted from [15].

Advertisement

2. Frequency of browsing

The frequency of browsing in an area can be measured objectively and is easily reproducible when quantified as browsing intensity, a specific form of browsing percentage [16]. Browsing intensity is the number of saplings with a browsed terminal shoot in the year preceding the survey (i.e., on the last completed terminal shoot), expressed as a percentage of all observed saplings [16]. Typically, it is assessed for saplings of individual tree species between 10 and 130 cm height since this is the tree size that is browsed most by wild ungulates [17].

Browsing intensity can be determined using different survey methods, mostly with systematically established grids of sample plots. Such surveys are used to determine the presence or absence of browsing on the terminal shoot of tree saplings:

  • in fixed circular areas around a sample plot center (e.g., fourth Swiss National Forest Inventory NFI4 [18]),

  • in a circular sector defined according to the density of tree saplings (e.g., indicator areas in Switzerland [19] and Austrian wildlife impact monitoring areas [20]), or

  • for the k saplings nearest to the plot center, where k = 1 (e.g., Swiss NFI3 [21]), or better k > 1 (e.g., Bavarian browsing inventory [22] and e.g., [23, 24]; see k-tree sampling [25, 26]).

Statistically, all survey methods are “correct.” Because saplings are irregularly distributed, however, the different survey methods lead to different values for the frequency of browsing. This issue is particularly evident in the case of a clumped sampling distribution [26]. This means that the results of individual survey methods are not directly comparable.

In general, survey methods in which the survey is stopped after a certain number of saplings, regardless of the tree species, or in which only the nearest tree per height class is considered, irrespective of tree species (see also discussion in [27]), are rather inappropriate, as their results depend strongly on the exact survey point, the spatial distribution of the saplings, and the tree species composition. More robust and thus more reproducible results are provided by the complete survey of all saplings per sampling plot, the tree-species-specific termination of the survey after x saplings of a certain species (e.g., 20 per species according to [27]), or the assessment of the k nearest trees to the sample plot center, with k trees defined per species and height class (where k > 1).

Since wild ungulates move in space and selectively choose their food, and as favorable areas for tree regeneration are irregularly distributed, the measured frequency of browsing depends on the spatial distribution of the sample plots. The measured values differ depending on whether the plots are sampled in a systematic grid over the entire forest (i.e., in all stages of development and across all forest types) or only in open sites where tree regeneration is more likely [20]. The area (e.g., forest, regeneration area, and problem area) for which a statement on browsing frequency is to be made must therefore be deliberately chosen and communicated accordingly.

Advertisement

3. Within-tree browsing intensity

The wild ungulates in Central Europe are concentrate selectors (roe deer) or mixed browsers (red deer and chamois), and they select not only specific plant species [4, 28], but also certain individual saplings and even plant parts. Particularly with dense tree regeneration, these ungulates select only the (terminal) buds or uppermost parts of the terminal shoot of the most vigorous and dominant saplings [29, 30]. The within-tree browsing intensity (or “browsing strength”) indicates whether only the bud(s), large parts of last year’s terminal shoot, or even older terminal shoots have been browsed away (Figures 2 and 3, Table 1). For an individual tree sapling, the within-tree browsing intensity determines:

  • the immediate loss in height caused when parts of the terminal shoot are browsed away,

  • the subsequent loss of height increment,

  • the loss of reserve material necessary for new shoots, and

  • the number of meristems remaining on the terminal shoot, that is, the number of regularly formed and dormant buds out of which a new terminal shoot can grow.

Figure 2.

Light terminal shoot browsing on silver fir (Abies alba; a and b), rowan (Sorbus aucurparia; c), and sycamore maple (Acer pseudoplatanus; d). only the terminal buds were eaten off, whereas the main part of the last annual height increment—including the lateral buds on the terminal shoot—is still present. These saplings suffer almost no reduction in tree height and can usually form a new terminal shoot without a delay. Photo by A. Kupferschmid.

Figure 3.

Moderate up to heavy terminal shoot browsing on silver fir (Abies alba; a and b) and rowan (Sorbus aucuparia; c). Almost all of the last year’s terminal shoot has been eaten away and with it many meristems. Photo by A. Kupferschmid.

Categories of within-tree browsing intensityDefinitionExample
Not browsedTerminal shoot not browsed and not damagedNorway spruce in Figure 5 and current year for silver fir in Figures 5 and 6
Lightly browsedTerminal shoot lightly browsed, that is, terminal bud eaten off but most of the last height increment, including lateral buds, still presentFigure 2 and previous years for silver fir in Figures 5 and 6
Heavily browsedTerminal shoot moderately to heavily browsed or even older terminal shoots eaten awayFigures 3 and 7
No shootNo terminal shoot present because no new terminal shoot formed after browsing in the preceding year or after frost or insect damage in the preceding yearFigure 8
Other damageTerminal shoot or terminal bud damaged due to reasons other than browsingBreakage, drought, frost (autumn assessment), insect damage

Table 1.

Categories of within-tree browsing intensity, with short definitions and examples.

