Applied Vegetation Science 17 (2014) 591–603
SPECIAL FEATURE: VEGETATION SURVEY
Grassland vegetation of the Molinio-Arrhenatheretea
class in the NW Balkan Peninsula
Urban Silc, Svetlana Acic, Zeljko Skvorc, Daniel Krstonosic, Jozo Franjic & Zora Dajic
Stevanovic
Keywords
Balkan region; Classification; Europe;
Grassland; Mesic meadows; Pastures;
Phytosociology; Vegetation survey
Nomenclature
Vascular plants Tutin et al. (1964–1993) except
for Scirpus georgianus Harper; Higher syntaxa
Mucina et al. (2014)
Received 10 July 2013
Accepted 7 December 2013
Co-ordinating Editor: Milan Chytry
Silc, U. (corresponding author,
urban@zrc-sazu.si): ZRC SAZU, Institute of
Biology, Novi trg 2, 1000, Ljubljana, Slovenia;
Biotechnical Centre Naklo, Strahinj 99, 4202,
Naklo, Slovenia
Acic, S. (acic@agrif.bg.ac.rs) &
Dajic Stevanovic, Z. (dajic@agrif.bg.ac.rs):
Faculty of Agriculture, University of Belgrade,
Nemanjina 6, 11080, Zemun, Serbia
Skvorc, Z. (zskvorc@sumfak.hr),
Krstonosic, D. (dkrstonosic@sumfak.hr) &
Franjic, J. (jfranjic@sumfak.hr): Faculty of
Forestry, University of Zagreb, Svetosimuska c.
25, 10000, Zagreb, Croatia
Virtual Special Feature: “Towards a consistent
classification of European grasslands” (Eds.
J€urgen Dengler, Erwin Bergmeier, Wolfgang
Willner & Milan Chytry).
Abstract
Questions: How does the floristic composition of plant species of meadows and
mesic pastures vary along a broad geographical gradient in the NW Balkans?
How does the current phytosociological classification of the Molinio-Arrhenatheretea
vegetation differ among the NW Balkan countries?
Location: NW Balkans (Slovenia, Croatia, Bosnia and Herzegovina, Serbia).
Methods: 3635 releves originally assigned to the class Molinio-Arrhenatheretea
were classified with a beta flexible method, and the crispness of classification
was checked. DCA ordination with Pignatti indicator values and climate data
were applied to show the influence of site conditions on species composition.
Results: The classification was best interpreted at the level of 13 clusters, but
could also be interpreted at the level of three groups of clusters. The first division
was according to geography and climate: the first and third groups were concentrated
in the NW part, while the second was restricted to the eastern part of the
study area. The most important variable was site moisture, followed by nutrients
and altitude, which corresponded with a west–east direction. The first group
was very diverse and included communities on the wettest and most nutrientrich
sites (Potentillion anserinae, Cynosurion cristati, Calthion palustris, Molinion
caeruleae, Molinio-Hordeion). The second group comprised mesophilous continental
grasslands (Trifolio-Ranunculion pedati, Trifolion pallidi, Trifolion resupinati),
while the third group consisted of grasslands from regions with abundant precipitation
(Arrhenatherion elatioris, Deschampsion cespitosae, Pancicion serbicae, Triseto
flavescentis-Polygonion bistortae).
Conclusions: Our analysis can be used to unify different phytosociological classifications
in different countries, also showing the transitional forms of wellknown
Central European vegetation types that have a different floristic composition
and ecology in the Balkans. This knowledge will enable classification of the
same vegetation types in neighbouring Balkan countries that are less studied.
Introduction
The Balkans is a peninsula in the form of an irregular
triangle that extends from Central Europe in the north
to the Eastern Mediterranean in the south (Reed et al.
2004). The complexity of its physical geography and
transitional location of floral (and faunal) influences
have made this region a European biodiversity hotspot.
The latest estimates have raised the number of autochthonous
vascular plant species to 8000 in the Balkan
Peninsula, which makes it one of the three (together
with the Italian and Iberian Peninsulas) most important
centres of floristic diversity in Europe (Stevanovic et al.
2003). The Balkans is the most species-rich region in
Europe in terms of plant endemism, with 2600–2700
endemic plant taxa (Stevanovic et al. 2007). Moreover,
the region is known for its very diverse vegetation,
including various types of natural and semi-natural
grasslands (Dajic-Stevanovic et al. 2010; Acic et al.
2013b).
Grasslands are habitats dominated by graminoids and
are the most widely distributed vegetation type in the
591
Applied Vegetation Science
Doi:10.1111/avsc.12094 © 2014 International Association for Vegetation Science
cultural landscape. In addition to natural grasslands, large
areas are covered with grasslands that are a product of
lengthy human influence. They thrive in various ecological
conditions (wet, mesic and dry) and under different
management regimes (mown, grazed, fertilized), all resulting
in different biological patterns (Sebastia 2004; Klimek
et al. 2007; Putfarken et al. 2008; Wesche et al. 2012) and
patterns of ecosystem functioning and diversity (Laliberte
et al. 2010). Most grasslands represent semi-natural vegetation
developed during a long history of human use. Different
types of historical and current grassland use have
resulted in various plant and vegetation diversity patterns.
Traditionally managed semi-natural grasslands (hay meadows
and pastures) are known to support a rich flora and
are recognized for high species diversity in various regions
of Europe, including the Balkans (Dajic-Stevanovic et al.
2010; Wilson et al. 2012). Many grassland communities
are now being threatened by rapid changes in agricultural
practices, especially related to the effects of either land
abandonment (Pyk€al€a et al. 2005; Dajic Stevanovic et al.
2008) or intensification (Stevens et al. 2004) as contrasting
management practices, whereby the latter often results in
the spread of P- and N-demanding competitive grasses tolerant
to mowing (Wesche et al. 2012; Ceulemans et al.
2013).
In the agricultural landscapes of Central Europe (as well
as in the Balkans), moist and mesic grasslands are among
the habitat types that have experienced the severest losses
(Prach 2008), thus attracting particular research and conservation
attention. The class Molinio-Arrhenatheretea
includes mown meadows and mesic pastures of temperate
regions of Europe and adjacent parts of Asia (Hajkova et al.
2007). In Central Europe, several alliances are widely recognized
and stable in international classifications (Arrhenatherion
elatioris, Molinion caeruleae, Calthion palustris), while
the classification of lowland wet meadows is less consistent
(Botta-Dukat et al. 2005). In the Balkans, due to large gradients
between sub-oceanic and continental climates with
a Mediterranean influence, these inconsistencies are even
more pronounced and also appear in well delimited syntaxa
in other parts of Europe. Another reason for ambiguous
classification, in addition to various climatic
influences, is that some syntaxa are at the limit of their distribution.
On the other hand, grasslands of Molinio-Arrhenatheretea
in the Central Balkans are rich in endemics (91
taxa; Tomovic 2007). The frequent occurrence of endemic
and rare plants has also been reported for several alliances
of this class in the Central European region (Ruzickova
et al. 2004). Species-rich mesic meadows of Molinio-Arrhenatheretea
were therefore included in the European Habitats
Directive (92/43/EEC; European Union 1992/1995;
NATURA 2000-Codes 64 and 65). Classification of grassland
vegetation of the NW Balkans is thus important for
incorporating these vegetation types into the European
scheme.
The territory of the NW Balkans (or former Yugoslavia)
has a long tradition of vegetation research, which, however,
as of recently has not been transnational (Kojic et al.
