Dry grasslands and thorn-cushion communities of Armenia: a first syntaxonomic classification

Dry grasslands and thorn-cushion communities of Armenia: a first syntaxonomic classification Denys Vynokurov1,2, Alla Aleksanyan3, Thomas Becker4, Idoia Biurrun1, Dariia Borovyk2,5, George Fayvush3, Itziar García-Mijangos1, Martin Magnes6, Salza Palpurina7, Ute Becker8, Asun Berastegi9, Beata Cykowska-Marzencka10, Iwona Dembicz11, Dieter Frank12, Andreas Hilpold13, Philipp Kirschner14, Helmut Mayrhofer6, Marine Oganesian15, Iuliia Vasheniak2, Jürgen Dengler16,17 1 Department of Plant Biology and Ecology, University of the Basque Country UPV/EHU, Bilbao, Spain 2 Department of Geobotany and Ecology, M.G. Kholodny Institute of Botany National Academy of Sciences of Ukraine, Kyiv, Ukraine 3 Department of Geobotany and Ecophysiology, Institute of Botany aft. A. Takhtajyan National Academy of Sciences, Yerevan, Armenia 4 Department of Geobotany, Faculty of Regional and Environmental Sciences, University of Trier, Trier, Germany 5 Department of Botany and Zoology, Faculty of Science, Masaryk University, Brno, Czech Republic 6 Department of Plant Sciences, Institute of Biology, University of Graz, Graz, Austria 7 National Museum of Natural History, Bulgarian Academy of Sciences, Sofia, Bulgaria 8 Botanic Garden of the Johannes Gutenberg-University, Mainz, Germany 9 Biodiversity Department, Environmental Management of Navarre, Pamplona-Iruña, Nafarroa, Spain 10 Wädenswil, Switzerland 11 Institute of Environmental Biology, Faculty of Biology, University of Warsaw, Warsaw, Poland 12 Environmental Protection Agency Saxony-Anhalt (LAU), Halle, Germany 13 Institute for Alpine Environment, Eurac Research, Bolzano/Bozen, Italy 14 Department of Botany, University of Innsbruck, Innsbruck, Austria 15 Department of Higher Plants Taxonomy and Geography, Institute of Botany after A. Takhtajyan of the National Academy of Sciences of Armenia, Yerevan, Armenia 16 Vegetation Ecology Research Group, Institute of Natural Resource Sciences (IUNR), Zurich University of Applied Sciences (ZHAW), Wädenswil, Switzerland 17 Plant Ecology, Bayreuth Center of Ecology and Environmental Research (BayCEER), University of Bayreuth, Bayreuth, Germany Corresponding author: Jürgen Dengler ([email protected]) Academic editor: Victor Chepinoga♦Linguistic editor: Michael Glaser Received 22 January 2024♦Accepted 15 March 2024♦Published 9 May 2024 Abstract Aim: To provide the first syntaxonomic, plot-based classification of the dry grasslands and thorn-cushion communities in Armenia. Study area: Armenia. Methods: We sampled 111 vegetation plots (10 m2) and recorded environmental and structural parameters. We collected additional 487 relevés from surrounding countries for a broad-scale comparison. We used modified TWINSPAN to derive a syntaxonomic classification system, whose units were then compared among each other regarding species composition, structure, site conditions and distribution. Results: The classification of Armenian vegetation plots resulted in a 12-cluster solution. Unsupervised classification of the broad-scale dataset yielded five main groups, which were used for the high-level syntaxonomic assignments of the Armenian data. We assigned about half of the plots of the Armenian dataset to the Festuco-Brometea, while the remaining represented a potential new class, preliminarily called “Ziziphora tenuior-Stipa arabica grasslands”. Most of the syntaxa below class level are new to science, therefore we provide formal descriptions of three orders (Plantagini atratae-Bromopsietalia variegatae, Onobrychido transcaucasicae-Stipetalia pulcherrimae, Cousinio brachypterae-Stipetalia arabicae), four alliances (Acantholimono caryophyllacei-Stipion holosericeae, Artemision fragrantis, Onobrychido michauxii-Stipion capillatae, Onobrychido transcaucasicae-Stipion pulcherrimae) and six associations. We found significant differences in the topographic, climatic and soil characteristics, and structural parameters, species life forms and distribution range types between the grassland types at different syntaxonomic levels. The mean species richness was 47.3 (vascular plants: 46.8, bryophytes: 0.4, lichens: 0.1). Conclusions: We found remarkable differences of the Armenian dry grasslands from the previously known units Copyright Denys Vynokurov et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC-BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Vegetation Classification and Survey 5: 39–73 doi: 10.3897/VCS.119253 RESEARCH PAPER International Association for Vegetation Science (IAVS) GRASSLANDS OF ASIA Denys Vynokurov et al.: Dry grasslands of Armenia40 and described most of the higher syntaxa and all the associations as new to science. Our study provides arguments for a potential new class of Ziziphora tenuior-Stipa arabica grasslands separate both from the Euro-Siberian Festuco-Brometea and