Trees whose terminal shoot has only been lightly browsed (end bud bitten off) suffer a smaller loss of height, reserves, and buds than those with heavy browsing. They can also relatively easily grow a new terminal shoot from buds or prematurely sprouted branches (prolepsis) on the remaining part of the terminal shoot [6, 33]. Lateral shoot browsing, on the other hand, does not have a direct impact on tree height or on the number of meristems on the terminal shoot. However, if many lateral shoots are browsed, reserve material in older needles or branches is lost [34], which can reduce the height increase in the following years [35, 36]. If, in an extreme case, all lateral shoots of a tree are browsed, the tree can no longer respond by bending lateral shoots upward, the most common type of response in Norway spruce trees severely browsed at the terminal shoot [7]. For the growth of the individual tree, however, the primary deciding factor is generally whether only the terminal bud or a long piece of the terminal shoot is browsed [37, 38, 39]. The within-tree browsing intensity is therefore a critical variable for assessing the loss of height (increment) and the ability of browsed trees to respond.

In detailed studies on terminal shoot browsing and regrowth in the cantons of St. Gallen and Graubünden in Switzerland, it was found that the terminal shoots of tree saplings were not browsed to the same extent in various forest types [40]. This had an impact on height increment loss in the year of browsing and on the height increments in the years thereafter [40]. The within-tree browsing intensity also affected the relative height increment (see Box 1) of the differently browsed tree species (Figure 4). For example, in a European beech (Fagus sylvatica) forest, lightly browsed silver fir grew better than Norway spruce, which was never browsed in these forests. Heavily browsed silver fir, on the other hand, fell significantly behind spruce in height growth (Figure 4, left). Only in the case of a greatly reduced height increment due to browsing (and/or a considerably delayed browsing response) can browsing be expected to change the growth rate ranking between species and thus have an impact on the entire tree regeneration process in the long term. That is, if lightly browsed silver fir trees grow better than Norway spruce despite browsing, they will not die as an indirect result of browsing. As half of the browsed silver firs in the studied beech forest were only lightly browsed, the number of “substantially” impacted saplings was also halved (Figure 4, left). With the classical browsing intensity approach, the impact of browsing on tree regeneration is significantly overestimated in such cases.

Advertisement

Box 1.

Relative height increment: definition, calculation, and interpretation.

By relative height increment, we refer to the length of the terminal shoot formed in the preceding year divided by the tree height. In the case of unbrowsed trees, the terminal shoot length is measured to the base of the terminal bud, while after winter browsing the remaining piece of the browsed shoot is measured.

The ratio of the relative height increments of two tree species can be calculated if an individual of the same height class is present for both species. For example, to calculate the ratio for silver fir and Norway spruce, the saplings closest to the center of the plot are considered. First, the relative height increment is calculated for each of these saplings (considering differences in tree height). Then, for each sample plot and height class, the relative height increment of the more frequently browsed tree species is divided by that of the less frequently browsed species. In the example, the relative height increment of silver fir is divided by the relative height increment of Norway spruce. If this ratio is equal to 1 (gray horizontal line in Figure 4), the two tree species have the same relative height increment, that is, they grow equally well in the plots. For values >1, the silver fir grows better than the Norway spruce in this example, and for values <1 it is the other way round. In the longer term, a ratio clearly >1 or clearly <1 leads to the loss of one of the two tree species. Regarding the influence of browsing, it is useful to know if the ratio falls from >1 or 1 to <1 as a result of browsing, in which case loss of the more frequently browsed tree species is to be expected over the long term. Figures depicting the relative growth rate between species on the y-axis and the within-tree browsing intensity of the selected species on the x-axis (e.g., Figure 4) illustrate whether and from which within-tree browsing intensity the browsing has an impact on the selected species and thus on the growth chances of the different tree species.

Figure 4.

Example of results for the ratio of the relative height increment of silver fir (Abies alba) to the relative height increment of Norway spruce (Picea abies) from two indicator areas (about 20–30 sampling plots) in Switzerland, with 86 (beech forest at Hasenstrick) and 51 (beech-fir forest at Spaltenstein) fir-spruce pairs, ordered by within-tree browsing intensity of silver fir (see box for calculation). Norway spruce were unbrowsed. Box plots feature quartiles (box boundary), median (thick horizontal line), ranging from 10th to 90th percentile (dashed lines) and outliers (dots). The width of the boxes is proportional to the number of measured tree pairs. At the gray line, the relative height increments of silver fir and Norway spruce are equal (ratio of 1). Source: Adapted from supplementary material of Ref. [40].

Compared with beech forests, most of the fir-beech and fir-spruce forests studied by Kupferschmid [40] contained fewer silver fir saplings per sample plot, and their terminal shoots were more heavily browsed. Therefore, most of the browsed silver fir saplings grew significantly less than the unbrowsed ones, though unbrowsed silver fir saplings grew as well as unbrowsed Norway spruce (Figure 4, right). This favored spruce over silver fir in terms of height growth [40]. In the longer term, this situation may lead to the dominance of Norway spruce and loss of silver fir. Such a browsing-induced growth difference needs to be assessed with measurements (e.g., comparisons between fenced and unfenced plots, i.e., [41]) to determine if there is indeed a “substantial” effect of browsing on tree regeneration.

