Content uploaded by Daniel Elias
Author content
All content in this area was uploaded by Daniel Elias on Oct 29, 2018
Content may be subject to copyright.
215
Tuexenia 38: 215–233. Göttingen 2018.
doi: 10.14471/2018.38.017, available online at www.zobodat.at
Goat paddock grazing improves the conservation status of
shrub-encroached dry grasslands
Verbesserung des Erhaltungszustandes von verbuschten Trockenrasen
durch Ziegenbeweidung auf Rotationsstandweiden
Daniel Elias1, *, Norbert Hölzel2 & Sabine Tischew1
1Anhalt University of Applied Sciences, Department for Nature Conservation and Landscape Planning,
Strenzfelder Allee 28, 06406 Bernburg, Germany;
2University of Münster, Institute of Landscape Ecology, Heisenbergstraße 2, 48149 Münster, Germany;
*Corresponding author, e-mail: Daniel.Elias@hs-anhalt.de
Abstract
European dry grasslands are predominantly semi-natural communities. Commonly, they have been
used as pastures for sheep and goats. Due to their richness in biological diversity and the large number
of threatened species, they are recognized as being of high conservation value. However, in the last few
decades, a dramatic decline in dry grassland areas has been observed throughout Europe. Pasture aban-
donment followed by grass and shrub encroachment are the main causes of dry grassland loss, especial-
ly at isolated marginal sites. The introduction of goat paddock grazing can be an efficient method for
restoring shrub-encroached dry grasslands, but most recommendations are still based on anecdotal
evidence. Hence, this study investigated the impact of goat paddock grazing with a relatively high
grazing pressure (0.6 to 0.8 LU/ha/yr) over seven years on habitat structure and species richness in six
encroached dry grassland localities of the lower Saale River valley in Central Germany. In each pad-
dock, we analyzed 25-m² plots in highly (initial woody species cover of ≥ 25%) and less encroached
areas (initial woody species cover < 25%) and compared the development of woody species, grass
encroachment indicators, litter cover and the share of bare soil as well as the coverage and number of
target and ruderal species with that of ungrazed control plots.
We observed the following effects of restorative goat paddock grazing: Woody plants and litter lay-
er cover declined significantly in all paddock plots. In addition, the share of bare soil increased signifi-
cantly in the less encroached paddock plots. The number of target species increased significantly in the
formerly highly encroached plots, whereas the already high target species richness slightly increased on
originally less encroached plots. Besides short-lived target species, even red list species were able to
colonize the grazed plots. Furthermore, we observed a slightly increasing number of ruderal species, but
at a very low level of coverage. An opposite trend was recorded in the ungrazed plots: The cover of
woody and competitive grass species, as well as litter coverage, increased significantly over time.
While the cover of target species decreased significantly in the highly encroached plots, the number of
target species stayed stable in both types of control plots, showing the restoration potential of the en-
croached grasslands.
Based on our findings, we conclude that goat paddock grazing with a relatively high grazing pres-
sure can be an effective tool to restore encroached dry grasslands. The reduction of encroachment and
the increasing number of target species correlates with the improved conservation status of these highly
valuable dry grassland habitat types. Finally, recommendations for the implementation of restorative
goat grazing are given.
Manuscript received 26 February 2018, accepted 04 June 2018
Co-ordinating Editor: Péter Török
216
Keywords: bare soil, browsing, goat paddocks, litter, livestock, marginal sites, target plant species,
restoration, trampling, woody plants
Erweiterte deutsche Zusammenfassung am Ende des Artikels
1. Introduction
Due to their high biodiversity (VEEN et al. 2009, WILSON et al. 2012) and the large num-
ber of threatened species (KORNECK et al. 1998, VAN SWAAY et al. 2006), dry grasslands are
recognized as being of high conservation value (CALACIURA & SPINELLI 2008, DENGLER et
al. 2014, JANSSEN et al. 2016). Dry grasslands mainly originated from a long history of sea-
sonal grazing by sheep and goats, occasionally in combination with other land use practices
(e.g., mowing) (POSCHLOD & WALLISDEVRIES 2002, POSCHLOD et al. 2009, ELLENBERG &
LEUSCHNER 2010). Shrub encroachment resulting from pasture abandonment is considered
to be an important driving factor for the unfavourable conservation status of many of these
habitats, especially in isolated marginal areas (WALLISDEVRIES et al. 2002, DIERSCHKE
2006, CALACIURA & SPINELLI 2008, HEGEDUŠOVÁ & SENKO 2011, EUROPEAN COMMISSION
2015, DEÁK et al. 2016). Similarly, inappropriate late summer grazing at low intensity may
also result in the spread of competitive tall grasses such as Arrhenatherum elatius, Brachy-
podium pinnatum or Bromus erectus, increasing litter accumulation and species loss (JÄGER
& MAHN 2001, WEDL & MEYER 2003, DOSTÁLEK & FRANTÍK 2012).
Many semi-natural dry grassland communities are listed as habitat types of community
interest in Annex I of the EU Habitats Directive (EUROPEAN COMMISSION 2013). These
habitat types have to be maintained at or, where appropriate, restored to a favourable conser-
vation status. Manual shrub removal can be a temporary solution to reduce shrubbery, but
must be repeated in relatively short intervals because of the fast re-growth of many woody
species (BACON 2003, MACCHERINI et al. 2007, ELIAS et al. 2014). In addition, mechanical
shrub removal requires a huge effort regarding manpower/machinery, and it is often ex-
tremely expensive particularly on steep slopes (recently up to 8000 €/ha in the lower Saale
River valley). Thus, in order to restore and maintain these dry grasslands, management plans
should be developed based on historical land use in which goat pasture often played a prom-
inent role, especially at marginally productive sites on steep slopes (GLAVAC 1983). Current-
ly, due to the preference of goats for browsing than grazing, they are commonly used to
control shrub encroachment worldwide (STRANG 1973, LUGINBUHL 1999, HOLST et al. 2004,
SMART et al. 2006, CELAYA et al. 2010, ASCOLI et al. 2013), but in Central Europe only
a few studies have dealt with goat grazing, especially in dry grasslands (RAHMANN 2000,
CZYLOK et al. 2013). Sheep grazing is often still preferred, even at heavily encroached sites.
Hence, there is little information about the effects of goats as browsers and grazers for bio-
diversity conservation in semi-natural dry grasslands.
For reducing shrubbery and dense grass cover, grazing with a comparatively high stock-
ing rate and an early start of grazing in spring is mandatory during a restoration phase of the
first five to seven years (CROFTS & JEFFERSON 1999, ELIAS & TISCHEW 2016). Experiences
in Central Germany has shown, that at small and isolated dry grassland localities, the eco-
nomic feasibility of such a grazing regime can only be achieved by paddock grazing with
permanent fences. However, nature conservationists are often concerned that paddock graz-
ing and an early start in spring may lead to negative effects through excessive trampling and
the spread of nitrophilous ruderal plant species, and finally to the loss of valuable target
species.
217
In this study, we evaluated the success of goat grazing in the restoration of strongly
encroached dry grasslands in Central Germany that still hold a high number of endangered
species. We focused on the following specific questions: (1) Can shrub and grass
encroachment be reduced significantly by restorative goat grazing and how do litter layer
and bare soil coverage develop under goat grazing? (2) How does goat grazing affect the
number and coverage of target species (characteristic species of dry grassland habitat types
and red list species) of dry semi-natural grasslands? (3) How does restorative goat grazing
affect the number and coverage of ruderal species?
2. Study area and goat grazing pilot project
The study was carried out in the lower Saale River valley between Halle and Könnern in
the south of the German federal state of Saxony-Anhalt in Central Germany. The climate is
sub-continental with an average annual precipitation of 476 mm and a mean annual tempera-
ture of 9.0 °C (climatic station Halle; see REICHHOFF et al. 2001). Elevations range from 70
to 150 m above sea level.
Sheep and goat livestock have been present since Neolithic times in Central Germany
(BENECKE 1994). The tradition of migratory herding of sheep and goats in combination with
varying geomorphological and soil conditions resulted in the formation of attractive mosaics
of different dry grassland habitat types. Some of the characteristic plant communities are
listed as natural habitat types in Annex I of the Habitats Directive (e.g., 6210 Semi-natural
dry grasslands on calcareous substrates, 6240* Sub-pannonic steppic grasslands).
Socio-economic changes after German reunification in 1990 led to an accelerated aban-
donment of sheep grazing, whereas goat breeding had already been in decline since the
middle of the last century (JÄGER & MAHN 2001, RICHTER et al. 2003). These actions have
resulted in species-rich dry grasslands being heavily endangered by shrub and grass en-
croachment.
However, species-rich remnants of dry grassland habitats as well as numerous popula-
tions of red list plants were still present on a number of sites when we started our study in
the lower Saale River valley in 2006. All sites were situated on steep slopes and partly sur-
rounded by dense shrubbery or woodland making migratory herding laborious or even im-
possible. Therefore, a pilot project for goat paddock grazing was initiated in cooperation
with local farmers. The paddocks were enclosed using permanent pasture fences to make
grazing of such marginal and isolated sites economically feasible. Farmers were urged to
pasture with high grazing pressure to reduce shrub and grass encroachment. Additionally,
they were invited to avoid supplementary feeding (except for minerals) which was largely
respected. The first paddocks were established in 2007. Over time, a total of 16 pastures are
being grazed by goats, partially together with sheep or cattle, supported by our model project
within the Saale River valley.