the Anatolian Astragalo-Brometea. Finally, we found plot scale richness of vascular plants clearly above the Palaearctic average of dry grasslands and that of non-vascular plants clearly below, which calls for further biodiversity analyses. Taxonomic reference: Euro+Med (2023) for vascular plants, Hodgetts et al. (2020) for bryophytes, Nimis et al. (2018) for lichens except for Xanthoparmelia camtschadalis (Ach.) Hale. Abbreviations: EDGG = Eurasian Dry Grassland Group; DCA = detrended correspondence analysis; ICPN = International Code of Phytosociological Nomenclature (Theurillat et al. 2021); TWINSPAN = two-way indicator species analysis. Keywords Armenia, classification, dry grassland, Festuco-Brometea, Irano-Turanian, mountain steppe, Onobrychido transcaucasicae-Stipetalia pulcherrimae, Plantagini atratae-Bromopsietalia variegatae, species richness, syntaxonomy, thorn-cushion community, Cousinio brachypterae-Stipetalia arabicae Introduction Armenia is a land-lock country located in the southern part of the Lesser Caucasus, belonging to two global biodiversity hotspots: the Caucasian and the Irano-Anatolian (Mittermeier et al. 2004). With 3,800 vascular plant species, including 144 country endemics, it has an outstandingly rich vascular plant flora for a territory of less than 30,000 km2 (Biodiversity and Landscape Conservation Union 2014). The specificity of the vegetation of Armenia is caused by the fact that the country is located on the border between the Euro-Siberian (or Circumboreal) and Irano-Turanian biogeographical regions (Takhtajan 1986; Manafzadeh et al. 2017; Loidi et al. 2022), which are belonging to different subkingdoms: Boreal and Ancient Mediterranean (Takhtajan 1986). Grasslands and shrublands in the Middle East and the Caucasus areas are of great interest as they host a high biodiversity of species and habitats (Ambarlı et al. 2020), and have been shown to be the area of origin of important elements of Western Asian and European dry grasslands (e.g. Euphorbia seguieriana, Frajman et al. 2019; Euphorbia nicaeensis group, Stojilkovič et al. 2022). Grasslands are linked to human activities, so can be heavily impacted by changes in land use that are leading to the loss and disturbance of these habitats, with the consequent loss of biodiversity (Török and Dengler 2018). In this area, unregulated (unsustainable) grazing, conversion to cropland, afforestation with non-native tree species, and mining and energy production projects are the major threats (Ambarlı et al. 2020). The knowledge of these habitats in Armenia is crucial for understanding their biodiversity and actual distribution, which will make it possible to establish the necessary management measures for conservation (Ambarlı et al. 2018). Due to the abovementioned importance of the typology of habitats and vegetation, there is a growing international consensus on the need for coherent vegetation classification systems based on the analysis of vegetation-plot data (De Cáceres et al. 2015). There are different plot-based vegetation classification approaches (De Cáceres et al. 2018), the most important globally being the phytosociological approach (Dengler et al. 2008) and the EcoVeg approach (Faber-Langendoen et al. 2014). The latter, which is applied by the International Vegetation Classification (IVC, Faber-Langendoen et al. 2020) formally incorporates two levels above the class level, which is the highest formal level in the phytosociological approach. The formation is based on physiognomic-structural features of the vegetation and thus it is very useful to define biomes. Recently Willner and Faber-Langendoen (2021) made a first attempt to link the European classification system, based on the phytosociological approach (EVC, Mucina et al. 2016), with the International Vegetation Classification. Up to date, there is no formalised plot-based classification system for Armenia, which was a Soviet Socialist Republic until 1991. As the Braun-Blanquet approach to vegetation classification (Braun-Blanquet 1964; Guarino et al. 2018) developed in Central and Western Europe, it did not play a role in the USSR for political reasons (Masing 1991). Hierarchical classifications based on syntaxa were virtually non-existent in Armenia, at least not for grasslands, as vegetation scientists followed the so-called dominance approach, which categorised vegetation according to dominant and ecologically significant species (Sukachev 1928). This approach does not usually produce complete species lists with abundance data for small sample areas of defined size, i.e. it does not produce relevés or vegetation plots. As a result, the first classification attempt of grassland habitats in Armenia identified three vegetation types (Grossgeim 1928): semi-deserts, mountain-xerophilous vegetation and mountain steppes. Afterwards, different classifications were developed for dry grasslands and