When browsing intensity is used as the only measure of the effect of browsing on tree regeneration, such a “substantial” effect is implicitly assumed because only the presence of terminal shoot browsing is considered. While this issue can be studied directly in the forest using “expert assessments,” an objective evaluation based on sample surveys requires information on the within-tree browsing intensity and growth rate ranking of the different tree species. The phenomenon described above (Figure 4) did not occur in all of the beech forests or in fir-beech and fir-spruce forests in the study [40], because the within-tree browsing intensity depends on the density of tree seedlings and saplings, tree species, site conditions, stage of development, and likely many other factors [40]. It is therefore important, in our view, to determine the within-tree browsing intensity with surveys.

For both spring and autumn, surveys using four to five categories of within-tree browsing intensity (Table 1) have been carried out at various locations and for various stages of development (e.g., [24, 40, 42, 43, 44]).

If the number of saplings that are lightly browsed on the terminal shoot and the number of saplings heavily browsed on the terminal shoot are added together and divided by the number of all saplings, the “classical” browsing intensity can be calculated. As described above, however, separation into light and heavy browsing—is helpful for identifying the cases in which the impact of browsing on tree saplings is relevant ecologically or for forestry.

The additional separation of saplings without a terminal shoot (Figure 8) provides evidence of the ability of the tree saplings of different tree species to respond after terminal shoot loss. Thus, if there is a time lag until new shoot formation, as often reported for silver fir (e.g., [7, 31, 37]). In addition, the separate category “no shoot” enables a different assignment of these trees to that of the unbrowsed (according to instructions in indicator plots, pers. communication D. Rüegg) or browsed trees (as recommended by Ref. [45]).

The category “other damage” (Table 1) gives an indication of further damage (e.g., from frost, insects, rabbits, mice, or lumbering) to trees at a given site and thus represents the relative frequency of browsing versus other damage. For instance, in some places in Switzerland, more frost damage than browsing impact on Norway spruce saplings was found [46].

Suitable methods for assessing within-tree browsing intensity are presented below (see also Table 2) and in the next section (Table 3). The usual survey of the presence or absence of a terminal shoot can be refined without much extra effort by assessing the within-tree browsing intensity in categories (Table 2).

Tree speciesHeight classWithin-tree browsing intensity
not browsedlightly browsedheavily browsedno terminal shoot present or other damage
Abies alba1IIIIII
2III
3IIII
4
Picea abies1III
2I
3I
4II

Table 2.

Excerpt from a fictitious survey table for records in circular sample areas, in which within-tree browsing intensity is addressed using four categories.

Tree speciesHeight classDistance [m]Tree height [m]Height increment [m]Within-tree browsing intensityFraying/bark strippingInsect damage
SummerWinter
Abies alba10.5122.50000
12.510.50.50200
20.8482.00200
20.94514.00100
Picea abies11.32204300
14.513.530000
2
2

Table 3.

Excerpt from a fictitious survey table for recording the nearest two saplings per tree species and height class to the plot center, including the within-tree browsing intensity in five categories (0 = not browsed, 1 = lightly browsed on terminal shoot, 2 = heavily browsed on terminal shoot, 3 = no terminal shoot present, and 4 = other damage on terminal shoot; definitions given in Table 1), including a distinction between summer and winter browsing. Here, other characteristics, such as fraying, bark stripping, and insect feeding on each sapling, can be integrated easily. For each individual tree, a single line is filled in.

Advertisement

4. Height (and height increment) loss due to browsing

Site conditions strongly influence the height increment of tree species and are therefore also decisive for the influence of ungulate browsing on tree regeneration [47]. Within-tree browsing intensity may indicate shifts in the growth rate ranking between tree species at a site (Figure 5), but only when combined with measurements of the effective annual height increment does it enable estimation of ungulate effects (Figure 4 and Box 1). In survey procedures in which the k nearest saplings to the plot center in each height class are assessed, it is possible to easily integrate measurements of height and height increment (approximated by terminal shoot length; Table 3 and Swiss NFI4 for an example [18]). As a result of differences in: (i) branching pattern, (ii) visibility of bud scars, and (iii) second and third flushes, the annual height increment may be assessed faster and with less error for certain tree species than for others. In most forests in Switzerland, many of the tree species with height increments that are relatively easy to measure are of particular interest for the estimation of browsing-induced species loss, such as P. abies (Norway spruce), A. alba (silver fir), Pinus sylvestris (Scots pine), Pinus mugo (mountain pine), Acer pseudoplatanus (sycamore maple), Acer platanoides (Norway maple), and Fraxinus excelsior (European ash). These measurements are more difficult with Larix decidua (European larch), Fagus sylvatica (European beech), Quercus petraea (sessile oak), Quercus pubescens (downy oak), Quercus robur (common oak), and Sorbus aucurparia (rowan). This is particularly true for European beech, due to the frequent formation of second to fifth flushes [48]. In principle, it is acceptable to reduce the height increment measurements to one or two preferably browsed main tree species (e.g., silver fir, maple, and ash) and one less frequently browsed species (usually spruce or, depending on the forest type, pine) to estimate the effect of browsing on species-specific height growth (Figure 4). For comparisons with beech, measurements in autumn are recommended, as all new flushes have leaves at that point, and thus, the delimitation of years can be seen.