3. Material and methods
3.1 Experimental design
The study was conducted at six paddocks (Table 1). Since we studied grazing management within
a real-life pilot project, where the paddocks were gradually established, the starting point of grazing
varied between paddocks, but have consistently lasted at least seven years up to 2018. Before the graz-
ing started, all sites were abandoned for many years and characterized by a mosaic of open grassland
218
patches and patches of more or less intensive shrub encroachment. The woody vegetation was dominat-
ed by thermophilic shrub communities (Berberidion), which include different thorny and spiny species
such as Berberis vulgaris, Crataegus spp., Prunus spinosa and Rosa spp. (shrub height was on average
2 meter, but some individuals reached 6.5 meter). These species are typical for abandoned pastures in
the whole region. Grass encroached parts of the paddocks were dominated especially by Arrhenatherum
elatius, Bromus erectus, Brachypodium pinnatum, Festuca rupicola and Poa angustifolia (compare also
PARTZSCH 2000). Despite the partly heavy encroachment, the paddocks were still characterized by
remnants of species rich dry grasslands (Table 1). Moreover, they still harboured populations of german
(KORNECK et al. 1996) or regional (FRANK et al. 2004) red list species such as Astragalus exscapus,
Oxytropis pilosa, Scabiosa canescens or Seseli hippomarathrum. The dry grassland communities of the
lower Saale River valley were described in detail by PARTZSCH (2000) and ELIAS et al. (2015). ELIAS
et al. (2015) included also a description of the goat paddock Nelbener Grund.
All investigated paddocks were grazed by Boer goats and crossbreeds. The grazing season extended
generally from March–April to October–November depending on weather conditions and subsequent
fodder availability. The paddock size varied between 1.0 and 8.3 ha (Table 1) and the annual stocking
rate ranged from approximately 0.6 to 0.8 LU/ha/yr, meaning 6–8 goats per hectare in relation to
a grazing period of eight months.
Vegetation development was analyzed on 25-m² permanent plots. The plots were established on
southeast to southwest exposed slopes in formerly less intensive (initial woody coverage < 25%) and
intensive encroached parts (initial woody coverage ≥ 25%) inside each paddock as well as outside each
paddock in abandoned dry grasslands (maximum distance 100 meter). We analyzed one paddock plot as
well as one control plot per structure type (in total 24 plots per year, 6 paddocks x 2 structural types x 2
grazing treatments) over a time period of seven years. The locations of the analyzed plots were selected
on the basis that the paddock and the control plots showed similar habitat features (topography, soil
depth, species composition).
Table 1. Characterization of the paddocks, initial level of shrub encroachment (LSE), target habitat
types and number of red list species (RL spec.: includes only grasses and forbs; KORNECK et al. 1996,
FRANK et al. 2004).
Tabelle 1. Charakterisierung der Weideflächen, Anfangsverbuschungsgrad (LSE), Ziel-Lebensraum-
typen und Zahl der vorhandenen Rote-Liste-Arten (RL spec.: nur Gräser und Kräuter; KORNECK et al.
1996, FRANK et al. 2004).
Paddock Geographical position Size
(ha)
Start of
observ.
LSE
(%)
Target Habitat types of
Annex I
RL
spec.
Nelbener Grund 51°39'60"N, 11°45'00"E 8.3 2008 40–50 6110*, 6130, 6210, 6240* 30
Zickeritz
51°38'54"N, 11°44'52"E
2.3
2008
40–50
6210, 6240*
5
Tannengrund
51°38'22"N, 11°45'27"E
4.8
2011
20–30
6210, 6240*
18
Dobis
51°37'30"N, 11°45'20"E
6.1
2007
60–70
6210, 6240*, 8230
12
Mücheln 51°34'45"N, 11°50'05"E 1.0 2007 20–30 4030, 6210, 6240*, 8230 14
Salzatal
51°29'45"N, 11°47'04"E
4.9
2007
50–60
6210, 6240*
18
4030: European dry heaths
6110*: Rupicolous calcareous or basophilic grasslands of the
Alysso-Sedion albi
6130: Calaminarian grasslands of the Violetalia calaminariae
6210: Semi-natural dry grasslands and scrubland facies on calcareous substrates (Festuco-Brometalia)
6240*: Sub-pannonic steppic grasslands
8230: Siliceous rock with pioneer vegetation of the
Sedo-Scleranthion
or of the
Sedo albi-Veronicion dillenii
219
With the beginning of goat grazing, the percentage cover of all species and layers was surveyed
annually (beginning of June to end of July) for seven years for each plot. Furthermore, we recorded
litter cover and the share of bare soil. Both parameters are good indicators of management intensity
assuming that high grazing intensity corresponds with low litter cover and a high proportion of bare
soil, while adandoment results in high litter cover and a low share of bare soil.
Nomenclature of vascular plants follows JÄGER (2011). When it was possible, specimens were
identified at the species level. Due to the difficulty of determination under the influence of grazing,
individuals of Rosa, Rubus and Cerastium were only identified to genus. Furthermore, due to the fre-
quent occurrence of the hybrid Potentilla x subarenaria in our study region (FRANK 2016), P. cinerea
subsp. incana and P. neumanniana were pooled together.
3.2 Data analysis
In line with the study objectives, we first evaluated the cover of all woody species (including the
genus Rubus L.) and the grass encroachment indicators (cumulative cover of Arrhenatherum elatius,
Brachypodium pinnatum, Bromus erectus, Festuca rubra, F. rupicola, Poa angustifolia).
Second, we selected particular target species to evaluate the species compostion. We defined as
target species all species that are characteristic of the following habitat types: 4030 (European dry
heaths), 6110* (Rupicolous calcareous or basophilic grasslands of the Alysso-Sedion albi), 6130
(Calaminarian grasslands of the Violetalia calaminariae), 6210 (Semi-natural dry grasslands and
scrubland facies on calcareous substrates, Festuco-Brometalia), 6240* (Sub-pannonic steppic
grasslands), 8230 (Siliceous rock with pioneer vegetation of the Sedo-Scleranthion or of the Sedo albi-
Veronicion dillenii) (SCHUBOTH & FRANK 2010), or that are listed in the red lists of Germany
(KORNECK et al. 1996), or Saxony-Anhalt (FRANK et al. 2004). The competitive grass species
B. pinnatum, B. erectus and F. rupicola, that often formed dense swards when grazing sites had been
abandoned, were analysed in the group of grass encroachment indicators although they are
characteristic for habitat type 6210.
Third, we selected ruderal species to evaluate possible negative effects of grazing disturbance.
Ruderal species are nitrophilous plant species which belong to the Agropyretea repentis, Sisymbrietea
officinalis or Artemisietea vulgaris. The classification of ruderal species was derived from literature
(SCHUBERT 2001, JÄGER 2011).
Since the data were not normally distributed (even after data transformation) and because of the
small sample size, non-parametric Friedman tests were applied to evaluate the development of habitat
structure and species group variables within the paddocks and ungrazed plots. Furthermore, we used
Mann-Whitney U-tests to compare the values between grazed and ungrazed plots in each year. All
analyses were conducted using IBM SPSS Statistics version 22.0.
4. Results
4.1 Development of habitat structure parameters
Within seven years, the browsing activity of goats resulted in a significant decrease of
woody species cover within paddocks, both in less and more intensively encroached plots
(Table 2). On the contrary, the coverage of woody species increased significantly in un-
grazed controls in both structure types (Table 3). This was confirmed by the paired U-tests
too; woody coverage showed no significant difference between the grazed and control plots
in the initial three years, but woody coverage was significantly lower in the grazed plots than
in the controls in the last four years (Fig. 1).