steppe vegetation. First, Makagian (1941) defined four types of steppes and steppe-like vegetation (stony semi-deserts, steppes, meadow-steppes and highland xerophytes). Later, Ziroyan (1989) categorised natural dry grasslands in Armenia into five vegetation types related to zonal and altitudinal distribution (deserts, semi-deserts, highland xerophytes, Vegetation Classification and Survey 41 mountain steppes and mountain meadow steppes). Finally, Fayvush (1992) recognised four steppe subtypes (true steppes, thorny-cushion steppes, shrubby steppes and meadow steppes) with 12 classes of formations. The lack of a unified framework of grassland typology impedes acquiring knowledge about their distribution and diversity, which is harmful to their conservation. Although the phytosociological approach has not been applied to the survey of Armenian grasslands, those of neighbouring countries of the Southern Caucasus have been at least fragmentarily studied in Transcaucasia (Azerbaijan: Peper et al. 2010; Etzold et al. 2016; Jabbarov et al. 2020; Georgia: Pyšek and Šrůtek 1989; Nakhutsrishvili et al. 2022, etc.), Eastern Anatolia in Turkey (Çetik and Tatlı 1975; Tatlı 1991; Gümüs 1992; Gümüs et al. 2003; Hamzaoğlu 2006, etc.) and North Iran (Klein 1982, 1987; Klein and Lacoste 1994; Noroozi et al. 2010, 2014, 2017, etc.). Many of the data used for these phytosociological surveys are stored in vegetation-plot databases. In the last decades small regional and/or personal databases have been compiled in large vegetation plots databases. The European Vegetation Archive (EVA, Chytrý et al. 2016) was the pioneer, followed by the global vegetation database (sPlot, Bruelheide et al. 2019) and by GrassPlot (Dengler et al. 2018). These databases are the basis for large-scale vegetation classification studies (Novák et al. 2023b; Peterka et al. 2023), but they also permit macroecological studies to great scales leading to a broad understanding of the distribution and diversity patterns (Graco-Roza et al. 2022; Sabatini et al. 2022; Večeřa et al. 2023). Especially for grasslands, the GrassPlot database hosts high-quality data from the Palaearctic realm, sampled on precisely delimited plots, including vascular plants and cryptogam data (Dengler et al. 2018; Biurrun et al. 2019). At the regional scale, the Transcaucasian Vegetation Database, a phytosociological database of the Southern Caucasus, was recently established (Novák et al. 2023a). However, it currently lacks vegetation plots of dry grassland and thorn-cushion communities in Armenia. It is therefore both an opportunity and a challenge to record vegetation plots and apply the Braun-Blanquet classification approach to Armenia – as its vegetation, to our knowledge, has never been studied according to this approach. Therefore, the Eurasian Dry Grassland Group (EDGG; www.edgg.org) conducted an international research expedition (called “Field Workshop”) in Armenia to collect standardised, high-quality vegetation-plot data from dry grasslands and thorn-cushion communities throughout the country. In principle the EDGG Field Workshops aim at collecting such data for regional studies on biodiversity patterns (Kuzemko et al. 2016; Dembicz et al. 2021b; Bergauer et al. 2022) and phytosociology (García-Mijangos et al. 2021; Magnes et al. 2021). Moreover, these data are provided to the GrassPlot database (Dengler et al. 2018) for biodiversity studies across all Palaearctic grasslands and other open habitats (Biurrun et al. 2021; Dembicz et al. 2021a; Zhang et al. 2021) as well as to the European Vegetation Archive (EVA; Chytrý et al. 2016) and the global vegetation-plot database sPlot (Bruelheide et al. 2019) to fill important data gaps in continental and global studies of biodiversity, global change and syntaxonomy. In this paper, we used the plot data sampled during the Field Workshop to provide the first syntaxonomic classification scheme of the dry grasslands and thorn-cushion communities of the country, using numerical methods of unsupervised classification and determination of diagnostic species. Specifically, we asked: (1) Which association-rank communities can be distinguished, and to which higher-rank syntaxa do they belong? (2) How do the syntaxa in Armenia compare to those in the neighbouring countries of Western Asia? (3) How are the syntaxa in Armenia differentiated from each other in terms of species composition, species richness, structure, site conditions and distribution? Study area Physiogeography, climate, soils, geology Armenia is a South Caucasian republic, bordering Georgia, Azerbaijan, Turkey, and Iran. It is a landlocked country with a total area of 29,740 km2, at about 145 km from the Black Sea and 175 km from the Caspian Sea. It lies between 38°50' and 41°18' northern latitude and between 43°27' and 46°37' eastern longitude, and measures 400 km along its main axis (north-west to