1998; Trinajstic 2008; Silc & Carni 2012). This has led to
syntaxonomic systems that are not comparable and, as Horvat
et al. (1974) have already mentioned, different authors
have classified grasslands into different syntaxa using local
experience. The study area is a transitional region between
different climatic influences, and this makes syntaxonomical
classification more difficult. Two climatic zones (the
Mediterranean and Central European) overlap in the
Balkans, with a large transitional area and pronounced
climate modifications, depending on the orientation of
mountain massifs, lowlands, valleys and altitude, and to a
large extent influenced by dominant winds. According to
Polunin (1987), lowlands and valleys are of great importance
for plant species distribution, since they enable
advance of the mediterranean climate into the interior.
Another problem is that stands are transitional between
syntaxa, due to changes to or even abandonment of management.
Such stands make phytosociological classification
even more ambiguous.
Use of large vegetation databases enables large-scale
comparisons and some efforts towards identifying distinct
vegetation types have already been made in the NW Balkans
(Kosir et al. 2008, 2013; Silc et al. 2008; Carni et al.
2009; Marinsek et al. 2013), but not for grasslands, and in
particular not for mesic grasslands. The main aim of this
study was to determine the main vegetation types of the
class Molinio-Arrhenatheretea in the NW Balkans, to delimit
alliances geographically and ecologically and to propose a
vegetation classification system on a higher level.
Material
A large data set of grassland vegetation releves (4615
releves originally assigned by authors into the class Molinio-Arrhenatheretea)
was compiled from the territory of the
NW Balkan Peninsula (Bosnia and Herzegovina, Croatia,
Serbia, Slovenia). The study area extends over
216 000 km2
(13°22′W to 23°00′E and 46°54′N to 41°50′
S). The terrain is predominantly hilly, with 72% of the
area (except for the large Pannonian plain) higher than
200 mÁa.s.l.; the bedrock is mainly carbonate. The climate
is very diverse: a temperate warm climate predominates,
while smaller regions have mountain (boreal) and true
mediterranean climates (Bertic 1987).
The data set consists mainly of published releves, but
includes some that are unpublished (GIVD EU-HR-002,
EU-RS-002, EU-SI-001). The releves are stored in TURBOVEG
format (Hennekens & Schaminee 2001). Outlier
Applied Vegetation Science
592 Doi:10.1111/avsc.12094 © 2014 International Association for Vegetation Science
Grassland vegetation of the NW Balkans U. Silc et al.
analysis was processed using PC-ORD 5.0 (MjM Software
Design, Gleneden Beach, OR, US) and 67 releves whose
species composition deviated more than Æ2 SD from
the mean calculated Euclidean distance of all plots were
omitted.
Releves were georeferenced a posteriori. To reduce the
over-sampling bias, the data set was subsequently geographically
stratified by 6′ latitude 9 10′ longitude grid
cells and ten releves were selected from each grid cell by
heterogeneity-constrained random resampling (Lengyel
et al. 2011) to obtain balanced floristic diversity. Finally,
3635 releves were included in the analysis.
Species nomenclature was checked and synonyms and
taxa determined at various ranks were aggregated to a
common taxonomic level (e.g. Agrostis stolonifera
agg. = A. gigantea, A. stolonifera; Brachypodium pinnatum
agg. = B. pinnatum, B. rupestre; Bromus racemosus agg. =
B. racemosus, B. commutatus; Carex flava agg. = C. flava,
C. lepidocarpa; Centaurea jacea agg. = C. jacea, C. angustifolia,
C. macroptilon; Festuca pratensis agg. = F. pratensis, F. arundinacea;
Galium mollugo agg. = G. album, G. mollugo; Leucanthemum
vulgare agg. = L. vulgare, L. praecox; Polygonum
aviculare agg. = P. aviculare, P. arenastrum; Vicia sativa
agg. = V. angustifolia, V. sativa; Vicia pannonica s. lat =
V. pannonica, V. striata). Records of species determined to
the genus level and bryophytes and lichens were deleted
from the data set since they were not consistently sampled
by all of the authors. Species cover values in original
releves were estimated on different scales. They were a
posteriori transformed to percentages and square-rooted.
Climatic variables (annual temperature, precipitation
and BIO 18 – precipitation of the warmest quarter) were
subjectively selected, since we presumed that they have
strong effects on species composition. They were compiled
for each releve from the WorldClim database (Hijmans
et al. 2005).
Environmental indicator values of Pignatti (2005) were
used for ecological interpretation of gradients. Unweighted
mean indicator values were calculated for each releve and,
in the same way, for columns in the synoptic table.
Methods
We classified the data set (3635 releves) using cluster
analysis in the PC-ORD 5.0. The Sørensen index as the distance
measure and beta flexible (^a = À0.25) for group
linkage were used. We made several classifications with
different numbers of clusters of releves. We used the OptimClass
method (Tichy et al. 2010) for identifying the
optimal partition and the peak of the OptimClass1 curve
showed the optimal number of clusters is 13 at threshold
P < 10EÀ10
. We accepted this classification with 13 clusters
as being the ecologically soundest. Other classifications
with more clusters produced groups with ambiguous ecological
and syntaxonomic significance.
The synoptic table was produced in the JUICE program
and the phi (u) coefficient was used as the measure of
fidelity. Each cluster was compared to the remaining
releves in the data set, which were taken as a single undivided
group. Since the clusters consisted of unequal numbers
of releves, higher u values for larger clusters were
expected. To avoid that, each of the 13 clusters was virtually
equalized to 1/13 of the size of the entire data set
(Tichy & Chytry 2006). The threshold of the u value was
subjectively selected at 0.10 for a species to be considered
diagnostic. This threshold was considered to be optimal for
interpreting a single cluster and for preventing inflation of
diagnostic species.
Relationships between clusters and environmental indicator
values were visualized using DCA ordination, in
which the square-rooted percentage covers of species were
used. The mean ecological indicator values (based on Pignatti
2005) were plotted a posteriori on the DCA ordination
diagram. Ordination results are presented on spider plots,
in which each releve is linked to the centroid of its cluster
by a line. Analysis was performed in R (R Foundation for
Statistical Computing, Vienna, AT; v 2.15.2, http://www.
r-project.org) using vegan package (http://cc.oulu.fi/~
jarioksa/softhelp/vegan.html).
Results
Classification
Classification could first be interpreted at the level of three
groups of clusters (Fig. 1, clusters 1–6, 7–9 and 10–13;
Table 1), which reflected geographic and climatic factors,
with the first and third group concentrated in the NW part
and the second group restricted to the E part of the study
area. The first group is very diverse and includes the wettest
communities on the one hand and the most nutrientrich
communities on the other. The second group comprises
mesophilous continental grasslands, while the third
group of clusters consists of grasslands on deeper soils from
regions with high precipitation. The most meaningful and
finer classification is at the level of 13 clusters (Fig. 1,
Table 1).
Cluster 1 comprises nitrophilous, semi-ruderal grasslands
along water bodies dominated by Agrostis stolonifera
agg.
Cluster 2 represents typical vegetation of the alliance Cynosurion
cristati from temperate Europe. Communities are
dominated by species adapted to trampling (Lolium perenne,
Plantago major), while the species Cynosurus cristatus in the
NW Balkans is characteristic for the association Bromo-Cynosuretum
(Cluster 12).
593
Applied Vegetation Science
Doi:10.1111/avsc.12094 © 2014 International Association for Vegetation Science
U. Silc et al. Grassland vegetation of the NW Balkans
Cluster 3 contains releves of the alliance Calthion palustris,
dominated by hygrophilous herbs. They are found on wet,
nutrient-rich soils with a large altitudinal range (they are
found in lowlands and up to 1800 m).