Figure 5.

Light browsing is not a decisive influencing factor if the lightly browsed species (silver fir in the photo) grows at least as well as or even better than an unbrowsed species (Norway spruce in the photo) occurring at the same site. Photo by A. Kupferschmid.

Advertisement

5. Grow-through time: time needed for a tree to grow out of the reach of browsers

If the annual height increment in different tree height classes is known, it is possible to estimate the amount of time during which tree saplings are vulnerable to grazing by wild ungulates. Thus, the approximate grow-through period (“hazard period” according to [17]) can be calculated. If the height increment is small, as in many mountain forests, the trees are browsed for a longer time span than if the height increment is large. This is important for estimating the effect of browsing. If, for example, it takes 10 years for the saplings to grow out of the reach of ungulate browsing and an average of 10% of the saplings are browsed on the terminal shoot per year, a sapling will be browsed, on average, once during this phase (assuming an equal distribution of browsing over time). If the grow-through time lasts 50 years with the same browsing intensity, then a sapling will be browsed five times on average. From this, it is easy to calculate the effect if not 10% but 20% or more of the trees are browsed each year.

The more often a sapling is browsed, the greater the effect tends to be. Consecutive (heavy) terminal-shoot browsing has a greater negative effect than one-time browsing [49, 50, 51] because: (i) most of the reserves for the response have already been mobilized in the preceding year and (ii) fewer meristems are available for the response [6]. As explained using the beech-forest example (Figure 4, left), the height increment is underestimated at some sites and thus, the grow-through time is Overestimated, if only unbrowsed saplings are considered in the calculation. Ungulates often eat only the buds of the most vigorous, dominant saplings when tree sapling abundance is high. For instance, if red deer or roe deer are offered different trees under experimental conditions, they choose the largest ones and the ones with the most spreading branches [5]. In forests, this choice of the most vigorous plants [52] often results in a situation where the residual shoot pieces of the lightly browsed terminal shoots are longer than the intact terminal shoots of the unbrowsed, less vigorous neighboring saplings [29, 30]. Therefore, the most vigorous saplings grow better than their neighbors despite browsing (Figure 4, left, [40]). However, their grow-through time is also prolonged, though not as much as with heavy and frequent terminal shoot browsing [46].

If measurements of height increment are omitted, the grow-through times must be estimated based on separate data in order to estimate the long-term impact of browsing on tree regeneration. This is not easy, as other factors must also be considered, such as light and soil conditions and tree species composition. These influences can be neglected when browsed and unbrowsed trees are measured within the same sample plots. In addition, height increment is also used to estimate the height increment loss due to browsing (see above) and is therefore well suited for estimating the average grow-through time, at least with high sapling densities in all height classes.

Advertisement

6. Density of tree saplings

Not every tree sapling is necessary for sufficient tree regeneration. With 10,000 saplings between 10 and 130 cm in height per hectare of forest, there is a greater chance that some will grow into large trees than with 1000 or only 100 saplings, given equivalent browsing frequency and site conditions (Figure 9). With a large number of saplings, there are more individuals in absolute terms that can grow “unhindered” into large trees, due to a variety of factors, such as their particular position (e.g., between tree trunks), their chemical composition, their accessibility, and the fact that they coincidentally have never been browsed, or at least less frequently.

In the study by Ref. [40], within-tree browsing intensity decreased, and thus, the loss of height increment decreased, with increasing density of silver fir saplings, or the density of silver fir saplings was higher at sites with light within-tree browsing intensity. These relationships remain insufficiently explored. However, it seems likely that roe deer, red deer, and chamois are less able to select specific trees when there is little tree regeneration and thus browses more often and more heavily on individual tree saplings (see films in [32]). These repeatedly browsed samplings, therefore, suffer a greater loss of height (and consequently height increment). Conversely, wild ungulates can choose between more trees at higher sapling densities. In addition, when tree saplings are abundant, conditions are often also favorable for other plant species (especially herbaceous plants; [53]), extending further the food choice for ungulates, at least in summer. This in turn reduces the effects of ungulates on tree regeneration.

From the point of view of forest services, it is irrelevant how often browsing occurs. Instead, the number of trees that successfully pass the sapling stage is of interest. If, however, we were to record only the trees that passed the sapling stage and pay no attention to browsing, we would not have an “early warning system” [54, 55]. In this case, we would only be able to assess the final outcome and thus would not be able to implement timely measures to protect and promote tree regeneration or to reduce the browsing rate.