south-east). Armenia is generally a mountainous country, having its lowest point at 375 m a.s.l. and culminating at 4,095 m a.s.l. in the Aragats, with an average elevation of 1,850 m a.s.l. The location of Armenia at the intersection of two phytogeographical subkingdoms (Boreal and Ancient Mediterranean), together with the diversity in climatic conditions and the active geological processes, have resulted in the formation of diverse ecosystems and high biodiversity with a high level of endemism (Fayvush and Aleksanyan 2016). On the small territory of the country, there are about 3,800 species of vascular plants, 497 species of soil and water algae, 433 species of bryophytes, 4,577 species of non-lichenized fungi, 619 species of lichens, 567 species of vertebrates and about 17,000 species of invertebrates (Fayvush 2023). A wide range of climatic zones are distinguished within Armenia, which experiences large climatic contrasts because of its intricate terrain and the big climatic gradients (Ministry of Nature Protection of the Republic of Armenia 2015). The basic climate types mainly follow the elevational gradient, from dry subtropical up to severe alpine. The average annual temperature ranges from -8°C in high-altitude mountainous regions (2,500 m a.s.l. and higher) to 12–14°C in low-traced valleys. The overall climate is best characterised as dry continental, in some areas with an Denys Vynokurov et al.: Dry grasslands of Armenia42 annual rhythm like the Mediterranean climate regime. The average annual precipitation in Armenia is 592 mm. The most arid regions are the Ararat valley and the region of Meghri with annual precipitation of 200–250mm. The highest annual precipitation of 800–1000 mm is observed in high-altitude mountain regions. From the orographical and physico-geographical points of view, Armenia forms the northern edge of the system of folded-block mountains of the Armenian Highland. Unlike the Greater Caucasus, Armenia and the Lesser Caucasus are not a single, distinct watershed ridge. It is a system of coulisse-spaced ridges that merge with the mountain formations of the inner parts of the Armenian Highland and adjacent high areas (Aslanyan 1958, 1985). Four main geomorphological regions can be recognized within Armenia. (1) Mountain ridges and valleys in the north-east of the country which bear witness of extensive erosion. (2) Areas covered by lava of relatively recent (upper Pliocene) origin within Asia Minor are characterised by gentle slopes with little evidence of erosion but, in which larger rivers have carved out deep gorges and canyons. (3) A series of ridged mountains in the south of Armenia, which constitute the Lesser Caucasus system and show intense erosion. (4) The Ararat Valley represents the lowest part of the Ararat depression covered with alluvial and proluvial sediments (Aslanyan 1958; Gabrielyan 1962; Dumitrashko 1979). In our study we tried to cover as much of the country’s dry grassland diversity as possible within 11 days, with a focus on the northwestern and central parts (Figure 1). In total, we sampled in five of the 11 administrative provinces of the country (Aragatsotn, Ararat, Gegharkunik, Shirak, Vayots Dzor) and seven of the 12 floristic regions. We covered an elevational gradient from 1,338 to 2,350 m a.s.l. National typologies of Armenian dry grasslands So far, the syntaxonomy of grassland or thorn-cushion vegetation of Armenia hasn’t been developed yet. The only existing vegetation typologies are based on the dominance approach. The first overview of the Armenian vegetation types was performed by Grossgeim (1928). He distinguished eight main types of vegetation: (1) aquatic and bog vegetation; (2) forest vegetation; (3) semi-deserts; (4) mountain-xerophilous vegetation (mountain semi-desert); (5) mountain steppes; (6) meadow-like vegetation; (7) solonetzs and solonchaks; (8) weeds. In the semi-desert type, he selected three subtypes: (a) alkaline-loamy semi-desert on the alluvium of the Aras River; (b) wormwood semi-desert on eruptive rocks; (c) sandy semi-desert. Takhtajan (1941) explored the phytogeographic patterns of Armenia, including the division of the region into phytogeographic provinces and districts, the historical Figure 1. Map of Armenia indicating zonal vegetation types based on Grossgeim (1928). Black dots indicate the sampling localities. The light blue area refers to the Lake Sevan. Vegetation Classification and Survey 43 development of its vegetation, and the classification of vegetation into broad types. Among these types, subalpine vegetation, mountain-steppes, xerophilous vegetation of skeleton mountains, and wormwood semi-deserts could be considered as the scope of our study. Makagian (1941) developed the vegetation typology of Armenia in more detail. The steppe and steppe-like vegetation included in his scheme was classified as: • Stony semi-deserts (wormwood semi-desert, wormwood-ephemeral