Cluster 4 comprises vegetation dominated by Molinia
caerulea agg. distributed mostly in Slovenia with some
occurrences also in the south (Caltho-Alopecuretum in Vojvodina
and at higher altitudes in Serbia, with a mix of Deschampsia
caespitosa and Polygonum bistorta). Stands are wet,
and compared to other clusters nutrient-poor and acidic.
Cluster 5 corresponds to the Illyrian-Mediterranean alliance
Molinio-Hordeion secalini, which is found on Karst poljes.
It is geographically limited to Slovenia, Croatia and
Bosnia and Herzegovina.
Cluster 6 represents a rare plant association Scirpo holoschoeni-Salicetum
rosmarinifoliae found on sandy sites along
rivers and classified into the Molinion caeruleae alliance.
Communities are dominated by Salix repens and xeric grassland
species.
Cluster 7 includes wet lowland grasslands in the eastern
continental part of the NW Balkans (E Croatia and central
Serbia). In addition to grasses, the most important species
group consists of clovers (Trifolium patens, T. pallidumi
T. resupinatum) and other Fabaceae (Medicago arabica,
M. polymorpha, Lathyrus tuberosus).
Cluster 8 covers mesophilous grassland vegetation on the
Pannonian plain that can thrive in sites with high variability
of soil moisture and slight salinity. The grass species Poa
angustifolia, Poa palustris and Alopecurus pratensis are characteristic
for the communities classified into the alliance Trifolio-Ranunculion
pedati.
Cluster 9 comprises plant communities classified into the
Balkan sub-halophytic alliance Trifolion resupinati. These
are sub-Mediterranean lowland meadows in the S and SE
part of the research area.
Cluster 10 includes releves with the largest altitudinal
range. There are communities from higher altitudes (Pancicion
serbicae) as well as stands that are poor in characteristic
species or were classified to higher syntaxa (mainly into
the order Arrhenatheretalia) and are similar in floristic composition
to stands from higher altitudes. The map of this
cluster therefore shows mixed geographic distribution with
communities of high altitudes and depauperate stands
found in lowlands.
Cluster 11 consists of typical grasslands of Central Europe
that are classified into alliances Arrhenatherion elatioris, Triseto
flavescentis-Polygonion bistortae and Poion supinae and
dominated by the grasses Arrhenatherum elatius, Trisetum
flavescens, Dactylis glomerata, Avenula pubescens.
Cluster 12 represents widespread wet stands of the alliance
Arrhenatherion elatioris, mostly the association BromoCynosuretum
comprising lowland grasslands temporarily
flooded in spring. Stands are characterized by Holcus lanatus
and Cynosurus cristatus. In Central Europe, the latter species
is characteristic for trampled and frequently mowed grass-
lands.
Fig. 1. Classification dendrogram of grassland dataset (3635 plots). 1 – Potentillion anserinae, 2 – Cynosurion, 3 – Calthion, 4 – Molinion, 5 – MolinioHordeion
secalini, 6 – Scirpo holoschoeni-Salicetum rosmarinifoliae, 7 – Trifolion pallidi, 8 – Trifolio-Ranunculion pedati, 9 – Trifolion resupinati, 10 –
Pancicion serbicae & species poor Arrhenatheretalia, 11 – Arrhenatherion s.str. & Triseto flavescentis-Polygonion bistortae, 12 – Bromo-Cynosuretum, 13 –
Deschampsion. The numbers of clusters refer to Table 1 and the number of releves included in each cluster is indicated.
Applied Vegetation Science
594 Doi:10.1111/avsc.12094 © 2014 International Association for Vegetation Science
Grassland vegetation of the NW Balkans U. Silc et al.
Table 1. Synoptic table of Molinio-Arrhenatheretea grasslands in the NW Balkans. The frequency values are shown, shaded are species with phi values
higher than 0.1. Only five species with highest fidelity are presented.
Cluster 1 2 3 4 5 6 7 8 9 10 11 12 13
No. of releves 240 128 279 311 198 21 452 199 258 457 473 346 273
% releves for country
Slovenia 31 25 23 68 10 0 5 0 0 24 44 33 55
Croatia 42 49 17 4 64 0 12 21 0 29 55 66 38
Bosnia and Herzegovina 4 0 1 1 26 0 3 0 0 11 0 0 0
Serbia 24 26 59 26 1 100 81 79 100 36 1 0 7
Cluster 1: Potentillion anserinae
Agrostis stolonifera agg. 93 38 21 23 22 – 29 18 10 7 3 26 25
Mentha longifolia 38 6 14 3 – – 6 2 6 6 5 1 1
Juncus inflexus 32 7 13 4 1 – 2 1 – 1 – 2 4
Alopecurus geniculatus 10 – 1 – – – – – – – – – –
Rorippa sylvestris 46 22 8 1 6 – 21 20 25 4 1 1 2
Cluster 2: Cynosurion
Lolium perenne 22 85 2 – 8 – 9 4 62 15 23 17 2
Cynodon dactylon 12 56 2 – 5 – 3 18 1 4 1 1 1
Plantago major 30 64 6 3 2 – 9 10 9 12 4 4 2
Trifolium repens 47 88 27 4 7 – 28 26 38 42 35 36 20
Poa annua 11 29 – – – – 1 4 – 8 1 1 –
Cluster 3: Calthion
Scirpus sylvaticus 3 – 45 3 – – 1 – 4 2 1 4 5
Filipendula ulmaria 5 – 34 29 3 – 1 – – 4 2 13 25
Eriophorum latifolium – – 21 15 1 – – – – 1 – – 1
Equisetum palustre 3 – 37 22 2 10 2 10 6 2 3 9 15
Caltha palustris 3 – 34 10 – – 1 4 1 3 – – 4
Cluster 4: Molinion coeruleae
Molinia caerulea agg. – – 15 93 31 29 – – – 2 2 3 15
Schoenus ferrugineus – – – 15 – – – – – – – – –
Potentilla erecta 2 – 28 73 8 – 1 1 – 13 15 26 44
Succisa pratensis 1 – 11 49 5 – 1 – – 1 1 19 29
Sanguisorba officinalis – – 10 48 20 – 4 1 1 1 3 13 24
Cluster 5: Molinio-Hordeion secalini
Scilla litardierei – – – 1 47 – – – – – – – 2
Deschampsia media – – – – 26 – 1 – – – – – –
Peucedanum coriaceum – – – 3 39 – 1 – – – – – 6
Centaurea pannonica 1 – – 8 33 – 1 – – 1 1 4 1
Sesleria caerulea – – – – 17 – 1 – – – – – –
Cluster 6: Scirpo holoschoeni-Salicetum rosmarinifoliae
Salix rosmarinifolia – – – 4 3 100 – – – – – 1 1
Calamagrostis epigejos – 1 1 3 – 100 2 2 – 5 2 1 2
Scirpus holoschoenus – – – 3 1 71 – – – – – – 1
Inula salicina – – 1 4 8 90 11 8 – 1 4 2 6
Scabiosa ochroleuca – – – 2 – 86 – 2 – 2 – – –
Cluster 7: Trifolion pallidi
Medicago arabica 1 1 – – – – 19 1 1 – – 1 –
Trifolium pallidum – – 1 – 1 – 17 9 8 2 1 1 1
Poa trivialis 20 2 32 10 1 – 49 16 – 13 25 32 19
Clematis integrifolia – – 1 1 5 – 16 11 – – 1 1 –
Poa pratensis 4 5 5 7 2 – 61 22 14 29 53 47 29
Cluster 8: Trifolio-Ranunculion pedati
Poa angustifolia – – – 1 4 5 – 28 – 3 1 1 3
Poa palustris 3 2 10 1 1 – 1 34 – – 1 1 8
Euphorbia lucida – – – – – – – 24 – – – – –
Festuca pseudovina 1 14 – – 5 – 2 25 – 2 1 – –
Trifolium angulatum 2 – – – – – 1 20 – – – – –
595
Applied Vegetation Science
Doi:10.1111/avsc.12094 © 2014 International Association for Vegetation Science
U. Silc et al. Grassland vegetation of the NW Balkans
Cluster 13 includes communities dominated by Deschampsia
cespitosa on clay soils with alternating wet and dry
phases. These grasslands are found in S Slovenia, the continental
part of Croatia and in some localities at higher elevations
in Serbia.