The cover of grass encroachment indicators declined slightly, but not significantly, in the
pastures (Table 2), whereas on the less intensively encroached control plots we observed
a significant increase of grass encroachment from 39.4 to 54.8% during the seven observation
220
1234567 df Chi- s. p1234567 df Ch i-s. p
Woody species (%) 39.7 28.3 21.6 15.5 16.8 17.0 16.4 621.546 0.001 13.2 5.6 5.3 4.5 4.0 3.3 2.8 620.108 0.003
±SD 11.5 18.1 15.9 13.0 15.7 15.4 14.2 7.5 4.3 3.3 2.7 3.4 2.0 1.9
Grass encr. ind. species (%) 33.8 25.7 26.0 24.9 28.4 29.2 28.1 68.101 0.231 25.8 26.8 25.1 25.3 23.7 23.3 22.7 62.540 0.864
±SD 20.2 10.0 7.5 10.1 8.8 8.5 10.0 18.0 16.3 15.8 16.6 17.1 15.8 15.8
Litt er layer (%) 40.0 30.8 26.7 26.8 26.5 26.7 26.3 613.928 0.03 25.0 17.7 14.6 14.2 14.3 14.3 12.8 613.091 0.042
±SD 19.5 24.2 24.7 24.5 21.3 20.8 18.6 7.7 10.5 13.8 10.1 9.4 8.9 8.9
Bare soil (%) 10.0 22.0 27.5 23.7 23.2 20.8 19.8 69.778 0.134 14.8 25.7 31.2 29.5 30.8 32.5 29.5 615.485 0.017
±SD 8.2 15.5 18.7 18.4 15.3 16.0 16.7 11.7 18.5 21.3 22.7 19.3 14.7 15.6
Target species (%) 14.3 14.3 15.0 16.2 17.0 18.8 19.3 66.429 0.377 29.0 21.3 20.1 17.9 19.8 21.8 20.8 69.714 0.137
±SD 7.5 9.7 12.4 13.0 11.2 13.7 13.3 13.9 12.4 13.0 12.1 10.8 15.2 14.7
Target species (number) 11.7 12.5 14.8 14.0 15.5 16.0 15.7 613.141 0.041 13.5 14.3 15.5 14.5 14.3 15.5 16.0 69.541 0.145
±SD 5.9 6.1 7.3 7.1 7.5 7.9 7.3 4.6 4.2 4.8 5.6 5.2 5.6 4.7
Ruderal species (%) 7.3 5.9 5.1 3.3 3.9 3.8 3.3 62.827 0.83 2.5 1.6 1.8 1.8 1.4 1.8 2.9 65.876 0.437
±SD 14.9 13.0 11.4 6.4 6.3 8.0 6.3 4.2 3.2 3.3 1.8 1.5 2.0 3.0
Ruderal species (number) 3.5 3.0 4.0 3.3 4.0 3.8 4.2 66.317 0.389 1.7 1.8 2.8 2.7 2.7 3.3 4.8 612.448 0.053
±SD 3.8 3.3 4.9 3.4 3.4 3.9 3.8 2.9 2.7 3.2 2.2 2.2 2.1 4.4
(Initial woody coverage ≥ 25%)
(Initial woody coverage < 25%)
Years
Friedman Test
Years
Friedman Test
Intensive encroachment
Less intensive encro achment
Table 2. Development of grazed plots (n = 6). M ean values and standard deviation (± SD). p ≤ 0.001 = extremely significant, 0.01 ≥ p > 0.001 = very significant,
0.05 ≥ p > 0.01 = significant, p > 0.05 = not significant.
Tabelle 2. Entwicklung der Weideflächen (n = 6). Mitt elwerte und Standardabweichung (± SD). p ≤ 0,001 = extremely significant; 0,01 ≥ p > 0,001 = very significant;
0,05 ≥ p > 0,01 = significant; p > 0,05 = not significant.
221
1 2 3 4 5 6 7 df Chi- s. p1234567df Ch i-s. p
Woody species (%) 42.6 44.8 42.1 45.0 49.3 54.8 58.9 617.857 0.007 11.0 14.6 16.4 19.5 20.3 21.1 28.2 618.500 0.005
±SD 17.4 14.8 20.0 17.9 16.8 19.6 17.2 7.7 12.2 12.1 11.3 12.7 12.1 20.2
Grass encr. ind. species (%) 33.6 35.6 35.6 40.3 42.6 41.6 43.4 69.878 0.13 39.4 46.3 48.9 49.0 50.0 53.3 54.8 621.860 0.001
±SD 13.9 14.4 16.1 16.4 17.0 13.0 14.5 13.9 14.2 11.3 12.8 13.9 17.3 17.3
Litt er layer (%) 49.2 50.8 53.3 51.7 54.2 53.7 54.2 63.058 0.802 45.0 52.2 56.7 63.3 63.3 65.0 62.8 622.393 0.001
±SD 23.8 24.2 20.7 18.6 20.8 22.0 21.5 25.9 30.0 29.6 26.4 24.0 25.7 25.8
Bare soil (%) 7.5 7.3 7.4 9.1 7.6 8.9 7.8 64.936 0.552 15.5 16.0 14.3 10.1 12.3 12.7 13.2 67.768 0.256
±SD 13.5 16.0 13.5 11.5 11.2 11.7 10.1 23.7 24.4 20.9 15.7 14.4 16.6 14.2
Target species (%) 17.0 13.1 10.1 11.1 10.8 9.6 7.7 623.588 0.001 17.6 15.0 14.4 18.5 17.7 15.7 13.4 67.504 0.277
±SD 15.3 12.0 11.6 10.4 11.5 10.5 7.8 12.0 11.4 10.1 16.0 15.7 12.5 12.4
Target species (number) 8.5 8.7 8.7 8.0 8.0 8.0 8.5 62.682 0.848 12.5 11.8 11.7 11.8 11.8 11.8 11.2 64.008 0.676
±SD 5.3 5.1 4.5 4.3 4.4 4.7 5.2 5.6 5.6 5.9 6.0 5.9 5.5 5.6
Ruderal species (%) 1.0 0.6 0.8 0.5 0.6 0.7 0.8 64.788 0.571 0.7 0.4 0.3 0.2 0.4 0.3 0.4 63.083 0.798
±SD 1.3 0.9 1.3 0.9 1.1 1.3 1.5 0.9 0.5 0.5 0.3 0.5 0.4 0.4
Ruderal species (number) 2.0 1.5 2.5 2.2 2.2 2.5 1.8 61.897 0.929 0.8 0.7 1.2 1.2 1.0 1.3 1.5 64.667 0.587
±SD 2.6 1.9 3.4 3.5 2.9 4.2 2.6 0.8 0.8 1.0 1.2 1.1 0.8 1.4
Years
Friedman Test
Years
Friedman Test
Tabell e 3. Entwicklung der unbeweideten Kontrollflächen (n = 6). Mittelwerte und Standardabweichung (± SD). p ≤ 0,001 = extremely significant; 0,01 ≥ p > 0,001 = very significant;
0,05 ≥ p > 0,01 = significant; p > 0,05 = not significant
Table 3. Development of ungrazed plots (n = 6). Mean values and standard deviation (± SD). p ≤ 0.001 = extremely significant, 0.01 ≥ p > 0.001 = very significant,
0.05 ≥ p > 0.01 = significant, p > 0.05 = not significant
Intensive encroachment
Less intensive encro achment
(Initial woody coverage ≥ 25%)
(Initial woody coverage < 25%)
222
years (Table 3). This was confirmed by paired U-tests; the cover of grass species was signif-
icantly higher in the control plots than in the pastures in the third to the seventh years (Fig. 1).
On the strongly encroached control plots, the grass encroachment also increased from 33.6 to
43.4%, but not significantly (Table 3, Fig. 1).
Within seven years, we found a significant reduction of litter layer cover in the paddocks
in both structure types (Table 2), whereas the litter layer significantly increased only within
the less shrub encroached control plots (Table 3). Accordingly, the paired U-tests showed
significant differences between the grazed and ungrazed plots in the less encroached struc-
ture type from year three to year seven (Fig. 1).
The share of bare soil significantly increased in the less encroached paddock plots from
14.8 to 29.5%, whereas we observed only a tendency for increasing shares of bare soil with-
in the stronger encroached grazed plots (Table 2, Fig. 1). In contrast, the mean values and
the results of the Friedman tests indicated no change in bare soil over time within the un-
grazed control plots (Table 3, Fig. 1).
4.2 Development of target species richness
During the seven observation years, a total of 73 target species (including 20 red list spe-
cies) were observed in all plots (25-m2). Although the average target species cover decreased
in the less encroached paddock plots, from 29.0% in the first observation year to 20.8% in
the seventh year, the average species number slightly, but not significantly, increased, from
13.5 to 16.0 in this structure type (Table 2). There was a marginal trend of reduction of tar-
get species number and cover within the ungrazed control plots (Table 3). In contrast, we
observed an increasing target species cover, from 14.3% to 19.3%, as well as a significant
rise of species number, from 11.7 to 15.7 (Friedman test: p = 0.041), in the formerly inten-
sively encroached paddock plots. The controls showed a significant reduction of target spe-
cies coverage, from 17.0% to 7.7% during the seven years of observation, but no change in
species number. U-tests showed no significant differences. A wide range of target species
benefited from grazing, especially short-lived therophytes such as Arenaria serpyllifolia,
Cerastium spp., Draba verna agg., but also perennial red list species such as Bothriochloa
ischaemum and Oxytropis pilosa were able to expand in grazed plots.
4.3 Development of ruderal species
We counted a total of 52 ruderal species within all 25-m2-plots over the seven observa-
tion years. These included nitrophilous ruderal species, such as Galium aparine or Urtica
dioica, but also ruderal species which are typical of dry grasslands, such as Convolvulus
arvensis or Erodium cicutarium. The cover of ruderal species did not change during the
observation years in all structure and management types. The number of ruderal species
slightly increased from 1.7 to 4.8 species within the less intensively encroached paddock
plots, but this development was not significant (Friedman test: p = 0.053). Several species
showed higher frequencies, for example Bromus sterilis, Urtica dioica and Viola arvensis.