semi-desert, wormwood semi-desert with perennial grasses, etc.) • Steppes (grass-forb and dry forb-grass steppe, feather-grass steppe, beardgrass steppe, fescue and fescue-junegrass steppe, mixed-grass steppe, forb and legume steppe) • Meadow-steppes (grass meadow-steppe, forb and forb-grass meadow-steppe, legume meadow-steppe, dwarf-sedge meadow-steppe) • Highland xerophytic vegetation (Minor-Asian thorn-cushion shrubs of Astracantha, Acantholimon etc., xeromorphic vegetation of screes and rocks) Afterwards, the classification of natural grasslands of Armenia was done by Ziroyan (1989) on the principles of the dominant approach. The author distinguished five vegetation types of desert and steppe vegetation which have strong zonal character and mainly are characteristic to a particular altitudinal belt: deserts (two classes of formations, 16 formations) in the lowest elevations up to 1,000 m a.s.l.; semi-deserts (two classes of formations, 12 formations) at the elevations 1,000–1,300 m a.s.l.; highland xerophytic vegetation (one class of formations, 12 formations), 1,200–1,600 m a.s.l.; mountain steppes (two classes of formations, 12 formations), 1,600–2,000 m a.s.l.; mountain meadow steppes (two classes of formations, 12 formations), 2,000–2,400 (occasionally up to 2,800) m a.s.l. Fayvush (1992) presents a detailed classification of the types of mountain steppes of Armenia based on the dominance approach. The author distinguishes four subtypes (true steppe, thorny-cushion steppe, shrubby steppe and meadow steppe) and 12 classes of formations within it. Overview on the described syntaxa from the surrounding countries In the Transcaucasus region, the first work with a description of syntaxa following the Braun-Blanquet approach was done by M. Guinochet in Azerbaijan and Georgia. Guinochet (1984) described two associations of steppe-like communities without an assignment to higher syntaxonomic units. The association Ziziphoro serpyllaceae-Scutellarietum orientalis Guinochet 1984 is characterized by the presence of many therophytes and is described from the lower elevations in Azerbaijan. Guinochet concludes that this vegetation type could be like the class Thero-Brachypodietea and emphasizes that a unique class and order should be described to include this association. The other association is described from higher elevations in the Pirqulu State Reserve (Azerbaijan), from the subalpine belt: Onobrychieto cyri-Festucetum sulcatae Guinochet 1984 nom. inval. (Article 2b ICPN, Theurillat et al. 2021). This association comprises mountain steppes and is similar, according to M. Guinochet, to the concept of the subalpine steppe of Gadzhiev (1962) from the national typology of Azerbaijan. Additionally, Guinochet (1984) described in Georgia another association from the alpine belt not belonging to the steppic vegetation and assigned to the class Carici rupestris-Kobresietea bellardii: Alchemillo caucasicae-Festucetum supinae, together with the new order Campanulo-tridentatae-Caricetalia tristis and the new alliance Alchemillo-Festucion supinae, all three being invalidly published due to insufficient original diagnosis (Article 2b ICPN, Theurillat et al. 2021). Lately, there was an attempt to classify the phryganoid vegetation of the Nakhchivan region of Azerbaijan (Jabbarov et al. 2020). The authors outlined several association-level units without a formal description (“Thymeto-Acantholimonetum bracteatae”, “Thymeto-Onobrychetum cornutae”), and assigned them to the order Astragalo-Brometalia Quézel 1973 and the class Astragalo-Brometea Quézel 1973. Recently, a new study on the syntaxonomy of alpine and subalpine grasslands has been conducted in Georgia (Nakhutsrishvili et al. 2022). The authors proposed a new class Bromopsio variegatae-Festucetea ovinae to unite subalpine meso-xeric and mesic grasslands, including one new order, three alliances and seven associations. None of the suggested units were published validly due to insufficient original diagnoses. For the Northern Caucasus, Tsepkova (1987) proposed a new class of high-mountain arid grasslands with the provisional name Bothryochloo-Salvietea, which according to Vynokurov et al. (2021) is a syntaxonomic synonym of the Festuco-Brometea. Other steppic grasslands of the Northern Caucasus have been traditionally assigned to the class Festuco-Brometea (Tsepkova 2005; Demina et al. 2017; Vynokurov et al. 2021). In Turkey, several high-level syntaxonomic units have been established for dry grasslands and thorn-cushion vegetation. Zohary (1973) united semi-desert and wormwood steppe grasslands into the class “Artemisietea fragrantis anatolica” Zohary 1973 nom. inv. (Art. 2b ICPN), and proposed the class “Astragaletea armeno-turcica” Zohary 1973 nom. inv. (Art. 2b ICPN) for subalpine tragacanthic vegetation in this region. Simultaneously, Quézel (1973) described another unit for hedgehog