Ordination analyses
The first DCA axis shows the gradient of altitude and nutrients,
and this gradient also corresponds to a W–E direction
(Figs 2 and 3). Temperature and BIO18 (precipitation of
the warmest quarter) as climatic variables also correspond
to the first axis. The most important variable is moisture of
the site and coincides with the second DCA axis; it is striking
that continentality shows an opposite trend to mois-
ture.
Discussion
Moisture and nutrients (fertility) were found to be the
most important gradients determining the floristic
composition of mesic grasslands in the NW Balkans. This is
congruent with several studies of this type of grassland
(Havlova et al. 2004; H€ardtle et al. 2006; Zelnik & Carni
2008). Next is the temperature (warmth) of the site, which
correlates with climatic variables, in particular precipitation
of warmest quarter of the year (Appendix S1). Climatic
variables, especially those linked to precipitation, are
important in forming plant communities, and this gradient
corresponds to the NW–SE gradient along the peninsula,
linking Central European communities to Pannonian and
(sub)-Mediterranean ones. The importance of climate for
the species composition of wet grasslands that are saturated
by surface or groundwater has already been demonstrated
(Ilijanic 1973; Hajek et al. 2008). The vegetation
structure and species composition in SE European peninsulas
is strongly influenced by biogeographic and ecological
differences (Blasi et al. 2012; Elias et al. 2013).
In addition to site conditions, management is very
important for the establishment of different grassland communities
(Dierschke & Briemle 2002; Zelnik 2005). Different
management practices, such as cutting, grazing,
application of artificial fertilizers, sowing species, can significantly
alter species composition (Waldhardt & Otte
2003; Havlova et al. 2004; Drobnik et al. 2011), which
results in different grassland types (e.g. species-poor comTable
1. (Continued).
Cluster 1 2 3 4 5 6 7 8 9 10 11 12 13
Cluster 9: Trifolion resupinati
Poa trivialis subsp. sylvicola – – – 1 17 – 1 – 83 – 1 1 1
Ranunculus velutinus – – – – – – 4 – 53 – – – –
Trifolium balansae – – – – – – – – 25 – – 1 –
Alopecurus rendlei – – – – 5 – 6 – 53 – 1 5 2
Tragopogon orientalis – – – 1 1 38 7 1 61 6 17 8 1
Cluster 10: Pancicion + Triseto-Polygonion bistortae + species poor Arrhenatheretalia
Agrostis capillaris 1 – 9 2 12 – 2 – 1 33 7 18 4
Festuca rubra agg. – – 4 18 9 – 3 – – 50 27 34 25
Hypochaeris maculata – – – – – – – – – 13 1 – –
Silene sendtneri – – – – – – – – – 11 – – –
Pimpinella serbica – – – – – – – – – 8 – – –
Cluster 11: Arrhenatherion
Arrhenatherum elatius – – 1 1 – – 4 – – 28 87 19 –
Trisetum flavescens – – 1 1 – – 1 – 1 18 75 32 1
Dactylis glomerata 5 5 2 10 6 – 9 4 1 49 85 31 7
Salvia pratensis – – – – 1 – 6 2 – 9 43 2 –
Knautia arvensis – – – 2 – – 5 1 – 16 47 12 –
Cluster 12: Bromo racemoso-Cynosuretum cristati
Cynosurus cristatus 4 5 27 13 16 – 38 – 44 39 32 86 42
Gaudinia fragilis – – 1 – – – 1 – – 1 3 23 3
Rhinanthus minor 1 – 25 11 24 – 4 3 27 13 34 56 18
Ranunculus acris 12 2 35 50 39 – 28 13 40 24 72 94 75
Ononis arvensis 1 9 2 1 – – 8 3 5 – 18 36 5
Cluster 13: Deschampsion caespitosae
Deschampsia cespitosa 6 – 37 30 21 – 3 4 5 8 5 30 77
Agrostis canina agg. 2 – 5 10 – – 2 10 4 1 – 16 47
Succisella inflexa – 2 8 5 1 – 1 – – – – 7 40
Ranunculus flammula 2 – 6 4 1 – 2 – – – – 8 36
Juncus conglomeratus 1 – 24 16 2 – 1 4 – 1 1 24 49
Applied Vegetation Science
596 Doi:10.1111/avsc.12094 © 2014 International Association for Vegetation Science
Grassland vegetation of the NW Balkans U. Silc et al.
munities). In large-scale analyses of vegetation databases,
it is difficult to incorporate management into the analysis
as an explanatory variable, and such wide geographic and
climatic analyses are rare (Moog et al. 2002; Wellstein
et al. 2007).
The syntaxonomic scheme of the class Molinio-Arrhenatheretea
(Appendix S2) is the first attempt of such scope in
the NW Balkans based on releve material from a large
database and on multivariate methods (classification and
ordination), and to the association level. The Balkans is a
transitional area in terms of climate and phytogeography,
and classification of certain grassland syntaxa is therefore
very difficult. Certain species with clear diagnostic values
in Central Europe, for example, are at the limit of their distribution
here and their coenological optima shifts. Rather
than offering definite answers, our analysis therefore generates
new questions about the syntaxonomic system.
Horvatic (1939) placed the geographic border between
the alliances Molinion caeruleae and Deschampsion cespitosae in
SE Slovenia and this is also confirmed by our analysis
(Fig. 3). In the NW Balkans, Molinion caeruleae is found in
the NW part (e.g. Slovenia), while Deschampsion cespitosae is
limited mainly to the continental part of Croatia along the
Sava River to the river Orljava (Ilijanic 1973), but occurs
also in Serbia and Bulgaria. Humid climate, higher altitudes
and higher amounts of organic matter are more characteristic
of the Molinion caeruleae alliance (Zelnik & Carni 2013).
The diversity of Molionion caeruleae is higher in the NW Balkans
than in neighbouring regions. One reason is the influence
of different phytogeographic regions on species
composition; another is the traditional classification into
several narrower associations (acido- and basophilous, on
Sphagnum mires). Classification also supports the findings
of Botta-Dukat et al. (2005) that Cnidion venosi and Deschampsion
cespitosae cannot be interpreted as two geographically
vicariant alliances, Pannonian and Illyrian,
respectively, as proposed by Ellmauer & Mucina (1993).
The alliance Agrostion albae is classified within the alliance
Deschampsion cespitosae in several national surveys
(Chytry 2007; Borhidi et al. 2012), while it was classified
as a separate alliance in Serbia (Kojic et al. 1998). In our
numerical classification, communities of Agrostion albae
from Serbia are classified into Potentillion anserinae (Acic
et al. 2013a).