However, U-tests showed also no significant differences in all cases.
223
Fig. 1. Development of habitat structure variables within encroached dry grasslands in grazed (dark
grey bars) and ungrazed (light grey bars) plots (n = 6). Values are means ± Standard error. Mann-
Whitney U-test: p ≤ 0.001 = extremely significant***, 0.01 ≥ p > 0.001 = very significant**, 0.05 ≥ p >
0.01 = significant*, p > 0.05 = not significant (n.s.).
Abb. 1. Entwicklung der Habitatstrukturen (Gehölzdeckung, Deckung brachetoleranter Gräser, Streu-
deckung, Offenbodenanteile) innerhalb beweideter (dunkelgraue Balken) und nicht beweideter (hell-
graue Balgen) Dauerbeobachtungsflächen (n = 6). Mittelwerte ± Standardfehler. Mann-Whitney-U-
Test: p ≤ 0,001 = höchst signifikant***; 0,01 ≥ p > 0,001 = sehr signifikant**; 0,05 ≥ p > 0,01 = signi-
fikant*; p > 0,05 = nicht signifikant (n.s.).
224
5. Discussion
5.1 Development of habitat structure parameters
In accordance with other observations in European dry grasslands (RAHMANN 2000,
ZEHM 2008, VEITH et al. 2012, CZYLOK et al. 2013), our results clearly demonstrated
a significant reduction of woody coverage by goat grazing, even when thorny and spiny
species dominated the shrubbery (ELIAS & TISCHEW 2016). Goats are opportunistic interme-
diate feeders (HOFMANN 1989); when available, they show a preference for browsing
(AHARON et al. 2007, EL AICH et al. 2007, ANIMUT & GOETSCH 2008). In addition, goats
remove bark from woody plants which can effectively damage trees (FAJEMISIN et al. 1996,
ZEHM 2008, HOLST et al. 2004). Furthermore, they often stand on their hind legs to maxim-
ize available forage height (RAHMANN 2000, EL AICH et al. 2007; Fig. 2a). Generally, there
is a greater reduction of shrub cover under increasing grazing pressure by goats (MELLADO
et al. 2003, JAUREGUI et al. 2008). Furthermore, spring grazing has a greater impact on the
reduction of woody species as shrubs are more palatable to livestock when the leaves are
fresh and the thorns of young shoots are still soft (DOSTÁLEK & FRANTÍK 2012, ELIAS &
TISCHEW 2016).
We observed only a minor decline of grass encroachment indicators, but a significant re-
duction of the litter layer within the goat paddocks, combined with an overall positive effect
on the share of bare soil (Fig. 2b, c). In contrast, competitive grass species and litter layer
increased significantly in ungrazed plots with less intensive initial shrub encroachment,
showing the positive effects of goat grazing in controlling competitive grasses. Generally, as
with the woody species, intensive spring grazing is also more effective than late summer
grazing in controlling competive grass species as their senescent herbage of low nutritive
value is infrequently grazed by small ruminants (CROFTS & JEFFERSON 1999, DOSTÁLEK &
FRANTÍK 2012, HEJCMANOVÁ et al. 2016). The observed decrease of the litter layer and
increase in bare soil patches within our goat paddocks must be considered as a combined
effect of feeding and trampling (e.g., ROSENTHAL et al. 2012).
Interestingly, in the intensively encroached control plots the spread of grass species and
subsequent accumulation of litter slowed with increasing shrubbery. Similarly, DIERSCHKE
(2006) observed decreased Bromus erectus coverage with increased shading by shrubs with-
in calcareous dry grasslands near Göttingen (Germany). Therefore, the unexpected low
reduction of grass encroachment indicators within our goat paddocks could be partly ex-
plained by the simultaneous improvement of light conditions by reduced shrub cover, which
abrogated the reduction of grass biomass by goat grazing.
5.2 Target species
We observed significantly increasing numbers of target species by goat grazing within
the formerly intensively encroached parts of the paddocks. This was primarily related to the
browsing and grazing behavior of the goats. The decreased shading and competition from
tall plants resulted in the expansion of suitable light conditions for low growing and less
competitive dry grassland species (Fig. 2d). Similary, within encroached calcareous dry
grasslands in Germany, RAHMANN (2000) documented decreased woody coverage and
simultaneously an increase in light-demanding target species number by goat grazing, while
woody and competitive grass species, particularly Brachypodium pinnatum, spread in
225
Fig. 2. Goat activities in the Nelbener Grund paddock. a) Browsing of shrubs by goats, frequently, they
stand up on their hind legs to maximize their browsing horizon up to two meters. b) Grazing of swards
is also a typical behavior of goats. c) In addition, the creation of bare soil patches by trampling is an
important effect of grazing animals. Goat activities result in sparsely covered, but species-rich swards.
Among other species, picture d) shows numerous target species such as Alyssum alyssoides,
A.montanum, Astragalus exscapus, Erysimum crepidifolium, Festuca csikhegyensis, Holosteum
umbellatum, Potentilla neumanniana, Salvia pratensis, Sanguisorba minor, Thymus praecox (Photos:
2a, b, d D. Elias, 2009, 2011 and 2013, respectively, and 2c S. Heinrich, 2009).
Abb. 2. Ziegenaktivitäten innerhalb der Weidefläche Nelbener Grund. a) Gehölzfraß an Gebüschen.
Die Ziegen stellen sich häufig auf die Hinterbeine um ihren Fraßhorizont auf bis zu zwei Metern zu
erhöhen. b) Aber auch der Fraß der Gras-/Krautschicht ist ein sehr typisches Weideverhalten von
Ziegen. c) Ein bedeutender Effekt stellen auch Triffeffekte und die Schaffung von Offenboden-
bereichen dar. Die Ziegenaktivitäten schaffen lichtreiche Konkurrenzverhältnisse in Bodennähe und
lückige, aber sehr artenreiche Trockenrasenbestände, wie auch das Bild unten rechts zeigt. Neben
anderen Arten beinhaltet der Bildausschnitt d) einige Zielarten, wie zum Beispiel Alyssum alyssoides,
A. montanum, Astragalus exscapus, Erysimum crepidifolium, Festuca csikhegyensis, Holosteum
umbellatum, Potentilla neumanniana, Salvia pratensis, Sanguisorba minor, Thymus praecox (Fotos: 2a,
b, d D. Elias, 2009, 2011 bzw. 2013, and 2c S. Heinrich, 2009).
ungrazed controls. Similar observations were also made within shrub-encroached dry grass-
lands grazed by goats in Poland, with decreased shrub coverage accompanied by the spread
of dry grassland species (CZYLOK et al. 2013). SCHWABE (1997) documented increased
numbers of gap indicator species under the influence of goat grazing within shrubby heath
vegetation of the montane zone of the southern Black Forest (Germany).
Generally, in addition to defoliation, trampling by livestock generates positive impacts
through suppression of tall-growing species, increased litter decomposition and the creation
of bare soil patches that provide opportunities for the recruitment of small, light-demanding
plant species (FLEISCHER et al. 2013, SCHWABE et al. 2013, TÖRÖK et al. 2014, KÖHLER et al.
2016, RUPPRECHT et al. 2016). Additionally, feeding in conjunction with trampling can lead
226
to the activation of the soil seed bank (RUPRECHT et al. 2010, SCHWABE et al. 2013). This is
particularly applicable in neglected pastures with dense grass and litter coverage. Trampling
by small ruminants can also enhance the incorporation of seeds into the soil, which may be
positively linked to germination success (EICHBERG et al. 2005, EICHBERG & DONATH 2018).
In addition, ruminant dung can contain germinable seeds of dry grassland species (WESSELS
& SCHWABE 2008, BENTJEN et al. 2016) and a high trampling frequency can enhance number
of dung-borne seedlings (FAUST et al. 2011).
As shown in our study, the creation of bare soil patches by grazing animals favoured tar-
get therophytes, which is in line with other studies (MCINTYRE et al. 1995, DUPRÉ & DIEK-
MANN 2001, ŠKORNIK et al. 2010). The recorded expansion of red list species was previously
documented within our goat paddocks (e.g., Astragalus exscapus, Gagea bohemica; see
ELIAS et al. 2014). As might be expected, we observed minor decreases of target species
coverage within the less intensively encroached goat paddocks as a result of the goat grazing
on the formerly abandoned sites. However, such results should not be interpreted as a nega-
tive effect, particularly as the number of target species did not decline but even increased
slightly.
In contrast, the results within the ungrazed control plots confirmed that shrub as well as
grass and litter encroachment pose a serious threat to dry grasslands of high conservation
value. However, even after seven years, the target species richness has not yet been reduced,
showing the high restoration potential of these encroached grasslands. Nevertheless, with
increased shading and competition as a consequence of succession, typical light-demanding
species and dry grassland plant communities will most certainly decrease in the near future,
a situation demanding appropriate measures such as restorative goat grazing.