plant communities in the subalpine zone of the Taurus Mountains in Turkey, occurring at elevations 2000–2500 m a.s.l., beyond the tree line – Astragalo-Brometea Quézel 1973. In the same publication, Quézel (1973) also published another class of xero-mesophytic and mesophytic subalpine grasslands, occupying higher altitudes than tragacanth vegetation communities: Trifolio anatolici-Polygonetea arenastri Quézel 1973. In the eastern part of Turkey, in Eastern Anatolia, there were some syntaxonomic investigations of Denys Vynokurov et al.: Dry grasslands of Armenia44 steppe vegetation (Çetik and Tatlı 1975; Tatlı 1991; Ocakverdi 1992; Hamzaoğlu 2006; Öztürk et al. 2015). Ocakverdi (1992) surveyed the vegetation (including steppe vegetation) in the region of Turkey bordering with Armenia. He distinguished two physiognomic types of steppe vegetation (grass steppe and tragacanth steppe) and three altitudinal variants: “plain” steppe (1,675–1,725 m a.s.l., mainly grass steppe vegetation), lower mountain steppe (1,750–1,850 m a.s.l., both grass steppe and tragacanth steppe) and high-mountain steppe (1,950–2,696 m a.s.l., grass steppe). He proposed four associations, none of them published validly (Art. 1 ICPN). Later, Ocakverdi et al. (2009) described 10 new associations from the same region (Kısır Mountain). Hamzaoğlu (2006) studied steppe communities of East Anatolia. He united the studied vegetation to a new order Festuco oreophilae-Veronicetalia orientalis Hamzaoğlu 2006, subordinated to the class Astragalo-Brometea Quézel 1973. In another bordering region, Iran, xerophilous grassland and scrub communities were first delineated by Zohary (1973), who proposed several vegetation classes: “Artemisietea fragrantis anatolica” Zohary 1973 nom. inv. (Art. 2b ICPN) for wormwood steppe grasslands in Northwestern Iran and Anatolia, “Artemisietea herbae-albae iranica” Zohary 1973 nom. inv. (Art. 2b ICPN) grouping wormwood semideserts in the Central Plateu of Iran, and “Astragaletea iranica” Zohary 1973 nom. inv. (Art. 2b ICPN) for tragacanth communities in Iran and Iraq. Later, the vegetation of the subalpine and alpine belts of the Alborz Mountains have been studied by Klein (1982, 1987). He described several new classes: Onobrychidetea cornutae Klein 1987 nom. inval. (Art. 2b ICPN) from the lower alpine belt of Alborz (3,200–3,500 m a.s.l.) and Prangetea ulopterae Klein 1987 nom. inval. (Art. 2b ICPN) from the subalpine belt of Alborz (2,500–3,200 m a.s.l.), aimed to unite high-mountain hedgehog communities and xeric tall-herb vegetation respectively. Also, from the subalpine belt of the northern macroslope of Alborz Mountains, Klein and Lacoste (1994) described one association subordinated to the Festuco-Brometea, Alchemilletum plicatissimae Klein et Lacoste 1994, but did not assign it to an alliance or order. In general, the class Astragalo-Brometea is the most widely used name to unite the dry grasslands and thorn-cushion vegetation in the western part of the Irano-Turanian region. However, its conceptual boundaries have undergone significant changes over time, both geographically and physiognomically. Many researchers now extend its scope to include tragacanth vegetation not only from the subalpine belt but also from the lower elevations, as well as chamaephyte-dominated phryganoid vegetation, non-tragacanth dry grasslands at lower elevations, saline steppes, and gypsophilous rocky grasslands (Ketenoğlu et al. 1983; Akman et al. 1984; Aydoğdu et al. 1994; Aydoğdu et a. 2004; Hamzaoğlu et al. 2004; Kaya 2011). Additionally, xero-mesophytic and mesophytic grasslands at higher elevations, which were classified by Quézel (1973) as a separate class Trifolio-Polygonetea, are sometimes included within the Astragalo-Brometea (Eren et al. 2004; Parolly 2004). Some authors have proposed extending the concept of the class also to the northern part of Iran, synonymizing the Onobrychidetea cornutae and Prangetea ulopterae described there (Parolly 2004). Recently, the dry feather-grass steppes of Tajikistan also were provisionally included into the Astragalo-Brometea (Nowak et al. 2016, 2018). Methods Field sampling in Armenia We sampled 111 plots of 10 m2size (Suppl. materials 1, 2) during the 13th Field Workshop of the Eurasian Dry Grassland Group (EDGG) in Armenia, from 26 June to 6 July 2019 (Aleksanyan et al. 2020; for distribution of sites, see Figure 1). Within each plot, we recorded vascular plants and terricolous bryophytes and lichens with the shoot presence method (Dengler 2008). Besides, we estimated their percentage cover on a continuous scale (for discussion of advantages of this method compared to ordinal scales, see Dengler and Dembicz 2023). Specimens that could not be determined in the field were dried and taken to the lab for further determination. Other environmental and structural parameters that were recorded in situ following the EDGG