Tree density can be reliably determined with surveys involving: (i) counting all saplings in a sample plot of a fixed size (plot-count method), (ii) counting a fixed number of saplings (e.g., 20 or 30 saplings) per tree species, and measuring the “break-off” azimuth of the circle (variable circle sectors), or (iii) measuring the distance to the k nearest saplings per tree species [26]. Which method is chosen is not critical. What is important regarding method (iii) is that the distance to the plot center of at least two saplings per tree species is measured, so that the density of saplings can be determined [25]. The precision (variation) of the results depends mainly on the selected survey radius or maximum search distance [26]. With a 2 m radius, the density is either 0 (if no sapling was found) or ≥ 796 saplings per ha (= 796 if exactly 1 tree was within the 2 m circle), whereas with a 10 m radius 0 or ≥ 32 saplings per ha would be reported. Therefore, the density estimate depends strongly on the radius or search distance, along with the number of sample plots and the spatial distribution of tree saplings [26].

Advertisement

7. Browsing-induced mortality

Information on browsing-induced mortality is also needed in order to make statements about the effect of browsing on tree regeneration. How many saplings were completely consumed by wild ungulates, and how many were browsed so heavily and/or so frequently that they died? This can be determined by experimentally studying browsing-related mortality. As the basis for their reference value calculations, Eiberle & Nigg [17] assumed, based on fencing experiments, that a silviculturally significant proportion of plants in mountain forests is browsed to the point of mortality when the height increment loss is >25%. However, the browsing-induced height increment loss depends on the site and the within-tree browsing intensity [40]. Therefore, it is also likely that mortality depends on these factors (e.g., Figure 6). At the same browsing intensity, browsing-induced mortality is likely to be significantly higher at nutrient-poor sites where tree growth is slow and heavy terminal shoot browsing occurs than at sites with vigorous growth and light within-tree browsing intensity [47]. If seedlings or small saplings are completely consumed (Figure 7), they are not considered in the calculation of browsing intensity. Thus, the supposedly moderate browsing intensity at some sites with poor growth may be the result of high browsing-related mortality. This leads to an underestimation of the actual browsing influence when assessments are made on the basis of browsing intensity alone. Data on height increment and within-tree browsing intensity can improve estimates. In order to clearly identify browsing-related mortality, either pair of fenced and unfenced plots should be established or tree seedlings and saplings should be individually marked and repeatedly assessed (see the experiment in Ref. [31]).

Figure 6.

Silver fir that has been lightly browsed several times on the terminal shoot (within-tree browsing intensity in the last 2 years = “light”). It continues to grow well and thus suffers almost no height loss and no mortality despite repeated browsing. Photo by A. Kupferschmid.

Figure 7.

Ungulate-induced mortality occurs in particular for seedlings that are consumed completely, heavily browsed down to very small residual pieces (image from the experiment by [31]) or are heavily and repeatedly browsed (see films in [32]).

Figure 8.

Silver fir (Abies alba) without a terminal shoot (“no terminal shoot present” category), presumably as a consequence of earlier heavy terminal-shoot browsing. The trees have grown vertically only through the upward growth of lateral shoots, thus showing a delay of 1 year so far. Photo by A. Kupferschmid.

Figure 9.

If the density of tree saplings is low (as in the foreground of the photo), each sapling is important for the subsequent number of stems in the stand. If, on the other hand, the density is high (as in the gap in the background of the photo), browsing on individual trees is not important, as long as a considerable proportion of the saplings of a target species are not heavily browsed and therefore are not delayed in their height growth compared with unbrowsed tree species. Photo by A. Kupferschmid.

Advertisement

8. Conclusion

The long-term effects of browsing by wild ungulates on tree saplings can only be objectively assessed if data is available on (i) the density of tree saplings, (ii) the within-tree browsing intensity, (iii) the height increment of at least the main tree species and thus the grow-through time and the browsing-induced height increment loss, (iv) the proportion of trees with no terminal shoot and thus the time lags after terminal shoot damage, and (v) the browsing-induced mortality (Figure 1). At least the first four factors can be easily measured/assessed and should be included in future browsing inventories. The within-tree browsing intensity (i.e., whether only the bud or large parts of the terminal shoot are consumed) can be integrated into any browsing survey with little additional effort. Apart from browsing-induced mortality, all of these factors can be objectively recorded or derived by means of repeated surveys of the k nearest trees to the plot center. The combination of these factors provides meaningful information on how wildlife browsing affects tree regeneration (e.g., [24]).

The aim of the proposals given above is to improve the methods used today to assess the influence of ungulate browsing on tree regeneration. Such an approach would enhance the existing data series on browsing intensity by including additional measurements or variables that would make it possible to estimate not only the frequency but also the impact of browsing in future surveys.

Large-scale pilot tests of the measures suggested here were carried out in wildlife sectors in the Swiss canton of St. Gallen in spring 2018 [24] and autumn 2022 [43] and in many national parks in Germany in spring 2019/2020 [44]. The authors are willing to accompany further implementations. Of great importance is the continuity of the data series even after the revision of the survey methods, at least for a transitional period, so that comparisons with previous surveys are possible.