5.3 Ruderal species
We did not observe any relevant changes in the coverage of ruderal species in the pad-
docks, but we recorded a slightly (but not significant) increase of species number within the
less encroached paddock plots. Due to the low abundance of ruderal species, this develop-
ment cannot be considered as threatening in regards to the dry grassland communities. Simi-
lary, DOSTÁLEK & FRANTÍK (2008) also recorded an increased presence of nitrophilous and
ruderal species after the introduction of co-grazing by sheep and goats, but, as in our study,
the observed level did not exceed a tolerable level. In fact, to a certain extent, ruderal species
are a typical component of grazed and sparsely covered dry grasslands (BRANDES &
PFÜTZENREUTHER 2013).
Ruderalisation by faeces deposits depends on embedded endozoochorous seeds, nutrient
input and their spatial distribution. It is likely that nitrophilous species (e.g., Urtica dioica:
BENTJEN et al. 2016) are frequent in goat faeces, but such competitive species are only spo-
radically occuring in the dry grassland patches of our paddocks, which indicates the im-
portance of water limitation as a filter for species not adapted to stressful environments such
as dry grasslands (EICHBERG et al. 2007), especially on shallow soils (KRUMBIEGEL et al.
1998). Higher abundances of nitrophilous species were only locally observed near watering
places, shelters and resting places in the studied paddocks, which were established outside of
vulnerable dry grassland mosaics.
227
6. Conclusion
Our study clearly demonstrated the positive impacts of paddock grazing by goats with
a relatively high grazing pressure on the habitat quality of encroached dry grasslands. Sub-
stantial processes initiated by goat grazing are browsing of woody species and grazing (es-
pecially grasses) as well as trampling (especially removal of litter) which results in a higher
habitat and therefore species diversity of formerly abandoned dry grasslands. The reduction
of shrub, grass and litter encroachment as well as the increasing number of target species
correlates with the improvement of the conservation status of dry grassland habitat types.
We are aware that the number of analyzed plots per pasture is relatively small and it
might be interesting to intensify the sampling effort to gain deeper insights into processes
related to the goat grazing activities. However, the vegetation changes within the permanent
plots were representative for the respective pasture.
7. Management implications
Generally, the selection of the livestock type and breed is crucial in biodiversity conser-
vation and management (CROFTS & JEFFERSON 1999, TÓTH et al. 2018). Overall, goats and
especially Boer goats, which were used for grazing in the investigation areas in the lower
Saale River valley, are well adapted to remove or control shrub encroachment in sloping dry
grasslands. Boer goats spend much time in exploratory behaviour that ensures browsing
activities in all paddock parts. Within our investigation areas, goats did not avoid thorny or
spiny shrub species (ELIAS & TISCHEW 2016), which are habitually neglected by other graz-
ing animals such as sheep. Compared to exclusively mechanical shrub removal, goat grazing
is a cost-effective and sustainable management tool (RAHMANN 2000, HART 2001, ELIAS et
al. 2014).
Based on our experiences with restorative goat grazing in the lower Saale River valley,
we suggest the following practical steps:
(i) If the shrubbery is dominated by plants higher than two meters (reachable browsing
zone), manual shrub removal before the grazing regime starts will accelerate the restora-
tion success.
(ii) For reducing shrub and grass encroachment, goat grazing should start in spring (between
end of March and mid-April depending on fodder availability).
(iii) High grazing pressure is decisive for restoring neglected dry grassland sites. We rec-
ommend stocking rates between 0.5 and 1.0 LU/ha/year during a restoration phase of the
first five to seven years, depending on the restoration goals and site conditions (e.g., lev-
el of encroachment, productivity). Due to rapid shrub re-growth, this is particularly ap-
plicable after mechanical clearance. If the grazing regime was successful in reducing the
shrubbery, the stocking rate can be gradually lowered.
(iv) Supplementary feeding (except of minerals) should be avoided, because it might de-
crease the browsing intensity due to the presence of more easily available high-quality
food.
(v) In order to hinder the goats from breaking out of the pastures, permanent electric fences
with four to five strands proved to be useful.
(vi) If possible, shelters and waterering places should be established only outside of vulnera-
ble dry grassland communities.
228
Erweiterte deutsche Zusammenfassung
Einleitung – Trocken- und Halbtrockenrasen stehen aufgrund ihrer Artenvielfalt (VEEN et al. 2009,
WILSON et al. 2012) und der hohen Anzahl an gefährdeten Arten (KORNECK et al. 1998, VAN SWAAY
et al. 2006) im besonderen Interesse des europäischen Naturschutzes (EUROPEAN COMMISSION 2013,
JANSSEN et al. 2016). Die Aufgabe der vielerorts traditionellen Weidenutzung mit Schafen und Ziegen
und die nachfolgende Vergrasung und Verbuschung der Bestände stellt aktuell eine Hauptgefährdungs-
ursache dar (WALLISDEVRIES et al. 2002, DIERSCHKE 2006, CALACIURA & SPINELLI 2008, EUROPEAN
COMMISSION 2015). Ziegen werden weltweit zunehmend für die Entbuschung eingesetzt (STRANG
1973, LUGINBUHL 1999, HOLST et al. 2004, SMART et al. 2006, CELAYA et al. 2010, ASCOLI et al.
2013), jedoch liegen aus dem mitteleuropäischen Raum nur wenige detaillierte Studien im Bereich von
Trockenrasen vor, weshalb vielerorts nach wie vor Schafbeweidung bevorzugt wird.
Für die Entbuschung mittels Ziegen ist anfänglich eine intensivere Beweidung und möglichst eine
Beweidung im Frühjahr erforderlich (ELIAS & TISCHEW 2016). Auf abgelegenen und verbuschten
Standorten ist ein solches Beweidungsregime nur mit Rotationsstandweiden mit Festzäunen ökono-
misch sinnvoll umsetzbar. Nach wie vor gibt es aber Bedenken Standweiden auf zwar bereits stark
verbuschten, aber noch artenreichen Trockenrasen einzusetzen, da negative Effekte auf die Zielarten
durch Trittschäden und Ruderalisierung befürchtet werden. In dieser Studie untersuchen wir den
Einfluss der Ziegenbeweidung auf die Deckung von Gehölzen und brachetoleranten Grasarten sowie
auf die Ausbildung der Streuschicht und die Anteile von Offenboden. Außerdem bewerten wir das
Auftreten von charakteristischen und gefährdeten Pflanzenarten der Trockenrasen (= Zielarten) sowie
von ruderalen Pflanzenarten.
Untersuchungsgebiet – Die Untersuchungen wurden auf sechs Ziegenweiden im Unteren Saaletal
zwischen Halle (Saale) und Könnern in Sachsen-Anhalt (Deutschland) durchgeführt. Das Untere Saale-
tal besitzt subkontinentales Klima. In Verbindung mit der über Jahrtausende andauernden Nutzung der
Trockenhänge, die überwiegend als Schaf- und Ziegenweide dienten, haben sich hier sehr artenreiche
und seltene Trockenrasengesellschaften entwickeln können. Spätestens nach der politischen Wende in
den 1990er Jahren wurde die Beweidung auf vielen Trockenrasenstandorten jedoch aufgegeben (JÄGER
& MAHN 2001, RICHTER et al. 2003). Die Folge waren Vergrasung sowie insbesondere Verbuschung
und der schrittweise Abbau des typischen Arteninventars. Dennoch sind nach wie vor artenreiche
Trockenrasenrestflächen vorhanden, jedoch sind diese akut durch fortschreitende Verbuschung bedroht.
Aus diesem Grund wurde im Jahr 2006 ein Ziegenbeweidungsprojekt im Unteren Saaletal initiiert. Die
Landwirte wurden aufgefordert, die Flächen mit einer höheren Besatzstärke zu beweiden und auf Zufüt-
terung zu verzichten, was weitgehend akzeptiert wurde. Detaillierte Beschreibungen zu Flora und Vegeta-
tion im Unteren Saaletal können PARTZSCH (2000) und ELIAS et al. (2015) entnommen werden.
Material und Methoden – Die Vegetationserfassung erfolgte über einen Zeitraum von sieben Jah-
ren auf jeweils zwei 25-m2-großen Dauerbeobachtungsflächen pro Weidefläche. Diese Dauerbeobach-
tungsflächen wurden in ehemals stark verbuschten (Ausgangsgehölzdeckung ≥ 25 %) und weniger stark
verbuschten Bereichen (Ausgangsgehölzdeckung < 25%) eingerichtet. Darüber hinaus wurde jeweils
eine Kontrollfläche pro Strukturtyp außerhalb jeder Weidefläche etabliert. Es wurden einmalig pro Jahr
alle Arten und deren Häufigkeiten sowie ausgewählte Strukturparameter (Offenboden, Streu) erfasst.
Die statistische Auswertung erfolgte mittels nichtparameterischer Verfahren (Friedman-Test, U-Test).
Ergebnisse – Nach sieben Jahren Ziegenbeweidung kann festgestellt werden, dass die Gehölz- und
Streudeckungen signifikant abgenommen haben, zu geringeren Anteilen auch die der brachetoleranten
Grasarten. Außerdem wurden signifikante Zunahmen bei den Offenbodenanteilen und der Artenzahl
der wertgebenden Trockenrasenarten erfasst. Neben kurzlebigen Therophyten konnten sich auch einige
Rote-Liste-Arten in den Dauerbeobachtungsflächen etablieren. Die beobachtete (nicht signifikante)
leichte Zunahme der Artenzahl der Ruderalarten ging einher mit unverändert sehr geringen Deckungs-
werten.