There is a striking difference in the distribution pattern
of Deschampsion cespitosae in the Balkans. In the NW part of
the peninsula, a typical region of this alliance, stands are
found on flooded lowlands up to 400 mÁa.s.l. (Ilijanic
1973; Botta-Dukat et al. 2005), while in the eastern part of
the Balkans they are found at higher elevations (700–
1000 mÁa.s.l.; Hajek et al. 2008; Milosavljevic et al. 2008;
Randelovic & Zlatkovic 2010). It was previously considered
that this alliance is absent to the east of the river Orljava
and very rare in Serbia (Ilijanic 1973; Kojic et al.
1998) but these stands were later included in Molinietum
caeruleae (Kojic et al. 2004). In the southern part of the
study area, Molinia and Deschampsia occur together in
stands, and the difference in nutrient availability is responsible
for their habitat differentiation (Hajek et al. 2008). In
the central Balkans, Molinia and Deschampsia are found at
higher elevations, where climatic conditions resemble
those in Central European lowlands.
Cynosurus-dominated communities are very common
in the Balkans, where several associations with Cynosurus
as a characteristic species are described. In the numerical
analysis, these Cynosurus-dominated communities are
classified into two distinct clusters. The first is the Cynosurion
cristati alliance typical of temperate Central Europe,
on grasslands that are intensively grazed and mown.
Another Cynosurus community is the association Bromo
racemosi-Cynosuretum found in continental Croatia and
Slovenia, which develops on temporarily inundated soils
and is the most widespread mown meadow in the region
(Horvatic 1930). The Cynosurion cristati alliance is widespread
in W and Central Europe, and concentrated in the
sub-montane and montane zone towards the SE (Zuidhoff
et al. 1995); it is represented by the well-delimited
association Lolio perennis-Cynosuretum. The presence of the
species Cynodon dactylon indicates the more southern and
0
0
–3 –2 –1 1 2 3
–3–2–11234
DCA1
DCA2
1
2
3
4
6
7
9
10
11
12
13
Light
Temperature
Continentality
Moisture
Reaction
Nutrients
Annual temperature
Precipitation
Bio 18
Altitude
Fig. 2. DCA ordination of the whole dataset. Centroids of clusters from
the classification are indicated and ecological variables (ecological
indicators and the most important climatic variables) are presented. Eigen
values of axes are: 1st axis 0.449 and 2nd axis 0.405 and total inertia
12.784.
597
Applied Vegetation Science
Doi:10.1111/avsc.12094 © 2014 International Association for Vegetation Science
U. Silc et al. Grassland vegetation of the NW Balkans
warmer character of these stands compared to this type
of vegetation in Central Europe. The position of BromoCynosuretum
is unclear; according to (Stancic 2008) characteristic
species have a broad ecological range and
should be divided into a number of associations. In our
analysis, these stands (together with some wet Arrhenatherion
elatioris stands) are separated as a single cluster
with a transitional position between Cluster 5 and 10 in
the DCA diagram (Fig. 2), as also pointed out by Lengyel
et al. (2012). The species composition and DCA analysis
suggest that these grasslands should be classified within
the alliance Arrhenatherion elatioris. Another possibility is
classification into the alliance Deschampsion cespitosae but
there are more species present that are characteristic for
the alliance Arrhenatherion elatioris. Classification of
Bromo-Cynosuretum into higher syntaxa is therefore not
clear; various authors classify it into the alliance Cynosurion
cristati or Arrhenatherion elatioris (Kojic et al. 1998; Trinajstic
2008; Velev et al. 2011a). Stancic (2008) indicated
that some of the releves of Bromo-Cynosuretum should be
classified into Arrhenatherion elatioris and some into Deschamspion
cespitosae.
Cluster 11 comprises the alliances Arrhenatherion elatioris
and Triseto flavescentis-Polygonion bistortae, which are typical
of the NW part of the study area, which is already part of
the SE Alps. Releves of these two alliances merged into
one cluster since their sites are similar in terms of moisture
and nutrients (Heged€usova et al. 2011). Communities of
Arrhenatherion elatioris from the south are less mesic and
lack characteristic species, while the alliance Triseto flavescentis-Polygonion
bistortae is not present in the central Balkans
(Apostolova et al. 2007). It is replaced by the alliance
Arrhenatherion elatioris, which occurs at elevations of 700–
1600 m. The altitudinal shift of plant communities towards
the south of the Balkans is well known (Horvat et al. 1974;
Hajek et al. 2008) and is the reason for the ambiguous classification
of vegetation that spreads over the long biogeographic
gradient.
Fig. 3. The distribution of releve´s included in particular clusters. The numbers of clusters refer to Table 1.
Applied Vegetation Science
598 Doi:10.1111/avsc.12094 © 2014 International Association for Vegetation Science
Grassland vegetation of the NW Balkans U. Silc et al.
The association Arrhenatheretum s. lat. is widespread in
Central and Western Europe, and the dominant species
gives the stands an easily recognizable physiognomy. In
many cases, stands were classified as belonging to Arrhenatheretum
s. lat. because of the presence of the dominant
grass species. The association changes towards the
south and communities become more ruderal (Velev
et al. 2011b). In Serbia these communities were previously
considered as sown grasslands (Jovanovic-Dunjic
et al. 1986) or were found mostly along roads (Parabucski
1990). They are also less commercially important,
found in untypical ecological conditions in the
Pannonian region (Vojvodina) (Kojic et al. 2005) or in
the montane belt in central Serbia (Acic et al. 2013b).
Arrhenatherion elatioris stands from Serbia are therefore
in the same cluster as ruderalized mesic grasslands from
the northern part of the study area. On the other hand,
stands of the alliance Arrhenatherion elatioris from Bulgaria
are more similar to those of Central European
(Velev et al. 2011b).
Pancicion serbicae is an alliance found at higher elevations,
rich in endemic species and with a Dinaric distribution
(Lakusic 1966). Pancicia serbica appears in various
plant communities (forest and non-forest) and has a
disjunct distribution in the Balkans (Obratov & Dukic
1998). Despite endemic species, the alliance was not
clearly delimited in our analysis and merged with speciespoor
stands and stands classified at higher syntaxonomic
levels without characteristic species. The factors filtering
the species composition at higher elevations seem similar
to those of human-disturbed sites at lower altitudes. Grazing
disturbance and altitudinal stress resemble high human
impact in the lowlands (e.g. road verges, intensive mowing).
Management can blur delimited communities, especially
if there is a combination of hay-cutting and grazing
or a change of pasture into meadow (or vice versa) (Havlova
et al. 2004).
The alliance Trifolion pallidi is mainly found in eastern
Croatia and central Serbia and the alliance Trifolion resupinati
is limited to southern Serbia, where the climate is warmer
and more arid. Our analysis shows the distribution of
Trifolion pallidi to be the most eastern part of Serbia, on the
border with Bulgaria, although it has not so far been identified
in Bulgaria (Elias et al. 2013). It is linked to the climax
vegetation Quercion frainetto, and when Ilijanic (1969)
described the alliance, he assumed its distribution in Bulgaria
and Romania.
Short descriptions of the alliances
On the basis of the literature review and our own results,
we present a unified syntaxonomic list, which is critically
commented and highlights the syntaxonomic problems for
further analysis (Appendix S2). Each alliance is presented
in brief.
Molinion caeruleae
These grasslands are found on wet or moist soils, sometimes
even on peaty soils. The water table is high during
one period of the vegetation season. Molinion grasslands
are concentrated in Slovenia, with some localities in
Pannonia and central Serbia. The characteristic and
dominant species is Molinia caerluea agg., accompanied
by Schoenus ferrugineus, Potentilla erecta and Sanguisorba of-
ficinalis.