Advertisement

Acknowledgments

We thank M. Dawes for providing linguistic suggestions.

The first version of this paper was published in May 2019 in the Schweizerische Zeitschrift für Forstwesen, DOI: https://doi.org/10.3188/szf.2019.0125 [15].

References

  1. 1. Cornelis J, Casaer J, Hermy M. Impact of season, habitat and research techniques on diet composition of roe deer (Capreolus capreolus): A review. Journal of Zoology. 1999;248:195-207. DOI: 10.1111/j.1469-7998.1999.tb01196.x
  2. 2. Tixier H, Duncan P, Scehovic J, Yant A, Gleizes M, M. L. Food selection by European roe deer (Capreolus capreolus): Effects of plant chemistry, and consequences for the nutritional value of their diets. Journal of Zoology. 1997;242:229-245. DOI: 10.1111/j.1469-7998.1997.tb05799.x
  3. 3. Gill RMA. A review of damage by mammals in North Temperate forests: 3. Impact on trees and forests. Forestry. 1992;65(4):363-388. DOI: 10.1093/forestry/65.4.363-a
  4. 4. Boulanger V, Baltzinger C, Saı S, Ballon P, Picard J-F, Dupouey J-L. Ranking temperate woody species along a gradient of browsing by deer. Forest Ecology and Management. 2009;258:1397-1406. DOI: 10.1016/j.foreco.2009.06.055
  5. 5. Iason GR, Duncan AJ, Hartley SE, Staines BW. Feeding behaviour of red deer (Cervus elaphus) on sitka spruce (Picea sitchensis): The role of carbon-nutrient balance. Forest Ecology and Management. 1996;88(1-2):121-129. DOI: 10.1016/S0378-1127(96)03817-0
  6. 6. Kupferschmid AD. Compensation capacity of Central European tree species in response to leader shoot browsing. In: Menendez A, Sands N, editors. Ungulates: Evolution, Diversity and Ecology. Animal Science, Issues and Research. Hauppauge, New York, USA: Nova Science Publishers; 2017. pp. 1-63
  7. 7. Kupferschmid AD, Heiri C. Recovery of Abies alba and Picea abies saplings to browsing and frost damage depends on seed source. Ecology and Evolution. 2019;2019(9):3335-3354. DOI: 10.1002/ece3.4955
  8. 8. Krueger LM, Peterson CJ, Royo A, Carson WP. Evaluating relationships among tee growth rate, shade tolerance, and browse tolerance following disturbance in an eastern deciduous forest. Canadian Journal of Forest Research. 2009;39:2460-2469. DOI: 10.1139/X09-155
  9. 9. Tremblay J-P, Huot J, Potvin F. Density-related effects of deer browsing on the regeneration dynamics of boreal forests. Journal of Applied Ecology. 2007;44(3):552-562. DOI: 10.1111/j.1365-2664.2007.01290.x
  10. 10. Imesch N, Stadler B, Bolliger M, Schneider O. Biodiversität im Wald: Ziele und Massnahmen. Vollzugshilfe zur Erhaltung und Förderung der Biologischen Vielfalt im Schweizer Wald. Bern: Bundesamt für Umwelt; 2015. p. 186
  11. 11. Apollonio M, Andersen R, Putman R. European Ungulates and their Management in the 21th Century. Cambridge, UK: Cambridge University Press; 2010. p. 604
  12. 12. Reimoser F, Putman R. Impact of wild ungulates on vegetation: Costs and benefits. In: Putman R, Apollonio M, Andersen R, editors. Ungulate Management in Europe: Problems and Practices. Cambridge: Cambrige University Press; 2011. pp. 144-191. DOI: 10.1017/CBO9780511974137.007
  13. 13. Kupferschmid AD, Bütikofer L, Hothorn T, Schwyzer A, Brang P. Ungulate species and abundance as well as environmental factors determine the probability of terminal shoot browsing on temperate forest trees. 2020;11(7):764. DOI: 10.3390/f11070764
  14. 14. Kupferschmid AD, Heiri C, Huber M, Fehr M, Frei M, Gmür P, et al. Einfluss wildlebender Huftiere auf die Waldverjüngung: Ein Überblick für die Schweiz. Schweizerische Zeitschrift für Forstwesen. 2015;166(6):420-431. DOI: 10.3188/szf.2015.0420
  15. 15. Kupferschmid AD, Brang P, Bugmann H. Abschätzung des Einflusses von Verbiss durch wildlebende Huftiere auf die Baumverjüngung. Schweizerische Zeitschrift für Forstwesen. 2019;170(3):125-134. DOI: 10.3188/szf.2019.0125
  16. 16. Berwert-Lopes R. Assessment of tolerable browsing by Eiberle’s method: Limitations and future prospects. Forest Ecology and Management. 1996;88(1-2):87-91. DOI: 10.1016/S0378-1127(96)03813-3