229
Demgegenüber wurden auf den unbeweideten Kontrollflächen signifikante Zunahmen bei den Ge-
hölzen, den brachetoleranten Grasarten und den Streudeckungen festgestellt. Die Deckung der Zielarten
war zumindest auf den stark verbuschten Kontrollflächen signifikant rückläufig, während bei den Ar-
tenzahlen in beiden Strukturtypen keine nennenswerten Veränderungen zu beobachten waren. Ru-
deralarten kamen auf den unbeweideten Kontrollflächen nur mit sehr geringen Abundanzen vor.
Diskussion – Die bei unseren Untersuchungen erfassten Gehölzrückgänge auf den Ziegenweiden
stimmen überein mit anderen Studien auf Trockenstandorten (RAHMANN 2000, ZEHM 2008, VEITH et
al. 2012, CZYLOK et al. 2013). Ziegen gelten als opportunistische Mischfresser (HOFMANN 1989) mit
Bevorzugung von Gehölznahrung, sofern verfügbar (AHARON et al. 2007, EL AICH et al. 2007, ANI-
MUT & GOETSCH 2008). Im Allgemeinen gilt, dass durch höhere Besatzstärken die Gehölze effektiver
zurückgedrängt werden können (MELLADO et al. 2003, JAUREGUI et al. 2008). Außerdem wird Früh-
jahrsbeweidung gegenüber Spätsommerbeweidung als effizientere Methode eingeschätzt, weil die
Gehölze dann frisch austreiben und die Dornen von jungen Trieben noch nicht verholzt sind
(DOSTÁLEK & FRANTÍK 2012, ELIAS & TISCHEW 2016).
Wir haben nur geringe Rückgänge bei den brachetoleranten Grasarten, aber signifikant
zurückgehende Streudeckungen erfasst. Insbesondere bei Vergleich mit den unbeweideten Kontroll-
flächen wird aber deutlich, dass Ziegenbeweidung die lokale Ausbreitung von konkurrenzstarken
Gräsern kontrollieren und bereits entwickelte Grasfilzdecken auflichten kann. Ebenso, wie bei den
Gehölzen bereits erwähnt, ist eine frühe und intensive Beweidung effizienter zum Zurückdrängen von
dichten Beständen aus konkurrenzstarken Gräsern, da im Spätsommer aufgrund der dann reduzierten
Futterwerte diese weniger stark verbissen werden (CROFTS & JEFFERSON 1999, DOSTÁLEK & FRANTÍK
2012, HEJCMANOVÁ et al. 2016).
Die erfasste Zunahme bei der Artenzahl der wertgebenden Trockenrasenarten steht in direktem Zu-
sammenhang mit der rückläufigen Beschattung und Konkurrenz durch Gehölze und der zunehmenden
lichteren Bestandsstruktur mit Offenbodenpatches. Ähnliche Beobachtungen gelangen auch RAHMANN
(2000) im Bereich von verbuschten Kalktrockenrasen. Neben der Fraßtätigkeit der Ziegen sind positive
Effekte durch Hufeinwirkung zu nennen, wodurch der Abbau des Streufilzes und die Entstehung von
Offenbodenbereichen zusätzlich unterstützt werden. Durch Beweidung entstandene Offenbodenpatches
können als Keim- und Etablierungsstellen für konkurrenzschwache und lichtliebende Trockenrasenar-
ten dienen (FLEISCHER et al. 2013, SCHWABE et al. 2013, KÖHLER et al. 2016, RUPPRECHT et al. 2016).
Außerdem kann durch Streuabbau als Folge des Huftritts die Diasporenbank aktiviert werden
(RUPRECHT et al. 2010, SCHWABE et al. 2013). Dies gilt insbesondere für länger aufgelassene Trocken-
rasenflächen mit dichten Streufilzdecken.
Die im Zuge des siebenjährigen Beobachtungszeitraumes nach Einführung der Ziegenbeweidung er-
fasste geringfügige Erhöhung der Artenzahl von Ruderalarten ist nicht als bedenklich im Sinne einer
Verschlechterung des Erhaltungszustandes einzustufen, da die Ruderalarten nur mit sehr geringen
Abundanzen aufgetreten sind (vgl. DOSTÁLEK & FRANTÍK 2008). In diesem Zusammenhang ist darauf
hinzuweisen, dass Ruderalarten bis zu einem gewissen Grad durchaus typische Komponenten beweide-
ter und lückig bewachsener Trockenrasen sind (BRANDES & PFÜTZENREUTHER 2013). Ruderalisie-
rungstendenzen durch Kotabsatz sind abhängig von den im Dung befindlichen Diasporen und der
räumlichen Verbreitung von Dungkonzentrationen auf der Weidefläche. Vermutlich sind Diasporen von
nitrophilen Arten im Ziegenkot durchaus häufig (z. B. Urtica dioica: BENTJEN et al. 2016). Solche
konkurrenzkräftigen Arten treten aber in den beweideten Trockenrasen im Unteren Saaletal nur spora-
disch auf, was mit den extremen Standorteigenschaften durch flachgründige und trockene Böden zu-
sammenhängen dürfte (vgl. KRUMBIEGEL et al. 1998, EICHBERG et al. 2007). Höhere Abundanzen von
nitrophilen Arten wurden auf den Weideflächen im Unteren Saaletal nur kleinflächig im Umfeld von
Unterstand, Tränke und teilweise auf Lagerplätzen festgestellt.
Schlussfolgerungen – Die vorliegende Studie zeigt deutlich den positiven Einfluss von fest instal-
lierten Ziegenrotationsweiden mit höheren Besatzstärken auf die Habitatqualität von verbuschten Tro-
ckenrasen. Wesentliche Pflegeeffekte ergeben sich durch Gehölzfraß und das Auflichten von Grä-
serdominanzen und Streudecken.
230
Managementempfehlungen – Ziegen und insbesondere Burenziegen sind hervorragend für die Zu-
rückdrängung von Gebüschaufwuchs im Bereich von Trockenrasen geeignet. Basierend auf den Erfah-
rungen mit Ziegenbeweidung können folgende Managementempfehlungen gegeben werden: 1) Sofern
die Gebüsche durch sehr hochwüchsige Gehölze dominiert werden, verbessert eine mechanische Entbu-
schung vor Beginn der Ziegenbeweidung den Renaturierungserfolg. Dies gilt vor allem dann, wenn die
Gehölze aus dem Fraßbereich der Ziegen bereits herausgewachsen sind. 2) Für eine effektive Zurück-
drängung von Gehölzen und konkurrenzstarken Gräsern in Verbindung mit deren Streufilzdecken ist
eine Frühjahrsbeweidung ratsam. 3) Wir empfehlen eine Besatzstärke von 0,5 bis 1,0 GVE/ha/Jahr
(abhängig vom Verbuschungsgrad und der Produktivität der Fläche) während einer Renaturierungspha-
se von fünf bis sieben Jahren. Insbesondere nach einer mechanischen Entbuschung sind hohe Besatz-
stärken aufgrund des raschen Wiederaustriebsvermögens vieler typischer Gehölzarten erforderlich.
Nach Rückführung der Verbuschung kann die Besatzstärke schrittweise reduziert werden. 4) Zufütte-
rung sollte weitgehend vermieden werden (Ausnahme Mineralien). 5) Empfehlenswert ist weiterhin die
Verwendung von Elektrozäunen mit vier bis fünf Litzen. 6) Unterstand und Tränke sollten, sofern
möglich, außerhalb wertgebender Bereiche der Weideflächen eingerichtet werden.
Acknowledgment
The project was sponsored by EAFRD (European Agricultural Fund for Rural Development) and the
state of Saxony-Anhalt (407.1.2.-60128/323012000005, 323012000040, 630116000009). We also
thank Sandra Mann and Matthias Necker from the Saaletal Association for implementing the grazing
management and supporting the investigations. We are grateful to Keith Edwards for language
corrections.
References
AHARON, H., HENKIN, Z., UNGAR, E.D., KABABYA, D., BARAM, H. & PEREVOLOTSKY, A. (2007):
Foraging behaviour of the newly introduced Boer goat breed in a Mediterranean woodland: A re-
search observation. – Small Rumin. Res. 69: 144–153.
ANIMUT, G. & GOETSCH, A.L. (2008): Co-grazing of sheep and goats: Benefits and constraints. – Small
Rumin. Res. 77: 127–145.
ASCOLI, D., LONATI, M., MARZANO, R., BOVIO, G., CAVALLERO, A. & LOMBARDI, G. (2013): Pre-
scribed burning and browsing to control tree encroachment in southern European heathlands. – For.
Ecol. Manag. 289: 69–77.
BACON, J. (Ed.) (2003): The Scrub Management Handbook: Guidance on the Management of Scrub on
Nature Conservation Sites. – English Nature, Peterborough: 45 pp.