Calthion palustris
Wet grasslands and tall herb communities that are frequently
unmanaged. They are found on flat lands along
streams or on saturated soils near headwaters. Stands are
distributed in Slovenia, Croatia and the mountainous parts
of southern Serbia. The predominant species are Scirpus sylvaticus,
Filipendula ulmaria, Juncus effusus and Eriophorum
latifolium (the last species is characteristic for specific stands
in Serbia).
Deschampsion cespitosae
Meadows on alluvial lowlands that are flooded from
rainwater and less from water courses. Soils have impermeable
layers and water stagnates for some period.
These meadows occur in Slovenia and Croatia, with one
location in SE Serbia. Meadows are dominated by Deschampsia
cespitosa and Agrostis canina agg., other characteristic
species are Succisella inflexa, Ranunculus flammula
and Gratiola officinalis.
Molinio-Hordeion secalini
Stands of this alliance are typical grasslands from the
humid part of the sub-Mediterranean region, mostly on
periodically flooded Karst poljes and locally on some
islands. Their distribution stretches from Slovenia through
Croatia to Bosnia and Herzegovina. Characteristic species
are Scilla litardiei, Deschampsia media, Peucedanum coriaceum
and Plantago altissima.
Trifolion resupinati
The alliance includes lowland wet grasslands of the subMediterranean
part of the central Balkans (southern Serbia
in the study area, with the centre of distribution in
Macedonia). The climate is extreme and sites dry out in
summer. The main species are Poa sylvicola, Trifolium
599
Applied Vegetation Science
Doi:10.1111/avsc.12094 © 2014 International Association for Vegetation Science
U. Silc et al. Grassland vegetation of the NW Balkans
resupinatum and several other species of the genus Trifolium
(T. balansae, T. nigrescens, T. fragiferum).
Trifolion pallidi
The alliance consists of wet flooded meadows with less
drought in summer than Trifolion resupinati, and colder
winters than Molinio-Hordeion secalini. The centre of distribution
of the alliance is eastern Croatia and western Serbia.
Characteristic species are Medicago arabica, Trifolium
patens, Festuca pratensis agg. and Trifolium pallidum.
Trifolio-Ranunculion pedati
The alliance includes grasslands of a mesic sub-halophytic
character. The alliance is restricted to the Pannonian plain.
Characteristic species are Poa angustifolia, P. palustris and
Trifolium angulatum and accompanied by the halophytes
Aster canus, Limonium gmelinii and Podospermum canum.
Cynosurion cristati
Intensively managed grasslands (pastures) on fertile soils
found in the Central European floristic region of the NW
Balkans (Slovenia, Croatia and Vojvodina). The main
species are Lolium perenne, Trifolium repens, Cynodon dactylon
and Trifolium fragiferum.
Arrhenatherion elatioris
The alliance includes mesic non-flooded meadows, regularly
mown and often fertilized. It is most common in
Central Europe; in the study area it is common in Slovenia
and Croatia, while in southern parts these grasslands
are also found in the mountain belt. Diagnostic species
are Arrhenatherum elatius, Dactylis glomerata, Galium mollugo
agg. and Pastinaca sativa.
Alchemillo-Ranunculion repentis
This vegetation type includes moderately trampled
mesophilous communities often in shaded sites. The
alliance is only found in the northern part of the study
area (Slovenia and Croatia). The characteristic species
are Prunella vulgaris, species characteristic of trampled
habitats and the neophytes Juncus tenuis and Duchesnea
indica.
Potentillion anserinae
The alliance comprises stands along water bodies with
alternating wet and dry phase. They are distributed in the
northern part of NW Balkans and on the Pannonian
lowlands. Stands are dominated by Agrostis stolonifera agg.,
Mentha longifolia and Juncus inflexus.
Triseto flavescentis-Polygonion bistortae
The alliance comprises mostly pastures from submontane
and montane belts (up to 1500 mÁa.s.l.) from northern
Slovenia. Stands are dominated by Trisetum flavescens, other
diagnostic species are Astrantia major, Bromus erectus and
Rhinanthus alectorolophus.
Pancicion serbicae
These are mesophilous grasslands of the upland and subalpine
belts (1100–1800 m) in mountain ranges of Bosnia
and Herzegovina, Kosovo, southern Serbia and also Montenegro.
The characteristic species of this endemic Dinaric
alliance are Anemone narcissiflora, Crepis aurea var. bosniaca,
Pimpinella serbica and Phleum alpinum.
Poion supinae
These are meadows that are mown once a year and then
grazed, in the mountain belt (800–1300 mÁa.s.l.). They are
found in Croatia and probably also in Slovenia. Stands are
dominated by Trisetum flavescens accompanied by Alchemilla
xanthochlora and species of the Arrhenatherion elatioris
alliance.
Acknowledgements
We are grateful to Ladislav Mucina who allowed us to use
the unpublished EuroVegChecklist. We thank Iztok Sajko
for producing the maps and for data extraction in GIS, and
Andrej Rozman for help with graphs. Martin Cregeen
kindly checked our English. We thank Milan Chytry, Michal
Hajek and two anonymous reviewers, whose comments
improved the manuscript. This project was funded
by the ARRS grant P1-0236, Ministry of Education and Science
of Serbia, grant 31057 and bilateral grants BI-RS/12-
13-035 and BI-HR/12-13-010.
References
Acic, S., Silc, U., Lakusic, D., Vukojicic, S. & Dajic Stevanovic, Z.
2013a. Typification and correction of syntaxa from the class
Molinio-Arrhenatheretea Tx. 1937 in Serbia. Hacquetia 12:
39–54.
Acic, S., Silc, U., Vrbnicanin, S., Cupac, S., Topisirovic, G., Stavretovic,
N. & Dajic-Stevanovic, Z. 2013b. Grassland communities
of Stol mountain (eastern Serbia): vegetation and
environmental relationships. Archives of Biological Sciences 65:
211–227.
Applied Vegetation Science
600 Doi:10.1111/avsc.12094 © 2014 International Association for Vegetation Science
Grassland vegetation of the NW Balkans U. Silc et al.
Apostolova, I., Meshinev, T. & Petrova, A.S. 2007. Hay meadows
with Trisetum flavescens in Bulgaria: syntaxonomy and implications
for nature conservation. Phytologia Balcanica 13: 401–
414.
Bertic, I. 1987. Veliki geografski atlas Jugoslavije. Sveucilisna naklada
Liber, Zagreb, HR.
Blasi, C., Tilia, A., Rosati, L., Del Vico, E., Copiz, R., Ciaschetti, G.
& Burrascano, S. 2012. Geographical and ecological differentiation
in Italian mesophilous pastures referred to the alliance
Cynosurion cristati Tx. 1947. Phytocoenologia 41: 217–229.
Borhidi, A., Kevey, B. & Lendvai, G. 2012. Plant communities of
Hungary. Akademiai Kiado, Budapest, HU.
Botta-Dukat, Z., Chytry, M., Hajkova, P. & Havlova, M. 2005.
Vegetation of lowland wet meadows along a climatic continentallity
gradient in Central Europe. Preslia 77: 89–111.
Carni, A., Kosir, P., Karadzic, B., Matevski, V., Redzic, S. &
Skvorc, Z. 2009. Thermophilous deciduous forests in Southeastern
Europe. Plant Biosystems 143: 1–13.
Ceulemans, T., Merckx, R., Hens, M. & Honnay, O. 2013. Plant
species loss from European semi-natural grasslands following
nutrient enrichment – is it nitrogen or is it phosphorus? Global
Ecology and Biogeography 22: 73–82.
Chytry, M. (ed.) 2007. Vegetace Ceske republiky 1. Travinna a
kerıckova vegetace. Academia, Praha, CZ.