  17. 17. Eiberle K, Nigg H. Grundlagen zur Beurteilung des Wildverbisses im Gebirgswald. Schweizerische Zeitschrift für Forstwesen. 1987;138:747-785. DOI: 10.5169/seals-766065
  18. 18. Keller M, editor. Schweizerisches Landesforstinventar, Feldaufnahmeanleitung 2013. Eidg. Forschungsanstalt WSL: Birmensdorf; 2013
  19. 19. Rüegg D. Verjüngungskontrolle: Methoden Stichproben in Indikatorflächen. Kaltbrunn: Dani Rüegg; 2008. p. 7
  20. 20. Reimoser F, Schodterer H, Reimoser S. Erfassung und Beurteilung des Schalenwildeinflusses auf die Waldverjüngung – Vergleich verschiedener Methoden des Wildeinfluss-Monitorings (“WEM-Methodenvergleich”). Vol. 17. Wien: Bundesforschungs- und Ausbildungszentrum für Wald, Naturgefahren und Landschaft; 2014. Report No.: BFW-Dokumentation Nr. 17. p. 177
  21. 21. Keller M, editor. Schweizerisches Landesforstinventar, Anleitung für die Feldaufnahmen der Erhebung 2004-2007. Eidg. Forschungsanstalt WSL: Birmensdorf; 2005
  22. 22. STMELF. Anweisung für die Erstellung der Forstlichen Gutachten zur Situation der Waldverjüngung 2018. München: Bayerisches Staatsministerium Ernährung Landwirtschaft Forsten; 2017. p. 56
  23. 23. Eiberle K, Bucher H. Interdependenzen zwischen dem Verbiss verschiedener Baumarten in einem Plenterwaldgebiet. Zeitschrift für Jagdwissenschaften. 1989;35:235-244. DOI: 10.1007/bf02241661
  24. 24. Kupferschmid AD, Gmür PA. Einschätzung des Verbisseinflusses in zwei Wildräumen des Kantons St. Gallen. Schweizerische Zeitschrift für Forstwesen. 2020;171(2):79-90. DOI: 10.3188/szf.2020.0079
  25. 25. Kleinn C, Vilčko F, Fehrmann L, Hradetzky J. Zur Auswertung der k-Baum-Probe. [About the evaluation of k-trees samples]. Allgemeine Forst- und Jagdzeitung. 2009;180:228-237
  26. 26. Huber M, Schwyzer A, Kupferschmid AD. A comparison between plot-count and nearest-tree method in assessing tree regeneration features. Current Trends in Forest Research. 2018;CTFR-122(4):10002. DOI: 10.29011/2638-0013
  27. 27. Eiberle K, Lanz A. Zur Erhebung des Wildverbisses mittels Stichproben. Schweizerische Zeitschrift für Forstwesen. 1989;140:171-187. DOI: 10.5169/seals-764217
  28. 28. Zweifel-Schielly B, Leuenberger Y, Kreuzer K, Suter W. A herbivore’s food landscape: Seasonal dynamics and nutritional implications of diet selection by a red deer population in contrasting Alpine habitats. Journal of Zoology. 2012;286:68-80. DOI: 10.1111/j.1469-7998.2011.00853.x
  29. 29. Kupferschmid AD, Zimmermann S, Bugmann H. Browsing regime and growth response of naturally regenerated Abies alba saplings along light gradients. Forest Ecology and Management. 2013;310:393-404. DOI: 10.1016/j.foreco.2013.08.048
  30. 30. Kupferschmid AD, Wasem U, Bugmann H. Ungulate browsing leads to height and diameter growth reduction on Abies alba saplings planted along light gradients. European Journal of Forest Research. 2015;134(1):75-87. DOI: 10.1007/s10342-014-0834-2
  31. 31. Kupferschmid AD, Wasem U, Bugmann H. Light availability and ungulate browsing determine growth, height and mortality of Abies alba saplings. Forest Ecology and Management. 2014;318:359-369. DOI: 10.1016/j.foreco.2014.01.027
  32. 32. Kupferschmid AD, Wasem U. Filme zum Verbiss an Weisstannen. ETH e-collection: ETH Zürich; 2014. DOI: 10.3929/ethz-a-010075029
  33. 33. Kupferschmid AD, Bugmann H. Timing, light availability and vigour determine the response of Abies alba saplings to leader shoot browsing. European Journal of Forest Research. 2013;132:47-60. DOI: 10.1007/s10342-012-0653-2
  34. 34. Millard P, Hester AJ, Wendler R, Wendler B. Interspecific defoliation responses of trees depend on sites of winter nitrogen storage. Functional Ecology. 2001;15:535-543
  35. 35. Ericsson A, Larsson S, O. T. Effects of early and late season defoliation on growth and carbohydrate dynamics in Scots pine. Journal of Applied Ecology. 1980;17(3):747-769. DOI: 10.2307/2402653
  36. 36. Honkanen T, Haukioja E, Suomela J. Effects of simulated defoliation and debudding on needle and shoot growth in Scots pine (Pinus sylvestris): Implications of plant source/sink relationships for plant-herbivore studies. Functional Ecology. 1994;8:631-639