BENECKE, N. (1994): Der Mensch und seine Haustiere: Die Geschichte einer jahrtausendealten Bezie-
hung. – Theiss, Stuttgart: 470 pp.
BENTJEN, O., BOBER, J., CASTENS, J. & STOLTER, C. (2016): Seed dispersal capacity of sheep and goats
in a near-coastal dry grassland habitat. – Basic Appl. Ecol. 17: 508–515.
BRANDES, D. & PFÜTZENREUTHER, S. (2013): Die Wechselbeziehungen zwischen Steppenrasen und
Adventiv- und Ruderalpflanzen in Deutschland. – In: BAUMBACH, H. & PFÜTZENREUTHER, S.:
Steppenlebensräume Europas - Gefährdung, Erhaltungsmaßnahmen und Schutz: 55–67. Thüringer
Ministerium für Landwirtschaft, Forsten, Umwelt und Naturschutz, Erfurt.
CALACIURA, B. & SPINELLI, O. (2008): Management of Natura 2000 habitats. 6210 Semi-natural dry
grasslands and scrubland facies on calcareous substrates (Festuco-Brometalia) (*important orchid
sites). – European Commission: 38 pp.
CELAYA, R., JÁUREGUI, B.M., GARCÍA, R.R., BENAVIDES, R., CARCÍA, U. & OSORO, K. (2010): Chang-
es in heathland vegetation under goat grazing: effects of breed and stocking rate. – Appl. Veg.
Sci.13: 125–134.
CROFTS, A. & JEFFERSON, R.G. (Eds.) (1999): The Lowland Grassland Management Handbook, 2nd ed.
– English Nature/The Wildlife Trusts, Peterborough: 529 pp.
231
CZYLOK, A., SLUSARCZYK, M. & TYC, A. (2013): Die Erfahrung aus der Revitalisierung von Trocken-
rasen mit Ziegenbeweidung im Krakau-Tschenstochaer Jura. – Natl.-Jahrb. Unteres Odertal 2013:
27–34.
DEÁK, B., VALKÓ, O., TÖRÖK, P. & TÓTHMÉRÉSZ, B. (2016): Factors threatening grassland specialist
plants - A multi-proxy study on the vegetation of isolated grasslands. –Biol. Conserv. 204: 255–262.
DENGLER, J., JANIŠOVÁ, M., TÖRÖK, P. & WELLSTEIN, C. (2014): Biodiversity of Palaearctic grass-
lands: a synthesis. – Agric. Ecosys. Environ. 182: 1–14.
DIERSCHKE, H. (2006): Sekundär-progressive Sukzession eines aufgelassenen Kalkmagerrasens –
Dauerflächenuntersuchungen 1987–2002. – Hercynia N.F. 39: 223–245.
DOSTÁLEK, J. & FRANTÍK, T. (2008): Dry grassland plant diversity conservation using low-intensity
sheep and goat grazing management: case study in Prague (Czech Republic). – Biodiv. Conserv. 17:
1439–1454.
DOSTÁLEK, J. & FRANTÍK, T. (2012): The Impact of Different Grazing Periods in Dry Grasslands on the
Expansive Grass Arrhenatherum elatius L. and on Woody Species. – Environ. Manag. 49: 855–861.
DUPRÉ, C. & DIEKMANN, M. (2001): Differences in species richness and life-history traits between
grazed and abandoned grasslands in southern Sweden. – Ecography 24: 275–286.
EICHBERG, C. & DONATH, T.W. (2018): Sheep trampling on surface-lying seeds improves seedling
recruitment in open sand ecosystems. – Restor. Ecol. 26: 211–219.
EICHBERG, C., STORM, C. & SCHWABE, A. (2005): Epizoochorous and post-dispersal processes in
a rare plant species: Jurinea cyanoides (L.) Rchb. (Asteraceae). – Flora 200: 477–489.
EICHBERG, C., STORM, C. & SCHWABE, A. (2007): Endozoochorous dispersal, seedling emergence and
fruiting success in disturbed and undisturbed successional stages of sheep-grazed inland sand eco-
systems. – Flora 202: 3–26.
EL AICH, A., EL ASSOULI, N., FATHI, A., MORAND-FEHR, P. & BOURBOUZE, A. (2007): Ingestive
behavior of goats grazing in the Southwestern Argan (Argania spinosa) forest of Morocco. – Small
Rumin. Res. 70: 248–256.
ELIAS, D. & TISCHEW, S. (2016): Goat pasturing – A biological solution to counteract shrub encroach-
ment on abandoned dry grasslands in Central Europe? – Agric., Ecosyst. Environ. 234: 98–106.
ELIAS, D., FRANK, D., MANN, S. & SCHÜTZE, P. (2015): Unteres Saaletal: Porphyrlandschaft bei Gim-
ritz, Perlgrashänge bei Rothenburg, Nelbener Grund und Georgsburg bei Könnern (Ziegenweide). –
Tuexenia Beiheft 8: 75–93.
ELIAS, D., MANN, S. & TISCHEW, S. (2014). Ziegenstandweiden auf degradierten Xerothermrasenstan-
dorten - Auswirkungen auf Flora und Vegetation. – Nat. Landsch. 89: 200–208.
ELLENBERG, H. & LEUSCHNER, C. (2010): Vegetation Mitteleuropas mit den Alpen – in ökologischer,
dynamischer und historischer Sicht. – Ulmer, Stuttgart: 1357 pp.
EUROPEAN COMMISSION (2013): Interpretation Manual of European Union Habitats. – EUR 28.
EUROPEAN COMMISSION (2015): The state of Nature in the EU: Reporting under the EU Habitats and
Birds Directives 2007–2012. – European Union, Luxembourg.
FAJEMISIN, B., GANSKOPP, D., CRUZ, R. & VAVRA, M. (1996): Potential for woody plant control by
Spanish goats in the sagebrush steppe. – Small Rumin. Res. 20: 229–238.
FAUST, C., EICHBERG, C., STORM, C. & SCHWABE, A. (2011): Post-dispersal impact on seed fate by
livestock trampling - A gap of knowledge. – Basic Appl. Ecol. 12: 215–226.
FLEISCHER, K., STREITBERGER, M. & FARTMANN, T. (2013): The importance of disturbance for the
conservation of a low-competitive herb in mesotrophic grasslands. – Biologia 68: 398–403.
FRANK, D. (unter Mitarbeit von Heino John & Anselm Krumbiegel) (2016): Gefäßpflanzen (Tracheo-
phyta: Lycopodiophytina, Pteridophytina, Spermatophytina) Bestandsentwicklung. – In: FRANK, D.
& SCHNITTER, P. (Eds.): Pflanzen und Tiere in Sachsen-Anhalt: Ein Kompendium der Biodiversität:
192–318. Natur+Text, Rangsdorf.
FRANK, D., HERDAM, H., JAGE, H. et al. (2004): Rote Liste der Farn- und Blütenpflanzen (Pteridophyta
et Spermatophyta) des Landes Sachsen-Anhalt. – Ber. Landesamtes für Umweltschutz Sachsen-
Anhalt 34: 91–110.
GLAVAC, V. (1983): Über die Wiedereinführung der extensiven Ziegenhaltung zwecks Erhaltung und
Pflege der Kalkmagerrasen: Überlegungen zu einem Modellversuch in Nordhessen. – Naturschutz
Nordhess. 6: 25–45.
HART, S.P. (2001): Recent Perspectives in Using Goats for Vegetation Management in the USA. –
J. Dairy Sci. 84 (E. Suppl.): E170–E176.
232
HEGEDUŠOVÁ, K. & SENKO, D. (2011): Successional changes of dry grasslands in southwestern Slo-
vakia after 46 years of abandonment. – Plant Biosys. 145: 666–687.
HEJCMANOVÁ, P., POKORNÁ, P., HEJCMAN, M. & PAVLŮ, V. (2016): Phosphorus limitation relates to
diet selection of sheep and goats on dry calcareous grassland. – Appl. Veg. Sci. 19: 101–110.
HOFMANN, R.R. (1989): Evolutionary steps of ecophysiological adaptation and diversification of rumi-
nants: a comparative view of their digestive system. – Oecologia 78: 443–457.
HOLST, P.J., ALLAN, C.J., CAMPBELL, M.H. & GILMOUR, A.R. (2004): Grazing of pasture weeds by
goats and sheep. 2. Scotch broom (Cytisus scoparius L.). – Aus. Exp. Agric. 44: 553–557.
JÄGER, C. & MAHN, E.-G. (2001): Die Halbtrockenrasen im Raum Questenberg (Südharz) in Beziehung
zu ihrer Nutzungsgeschichte. – Hercynia N.F. 34: 213–235.
JÄGER, E.J. (Ed.) (2011): Rothmaler - Exkursionsflora von Deutschland. Gefäßpflanzen: Grundband. –
Spektrum, Heidelberg: 944 pp.
JANSSEN, J.A.M., RODWELL, J.S., GARCÍA CRIADO, M. et al. (2016): European Red List of Habitats -
Part 2. Terrestrial and freshwater habitats. – Publications Office of the European Union, Luxem-
bourg: 38 pp.