Dajic Stevanovic, Z., Peeters, A., Vrbnicanin, S., Sostaric, I. &
Acic, S. 2008. Long term grassland vegetation changes: case
study Nature Park Stara Planina (Serbia). Community Ecology
9: 23–31.
Dajic-Stevanovic, Z., Lazarevic, D., Petrovic, M., Acic, S. & Tomovic,
G. 2010. Biodiversity of natural grasslands of Serbia:
state and prospects of utilization. In: XII International Symposium
on Forage Crops of Republic of Serbia “Forage Crops Basis of
the Sustainable Animal Husbandry Development, pp. 235–247.
Krusevac, RS.
Dierschke, H. & Briemle, G. 2002. Kulturgrasland. Ulmer, Stuttgart,
DE.
Drobnik, J., R€omermann, C., Bernhardt-R€omermann, M. & Poschlod,
P. 2011. Adaptation of plant functional group composition
to management changes in calcareous grassland.
Agriculture, Ecosystems & Environment 145: 29–37.
Elias, P. Jr, Sopotlieva, D., Dıte, D., Hajkova, P., Apostolova, I.,
Senko, D., Meleckova, Z. & Hajek, M. 2013. Vegetation
diversity of salt-rich grasslands in Southeast Europe. Applied
Vegetation Science 16: 521–537.
Ellmauer, T. & Mucina, L. 1993. Molinio-Arrhenatheretea. In: Mucina,
L., Grabherr, G. & Ellmauer, T. (eds.) Pflanzengesellschaften
€Osterreichs, Teil I. Anthropogene Vegetation, pp. 197–401.
Gustav Fischer, Jena, DE.
European Commission1992/1995. Council Directive 92/43/EEC
of 21 May 1992 on the conservation of natural habitats and of
wild fauna andflora.ECOfficial JournalL I.1.1,Brussels.
Hajek, M., Hajkova, P., Sopotlieva, D., Apostolova, I. & Velev, N.
2008. The Balkan wet grassland vegetation: a prerequisite to
better understanding of European habitat diversity. Plant
Ecology 195: 197–213.
Hajkova, P., Hajek, M., Blazkova, D., Kucera, T., Chytry, M.,
Havlova, M., Sumberova, K., Cerny, T., Novak, J. & Simonova,
D. 2007. Louky a mezofilnı pastviny (Molinio-Arrhenatheretea).
In: Chytry, M. (ed.) Vegetace Ceske republiky, 1.
Travinna a kerıckova vegetace, pp. 165–280. Academia,
Praha, CZ.
H€ardtle, W., Redecker, B., Assmann, T. & Meyer, H. 2006. Vegetation
responses to environmental conditions in floodplain
grasslands: prerequisites for preserving plant species diversity.
Basic and Applied Ecology 7: 280–288.
Havlova, M., Chytry, M. & Tichy, L. 2004. Diversity of hay
meadows in the Czech Republic: major types and environmental
gradients. Phytocoenologia 34: 551–567.
Heged€usova, K., Ruzickova, H., Senko, D. & Zuccarini, P. 2011.
Plant communities of the montane mesophilous grasslands
(Polygono bistortae-Trisetion flavescentis alliance) in central Europe:
formalized classification and syntaxonomical revision.
Plant Biosystems 145: 1–16.
Hennekens, S.M. & Schaminee, J.H.J. 2001. TURBOVEG, a comprehensive
data base management system for vegetation
data. Journal of Vegetation Science 12: 589–591.
Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P.G. & Jarvis, A.
2005. Very high resolution interpolated climate surfaces for
global land areas. International Journal of Climatology 25:
1965–1978.
Horvat, I., Glavac, V. & Ellenberg, H. 1974. Vegetation S€udosteuropas.
Gustav Fischer, Stuttgart, DE.
Horvatic, S. 1930. Soziologische Einheiten der Niederungswiesen
in Kroatien und Slavonien. Acta Botanica Croatica 5:
57–118.
Horvatic, S. 1939. Splosna primerjava vegetacije nizinskih travnikov
Slovenije z ono Hrvatske in Slavonije. Zbornik Prirodoslovnega
drustva 1: 40–43.
Ilijanic, L. 1969. Das Trifolion pallidi, ein neuer Verband der
Ordnung Trifolio-Hordeetalia H-ic. Acta Botanica Croatica 28:
151–159.
Ilijanic, L. 1973. Allgemeiner €Uberblick €uber die Wechselfeuchten
Niederungswiesen Jugoslawiens im Zusammenhang mit
den klimatischen Verh€altnissen. Acta Botanica Academiae Scientiarum
Hungaricae 19: 165–179.
Jovanovic-Dunjic, R., Stefanovic, K., Popovic, R. & Dimitrijevic,
J. 1986. Prilog poznavanju livadskih ekosistema na podrucju
Velikog Jastrebca. Glasnik Instituta za botaniku i Botanicke baste
Univerziteta u Beogradu 20: 7–31.
Klimek, S., Kemmermann, A.R., Hofmann, M. & Isselstein, J.
2007. Plant species richness and composition in managed
grasslands: the relative importance of field management
and environmental factors. Biological Conservation 134: 559–
570.
Kojic, M., Popovic, R. & Karadzic, B. 1998. Sintaksonomski pregled
vegetacije Srbije. Inst. biol. istraz. Sinisa Stankovic, Beograd,
RS.
Kojic, M., Mrfat-Vukelic, S., Dajic, Z. & Dordevic-Milosevic, S.
2004. Livade i pasnjaci Srbije. Institut za istrazivanja u poljoprivredi
Srbija, Beograd, RS.
601
Applied Vegetation Science
Doi:10.1111/avsc.12094 © 2014 International Association for Vegetation Science
U. Silc et al. Grassland vegetation of the NW Balkans
Kojic, M., Mrfat-Vukelic, S. & Dordevic-Milosevic, S. 2005. Osnovne
fitocenoloske i ekonomske karakteristike prirodnih
travnjaka i pasnjaka Srbije. Biotechnology in Animal Husbandry
21: 187–191.
Kosir, P., Carni, A. & Di Pietro, R. 2008. Classification and phytogeographical
differentiation of broad-leaved ravine forests in
southeastern Europe. Journal of Vegetation Science 19: 331–384.
Kosir, P., Casavecchia, S., Carni, A., Skvorc, Z., Zivkovic, L. & Biondi,
E. 2013. Ecological and phytogeographical differentiation
of oak–hornbeam forests in southeastern Europe. Plant
Biosystems 147: 84–98.
Lakusic, R. 1966. Vegetation of grasslands and pastures at the
Bjelasica Mt. Godisnjak Bioloskog instituta Univerziteta Sarajevo
19: 25–186.
Laliberte, E., Wells, J.A., DeClerck, F., Metcalfe, D.J., Catterall,
C.P., Queiroz, C., Aubin, I., Bonser, S.P., Yi, D., (…) & Mayfield,
M.M. 2010. Land-use intensification reduces functional
redundancy and response diversity in plant
communities. Ecology Letters 13: 76–86.
Lengyel, A., Chytry, M. & Tichy, L. 2011. Heterogeneity-constrained
random resampling of phytosociological databases.
Journal of Vegetation Science 22: 175–183.
Lengyel, A., Purger, D. & Csiky, J. 2012. Classification of mesic
grasslands and their transitions of South Transdanubia (Hungary).
Acta Botanica Croatica 71: 31–50.
Marinsek, A., Silc, U. & Carni, A. 2013. Geographical and ecological
differentiation of Fagus forest vegetation in SE Europe.