  37. 37. Häsler H, Senn J, Edwards PJ. Light-dependent growth responses of young Abies alba to simulated ungulate browsing. Functional Ecology. 2008;22:48-57
  38. 38. Mayhead GJ, Jenkings TAR. Growth of young Sitka spruce (Picea sitchensis (Bong.) Carr.) and the effect of simulated browsing, staking and treeshelters. Forestry. 1992;65(4):453-462
  39. 39. Metzger FT. Sugar maple and yellow birch seedling growth after simulated browsing. USDA Forest Service Research Paper NC. 1977;140:1-6
  40. 40. Kupferschmid AD. Selective browsing behaviour of ungulates influences the growth of Abies alba differently depending on forest type. Forest Ecology and Management. 2018;429:317-326. DOI: 10.1016/j.foreco.2018.06.046
  41. 41. Reimoser F. Verbiss-Kontrollgatter: Eine Methode zur objektiven Erfassung des Einflusses von Schalenwild und Weidevieh auf die Waldverjüngung (System Voralberg). Bündnerwald. 1992;45(3):21-45
  42. 42. Kupferschmid AD, Lachat T. Einfluss des Wildverbisses auf die Verjüngung: ein Tutorial [Film]. DOI: 10.24451/arbor.11872
  43. 43. Kupferschmid AD. Verbisseinfluss - Monitoring in Buchenwäldern. Birmensdorf: Eidg. Forschungsanstalt WSL. 2022. Available from: https://www.wsl.ch/de/projekte/verbisseinflussmonitoring-in-buchenwaeldern.html
  44. 44. Fiderer C, Heurich M. F+E Vorhaben Schalenwildmonitoring in den Deutschen Nationalparken. Grafenau: Nationalparkverwaltung Bayerischer Wald; 2019. Available from: https://www.nationalpark-bayerischer-wald.bayern.de/forschung/projekte/schalenwildmonitoring.htm
  45. 45. Reimoser F. Anmerkungen zur Feststellung von Wildverbiss und zum Vergleich von Verbisskennzahlen. Zeitschrift für Jagdwissen. 2000;46:51-56. DOI: 10.1007/BF02240664
  46. 46. Kupferschmid AD. Verjüngungskontrolle St. Gallen: Auswertungen zur Piloterhebung 2018 in den Wildräumen 2 und 8. Birmensdorf: Eidg. Forschungsanstalt WSL; 2019. p. 103. DOI: 10.13140/RG.2.2.33154.02242
  47. 47. Edenius L, Danell K, Nyquist H. Effects of simulated moose browsing on growth, mortaltiy, and fecundity in Scots pine: Relations to plant productivity. Canadian Journal of Forest Research. 1995;25:529-535. DOI: 10.1139/x95-060
  48. 48. Frank A, Heiri C, Kupferschmid AD. Growth and quality of Fagus sylvatica saplings depend on seed source, site, and browsing intensity. Ecosphere. 2019;10(1):1-19. DOI: 10.1002/ecs2.2580
  49. 49. Eiberle K. Folgewirkungen eines simulierten Wildverbisses auf die Entwicklung junger Waldbäume. Schweizerische Zeitschrift für Forstwesen. 1978;129(9):757-768. DOI: 10.5169/seals-764381
  50. 50. Kristöfel F, Pollanschütz J. Entwicklung von Fichtenpflanzungen nach Triebrückschnitten. Berichte Forstliche Bundesversuchsanstalt Wien, Waldforschungszentrum. 1995;85:1-16
  51. 51. Wallgren M, Bergquist J, Bergström R, Eriksson S. Effects of timing, duration, and intensity of simulated browsing on Scots pine growth and stem quality. Scandinavian Journal of Forest Research. 2014;29(8):734-746. DOI: 10.1080/02827581.2014.960896
  52. 52. Price PW. The plant vigor hypothesis and herbivore attack. Oikos. 1991;62:244-251. DOI: 10.2307/3545270
  53. 53. Kupferschmid AD, Bütikofer L, Hothorn T, Schwyzer A, Brang P. Quantifying the relative influence of terminal shoot browsing by ungulates on tree regeneration. Forest Ecology and Management. 2019;446:331-344. DOI: 10.1016/j.foreco.2019.05.009
  54. 54. Odermatt O, Rüegg D. Planung im Bereich Wald Wild - Verbissprozente im Praxistest. Wald und Holz. 2007;4:55-57
  55. 55. Rüegg D, Burger T, Brang P. Methoden zur Erhebung und Beurteilung des Wildeinflusses auf die Waldverjüngung. Wald und Wild – Grundlagen für die Praxis Wissenschaftliche und methodische Grundlagen zum integralen Management von Reh, Gämse, Rothirsch und ihrem Lebensraum Umwelt-Wissen Nr. 1013. Bern: Bundesamt für Umwelt BAFU (Hrsg.). 2010:67-92

Written By

Andrea Doris Kupferschmid, Raphaël Greilsamer, Peter Brang and Harald Bugmann

Submitted: 18 September 2022 Reviewed: 19 October 2022 Published: 16 November 2022