JAUREGUI, B.M., ROSA-GARCIA, R., GARCIA, U., WALLISDEVRIES, M.F., OSORO, K. & CELAYA, R.
(2008): Effects of stocking density and breed of goats on vegetation and grasshopper occurrence in
heathlands. – Agriculture, Ecosystems and Environment 123: 219–224.
KÖHLER, M., HILLER, G. & TISCHEW, S. (2016): Year-round horse grazing supports typical vascular
plant species, orchids and rare bird communities in a dry calcareous grassland. – Agric., Ecosys.
Environ. 234: 48–57.
KORNECK, D., SCHNITTLER, M., KLINGENSTEIN, F., LUDWIG, G., TAKLA, M., BOHN, U. & MAY, R.
(1998): Warum verarmt unsere Flora? Auswertung der Roten Liste der Farn- und Blütenpflanzen
Deutschlands. – Schriftenr. Vegetationskd. 29: 299–444.
KORNECK, D., SCHNITTLER, M. & VOLLMER, I. (1996): Rote Liste der Farn- und Blütenpflanzen (Pte-
ridophyta et Spermatophyta) Deutschlands. Bonn-Bad Godesberg: Bundesamt für Naturschutz. –
Schriftenr. Vegetationskd. 28: 21–187.
KRUMBIEGEL, A., SCHMIDT, T. & KLOTZ, S. (1998): Artenverschiebung und Einwanderungsprozesse
an einer Brache-Trockenrasen-grenze im Mitteldeutschen Trockengebiet. – Tuexenia 18: 313–330.
LUGINBUHL, J.M., HARVEY, T.E., GREEN JR., J.T., POORE, M.H. & MUELLER, J.P. (1999): Use of goats
as biological agents for the renovation of pastures in the Appalachian region of the United States. –
Agrofor. Sys. 44: 241–252.
MACCHERINI, S., MARIGNANI, M., CASTAGNINI, P. & VAN DEN BRINK, P.J. (2007): Multivariate analy-
sis of the response of overgrown semi-natural calcareous grasslands to restorative shrub cutting. –
Basic Appl. Ecol. 8: 332–342.
MCINTYRE, S., LAVOREL, S. & TREMONT, R.M. (1995): Plant life-history attributes: their relationship
to disturbance response in herbaceous vegetation. – J. Ecol. 83: 31–44.
MELLADO, M., VALDEZ, R., LARA, L.M. & LOPEZ, R. (2003): Stocking rate effects on goats: A research
observation. – J. Range Manag. 56: 67–173.
PARTZSCH, M. (2000): Die Porphyrkuppenlandschaft des unteren Saaletales – Strukturwandel ihrer
Vegetation in den letzten vier Jahrzehnten. – Tuexenia 20: 153–187.
POSCHLOD, P., BAUMANN, A. & KARLÍK, P. (2009): Origin and development of grasslands in central
Europe. – In: VEEN, P., JEFFERSON, R., DE SMIDT, J. & VAN DER STRAATEN, J. (Eds.): Grasslands in
Europe of high nature value: 15–25. KNNV Publishing, Zeist.
POSCHLOD, P. & WALLISDEVRIES, M.F. (2002): The historical and socioeconomic perspective of
calcareous grasslands–lessons from the distant and recent past. – Biol. Conserv. 104: 361–376.
RAHMANN, G. (2000): Biotoppflege als neue Funktion und Leistung der Tierhaltung – dargestellt am
Beispiel der Entbuschung von Kalkmagerrasen durch Ziegenbeweidung. – Schriftenr. Agrar. 28:
1–384.
REICHHOFF, L., KUGLER, H., REFIOR, K. & WARTHEMANN, G. (2001): Die Landschaftsgliederung
Sachsen-Anhalts. Ein Beitrag zur Fortschreibung des Landschaftsprogrammes des Landes Sachsen-
Anhalt. – Magdeburg, Halle: 332 pp.
RICHTER, B., PARTZSCH, M. & HENSEN, I. (2003). Vegetation, Kultur- und Nutzungsgeschichte der
xerothermen Hügellandschaft bei Mücheln/Wettin (Sachsen-Anhalt). – Hercynia N.F. 36: 91–121.
233
ROSENTHAL, G., SCHRAUTZER, J. & EICHBERG, C. (2012): Low-intensity grazing with domestic herbi-
vores: A tool for maintaining and restoring plant diversity in temperate Europe. – Tuexenia 32:
167–205.
RUPPRECHT, D., GILHAUS, K. & HÖLZEL, N. (2016): Effects of year-round grazing on the vegetation of
nutrient-poor grass-and heathlands—Evidence from a large-scale survey. – Agric. Ecosys. Environ.
234: 16–22.
RUPRECHT, E., ENYEDI, M.Z.; ECKSTEIN, L. & DONATH, T.W. (2010): Restorative removal of plant
litter and vegetation 40 years after abandonment enhances re-emergence of steppe grassland vegeta-
tion. – Biol. Conserv. 143: 449–456.
SCHUBERT, R. (Ed.) (2001): Prodromus der Pflanzengesellschaften Sachsen-Anhalts. – Mitt. Flor. Kart.
Sachsen-Anhalt, Sonderheft 2: 1–688.
SCHUBOTH, J. & FRANK, D. (2010): Kartieranleitung Lebensraumtypen Sachsen-Anhalt: Teil Offen-
land. Zur Kartierung der Lebensraumtypen nach Anhang I der FFH-Richtlinie. Stand 11.05.2010. –
Landesamt für Umweltschutz Sachsen-Anhalt, Halle/Saale: 166 pp.
SCHWABE, A. (1997): Zum Einfluß der Ziegenbeweidung auf gefährdete Bergheide-Vegetations-
komplexe: Konsequenzen für Naturschutz und Landschaftspflege. – Nat. Landsch. 72: 183–192.
SCHWABE, A., SÜSS, K. & STORM, C. (2013): What are the long-term effects of livestock grazing in
steppic sandy grasslands with high conservation value? Results from a 12-year field study. –
Tuexenia 33: 189–212.
ŠKORNIK, S. VIDRIH, M. & KALIGARIČ, M. (2010): The effect of grazing pressure on species richness,
composition and productivity in North Adriatic Karst pastures. – Plant Biosys. 144: 355–364.
SMART, A.J., DANIEL, J., BRUNS, K. & HELD, J. (2006): Browsing of Western Snowberry by Goats and
Sheep. – Sheep Goat Res. J. 21: 1–5.
STRANG, R.M. (1973): Bush Encroachment and Veld Management in South-central Africa: the Need
for a Reappraisal. – Biol. Conserv. 5: 96–104.
TÖRÖK, P., VALKÓ, O., DEÁK, B., KELEMEN, A. & TÓTHMÉRÉSZ, B. (2014): Traditional cattle grazing
in a mosaic alkali landscape: Effects on grassland biodiversity along a moisture gradient. – PlosONE
9: e97095.
TÓTH, E., DEÁK, B., VALKÓ, O., KELEMEN, A., MIGLÉCZ, T., TÓTHMÉRÉSZ, B. & TÖRÖK, P. (2018):
Livestock type is more crucial than grazing intensity: Traditional cattle and sheep grazing in short-
grass steppes. – Land Degrad. Dev. 29: 231–239.
VAN SWAAY, C., WARREN, M. & GRÉGOIRE, L. (2006): Biotope use and trends of European butterflies.
– J. Insect Conserv. 10: 189–209.
VEEN, P., JEFFERSON, R., DE SMIDT, J. & VAN DER STRAATEN, J. (Eds.) (2009): Grasslands in Europe
of high nature value. – KNNV Publishing, Zeist: 319 pp.
VEITH, M., BONN, S., SANDER, U., ALBRECH, J. & POSCHLOD, P. (2012): Nachhaltige Entwicklung
xerothermer Hanglagen am Beispiel des Mittelrheintals - eine naturschutzfachliche, ökonomische
und sozio-kulturelle Bewertung: Ergebnisse des gleichnamigen E+E-Vorhabens. – Naturschutz und
Biologische Vielfalt 121. 357 pp.
WALLISDEVRIES, M.F., POSCHLOD, P. & WILLEMS, J.H. (2002): Challenges for the conservation of
calcareous grasslands in northwestern Europe: integrating the requirements of flora and fauna. –
Biol. Conserv. 104: 265–273.
WEDL., N. & MEYER, E. (2003): Beweidung mit Schafen und Ziegen im NSG Oderhänge Mallnow. –
Naturschutz Landschaftspfl. Brandenbg. 12: 137–143.
WESSELS, S. & SCHWABE, A. (2008): Testing the potential seed availability in dung samples: compari-
son of two seedling emergence methods. – Flora 203: 429–436.
WILSON, J.B., PEET, R.K., DENGLER, J. & PÄRTEL, M. (2012): Plant species richness: the world rec-
ords. – J. Veg. Sci. 23: 796–802.
ZEHM A. (2008): Praxis der Erstpflege von gehölzreichen, basenreichen Sandrasen. – Nat. Landsch. 83:
541–547.