Applied Vegetation Science 16: 131–147.
Milosavljevic, N.V., Randelovic, N.V., Zlatkovic, B. & Randelovic,
V.N. 2008. Phytocenologic diversity of Krajiste in
southeastern Serbia. Natura Montenergrina 7: 193–204.
Moog, D., Poschlod, P., Kahmen, S. & Schreiber, K.F. 2002.
Comparison of species composition between different grassland
management treatments after 25 years. Applied Vegetation
Science 5: 99–106.
Mucina, L., B€ultman, H., Dierssen, K., Theurillat, J.-P., Dengler,
J., Carni, A., Sumberova, K., Raus, T., Di Pietro, R., (…) &
Tichy, L. 2014. Vegetation of Europe: hierarchical floristic
classification system of plant, lichen, and algal communities.
Applied Vegetation Science. in review.
Obratov, D. & Dukic, M. 1998. Plant communities with endemic
species Pancicia serbica Vis. (Apiaceae, Araliales) in Yugoslavia.
In: Tsekos, I. & Moustakas, M. (eds.) Progress in Botanical
Research, pp. 97–100. Kluwer Academic, Dordrecht, NL.
Parabucski, S. 1990. Neke odlike dolinskih livada na podrucju
Vojvodine. Zbornik Matice Srpske za prirodne nauke 79: 107–
118.
Pignatti, S. 2005. Valori di bioindicazione delle piante vascolari
della flora d‘Italia. Braun-Blanquetia 39: 1–97.
Polunin, O. 1987. Flowers of Greece and the Balkans: A field guide.
Oxford University Press Oxford, UK.
Prach, K. 2008. Vegetation changes in a wet meadow complex
during the past half century. Folia Geobotanica 43: 119–130.
Putfarken, D., Dengler, J., Lehmann, S. & H€ardtle, W. 2008. Site
use of grazing cattle and sheep in a large-scale pasture landscape:
a GPS/GIS assessment. Applied Animal Behaviour Science
111: 54–67.
Pyk€al€a, J., Luoto, M., Heikkinen, R.K. & Kontula, T. 2005. Plant
species richness and persistence of rare plants in abandonment
semi-natural grasslands in northern Europe. Basic and
Applied Ecology 6: 25–33.
Randelovic, V. & Zlatkovic, B. 2010. Flora i vegetacija Vlasinske visoravni.
Prirodno-matematicki fakultet, Nis, RS.
Reed, J.M., Krystufek, B. & Eastwood, W.J. 2004. The physical
geography of the Balkans and nomenclature of place names.
In: Griffiths, H.I., Krystufek, B. & Reed, J.M. (eds.) Balkan
biodiversity. Pattern and process in the European hotspot, pp. 9–
22. Kluwer, Dordrecht, NL.
Ruzickova, H., Banasova, V. & Kalivoda, H. 2004. Morava River
alluvial meadows on the Slovak-Austrian border (Slovak
part): plant community dynamics, floristic and butterfly
diversity – threats and management. Journal for Nature Conservation
12: 157–169.
Sebastia, M.T. 2004. Role of topography and soils in grassland
structuring at the landscape and community scales. Basic and
Applied Ecology 5: 331–346.
Silc, U. & Carni, A. 2012. Conspectus of vegetation syntaxa in
Slovenia. Hacquetia 11: 113–164.
Silc, U., Vrbnicanin, S., Bozic, D., Carni, A. & Dajic Stevanovic,
Z. 2008. Phytosociological alliances in the vegetation of arable
fields in the northwestern Balkan Peninsula. Phytocoenologia
38: 241–254.
Stancic, Z. 2008. Classification of mesic and wet grasslands in
northwest Croatia. Biologia 63: 1089–1103.
Stevanovic, V., Tan, K. & Iatrou, G. 2003. Distribution of the
endemic Balkan flora on serpentine I. –obligate serpentine
endemics. Plant Systematics and Evolution 242: 149–170.
Stevanovic, V., Tan, K. & Petrova, A. 2007. Mapping the endemic
flora of the Balkans – a progress report. Bocconea 21: 131–137.
Stevens, C.J., Dise, N.B., Mountford, J.O. & Gowing, D.J. 2004.
Impact of nitrogen deposition on the species richness of
grasslands. Science 303: 1876–1879.
Tichy, L. & Chytry, M. 2006. Statistical determination of diagnostic
species for site groups of unequal size. Journal of Vegetation
Science 17: 809–818.
Tichy, L., Chytry, M., Hajek, M., Talbot, S.S. & Botta-Dukat, Z.
2010. OptimClass: using species-to-cluster fidelity to determine
the optimal partition in classification of ecological communities.
Journal of Vegetation Science 21: 287–299.
Tomovic, G. 2007. Fitogeografska pripadnost, distribucija i centri
diverziteta Balkanske endemicne flore u Srbiji. Doktorska disertacija.
Bioloski fakultet, Beograd, RS.
Trinajstic, I. 2008. Biljne zajednice Republike Hrvatske. Akademija
sumarskih znanosti, Zagreb, HR.
Velev, N., Apostolova, I. & Fajmonova, Z. 2011a. Cynosurus
cristatus grasslands in West Bulgaria. Phytologia Balcanica 17:
221–236.
Velev, N., Apostolova, I. & Rozbrojova, Z. 2011b. Alliance Arrhenatherion
elatioris in West Bulgaria. Phytologia Balcanica 17:
67–78.
Applied Vegetation Science
602 Doi:10.1111/avsc.12094 © 2014 International Association for Vegetation Science
Grassland vegetation of the NW Balkans U. Silc et al.
Waldhardt, R. & Otte, A. 2003. Indicators of plant species and
community diversity in grasslands. Agriculture, Ecosystems and
Environment 98: 339–351.
Wellstein, C., Otte, A. & Waldhardt, R. 2007. Impact of site and
management on the diversity of central European mesic
grassland. Agriculture, Ecosystems and Environment 122: 203–
210.
Wesche, K., Krause, B., Culmsee, H. & Leuschner, C. 2012. Fifty
years of change in Central European grassland vegetation:
large losses in species richness and animal-pollinated plants.
Biological Conservation 150: 76–85.
Wilson, J.B., Peet, R.K., Dengler, J. & P€artel, M. 2012. Plant species
richness: the world records. Journal of Vegetation Science
23: 796–802.
Zelnik, I. 2005. Meadows of the order Molinietalia caeruleae
Koch 1926 in south-eastern Slovenia. Fitosociologia 42: 3–32.
Zelnik, I. & Carni, A. 2008. Wet meadows of the alliance Molinion
and their environmental gradients in Slovenia. Biologia 63:
187–196.
Zelnik, I. & Carni, A. 2013. Plant diversity and composition of
wet grasslands in relation to environmental factors. Biodiversity
Conservation 22: 2179–2192.
Zuidhoff, A.C., Rodwell, J.S. & Schaminee, J.H.J. 1995. The Cynosurion
cristati Tx. 1947 of central, southern and western
Europe: a tentative overview, based on the analysis of individual
releves. Annali di botanica 53: 25–48.
Supporting Information
Additional supporting information may be found in the
online version of this article:
Appendix S1. Map of releve distribution and precipitation
of the warmest quarter as the climatic variable.
Appendix S2. Syntaxonomic checklist of the MolinioArrhenatheretea
class in the NW Balkans.
Appendix S3. Full synoptic table in pdf and csv format.
603
Applied Vegetation Science
Doi:10.1111/avsc.12094 © 2014 International Association for Vegetation Science
U. Silc et al. Grassland vegetation of the NW Balkans