sampling methodology (Dengler et al. 2016), included: geographical position (latitude, longitude), elevation (m), slope aspect (°), slope inclination (°), maximum microrelief (cm), soil depth (cm, mean and SD of five measurements per plot), vegetation covers (%; total vegetation, shrub layer, herb layer, cryptogam layer), cover of litter (%), covers of stones and rocks, gravel, and fine soil (all three fractions summing up to 100%), maximum height of shrubs (m), maximum height of herbs (cm), height of herb layer (cm, mean and SD of five measurements with a falling disc per plot), and land use details (grazing, mowing, burning, abandonment). Soil was collected as mixed samples from the uppermost 15 cm of the soil in five random points inside each plot. After air drying and sieving to the fine-soil fraction, the following parameters were measured in the lab: pH (in H2O), electrical conductivity (μS cm-1), organic C content (%), humus content (%), N content (%), and C/N ratio. Southing was calculated from aspect as -cos (aspect). The nomenclature of vascular plants was standardised to Euro+Med (2023) for vascular plants, Hodgetts et al. (2020) for bryophytes, Nimis et al. (2018) for lichens. For some groups of closely related species that could not always be distinguished, we defined additional species aggregates (“aggr.”; see Suppl. material 3). The value distribution of all recorded and analysed numerical environmental, structural and biodiversity variables is given in Suppl. material 4. The complete data are stored in and available from the GrassPlot database (Dengler et al. 2018; Biurrun et al. 2019; https://edgg.org/databases/GrassPlot) as dataset “AM_B”. Vegetation Classification and Survey 45 Data from external sources Plot data from the surrounding countries To be able to identify the high-level syntaxonomic units, we digitised from literature and used for comparison relevés from the bordering regions, focusing on the original diagnosis of the high-level units of similar vegetation types: 230 plots from Anatolia (Turkey) and 51 plots from Northern Iran. Among them, the original diagnosis of the class Astragalo-Brometea Quézel 1973 with the type order Astragalo-Brometalia Quézel 1973 and the other orders Drabo-Androsacetalia Quézel 1973, Hyperico linarioidis-Thymetalia scorpilii Akman et al. 1987, Onobrychido armenae-Thymetalia leucostomi Akman et al. 1985, Festuco oreophilae-Veronicetalia orientalis Hamzaoğlu 2006, and other lower rank units (Quézel 1973; Akman et al. 1984, 1987; Gümüs 1992; Klein and Lacoste 1994; Hamzaoğlu 2006; Ocakverdi et al. 2009). Also, 206 plots from Northern Caucasus (Russia) were taken from the Eastern European Steppe Database (Vynokurov et al. 2020). Combined with our own plots from Armenia, a dataset of 598 relevés resulted (Suppl. material 5). After the unification of the species taxonomy, removing the taxa determined to the genus level, and combining the aggregates, the final dataset contained 1,556 taxa. Environmental data from external sources Climatic data were extracted from the CHELSA Climate database (Karger et al. 2017, 2018). We used the following variables: BIO01 – mean annual air temperature, BIO07 – annual range of air temperature, BIO12 – annual precipitation amount, BIO17 – mean monthly precipitation amount of the driest quarter. Maps were created using QGIS software (QGIS Development Team 2009). As basemaps, we used the map of main vegetation types of Armenia by Grossgeim (1928), and the SRTM (Shuttle Radar Topography Mission) elevation model for Armenia (EROS Center 2017). Attributes for the Armenian species In order to assess the distribution ranges of species, we analysed their distribution maps (according to GBIF 2023 and POWO 2023) and classified their distribution ranges into the following broad categories: European (occurring mainly in the temperate regions of Europe and Western Siberia), Mediterranean (covering the southern part of Europe in areas with Mediterranean climate), Irano-Turanian (occurring in Western and Central Asia), Transcaucasian (narrow endemics of Armenia and surrounding regions of Transcaucasia), Caucasian (broader endemics of the whole Caucasus Range, including the Northern Caucasus and adjacent parts of Eastern Anatolia in Turkey and North-Western Iran). If the species occurred predominantly in one of the mentioned regions, we assigned a value of one (1) to the corresponding category. In case that the species occurred in two regions simultaneously (European and Irano-Turanian, European and Mediterranean, or Mediterranean and Irano-Turanian), we assigned a value of 0.5 to the corresponding categories. If the species distribution covered three regions (European, Mediterranean and Irano-Turanian), we assigned a value ‘0.3’ to each of these categories. In all other cases, when the species distribution area was greater than the mentioned categories (e.g., Palaearctic, Holarctic, etc.), we classified these species into the category ‘Other’. The assigned distribution ranges of species are available in Suppl. material 6. In addition, we classified all the species into one of the Raunkiær plant life forms: therophytes, geophytes, hemicryptophytes, chamaephytes, and phanerophytes. The data are also available in the Suppl. material 6. Statistical analysis Unsupervised classification Unsupervised classification for both the West Asian and Armenian dataset was done in JUICE 7.0 (Tichý 2002) using the modified TWINSPAN method (Roleček et al. 2009) with three pseudospecies cut levels (0, 5, and 15), and Whittaker’s beta-diversity index as a measure of internal cluster heterogeneity. Diagnostic species were determined based on phi values (Chytrý et al. 2002), standardised to equal plot numbers at association level (Tichý and Chytrý 2006). This was done hierarchically at the four syntaxonomic levels from association to class (García-Mijangos et al. 2021). Since this approach is not implemented in JUICE (Tichý 2002) yet, we had to do it in Microsoft Excel, which precluded the use of Fisher’s exact test for significance. To avoid selecting non-significant diagnostic species, we put the thresholds for phi values rather high. For associations and alliances, we used > 0.4 for diagnostic species and > 0.6 for highly diagnostic species, while for orders and classes the thresholds were > 0.3 and > 0.5, respectively. Moreover, we also ensured that the phi values were at least 0.2 higher in the target syntaxon than the syntaxon of the same rank with the next-lower phi value (see García-Mijangos et al. 2021). If a species fulfilled the criteria to be diagnostic at several hierarchical levels, it was assigned to the level with the highest phi value. In case of monotypic syntaxa, diagnostic species were only assessed at the higher level. Ordination DCA-Ordination was performed with Canoco 5 (ter Braak and Šmilauer 2012) with log-transformed percent cover values of species and downweighting rare species and post hoc fitted variables (environmental, calculated or measured). Comparison of syntaxon characteristics Differences in variables between syntaxa were tested by univariate ANOVA using SPSS 22 (IBM, Armonk, NY, US). We tested whether the assumptions of ANOVA (normal distribution, equal variance) were sufficiently met by visually inspecting the frequency distribution of the residuals and by testing for homogeneity of variance according to Levene (Quinn and Keough 2002). Where ANOVA revealed a significant pattern, Tukey’s post-hoc Denys Vynokurov et al.: Dry grasslands of Armenia46 test at p < 0.05 was used to identify homogeneous groups of syntaxa. Results were presented as box-whisker plots with median and mean, and 25th/75th (boxes) and 10th/90th (whiskers) percentiles as well as outliers. Syntaxonomic assignment We selected a TWINSPAN resolution where the terminal clusters were floristically still well-characterised and not too small. These clusters were then assigned to the rank of association. Alliance, order and class levels were assigned to higher cut levels of the dendrogram, with the double aim to have floristically well differentiated and ecologically and chorologically interpretable units on a comparable level as these hierarchies have in Mucina et al. (2016). After defining the hierarchical units, we carefully checked the syntaxonomic literature of the neighbouring countries to determine whether syntaxa with this content already existed. If this was the case, we took the established name. If not, we formally described new syntaxa according to the ICPN (Theurillat et al. 2021). Following Recommendation 7A of the ICPN, we refrained from establishing new associations when we had fewer than 10 relevés and instead treated the respective cluster as an informal community at association rank. Likewise, we refrained from a formal description of the new class that was supported by our analyses, suggesting that this should first be “validated” in a broader-scale analysis involving the neighbouring countries of Armenia. Results Unsupervised classification and ordination of the West Asian and Caucasian dataset Modified TWINSPAN resulted into 16 clusters with five main groups of clusters: A (clusters 1–3), B (4–6), C (7), D (8–10) and E (11–16) (Figure 2). The synoptic table built with these five main groups is shown in Suppl. material 7. The group A (clusters 1–3) completely consisted of the relevés originally assigned to the class Astragalo-Brometea, including the type order Astragalo-Brometalia Quézel 1973 (in the cluster 2), described from the Taurus Mountains in South-Western Turkey (Quézel 1973). The plots assigned to the other orders of this class with their corresponding types were also included in this group of clusters: Drabo-Androsacetalia Quézel 1973 (cluster 2) from the same Taurus Mountain range; Hyperico linarioidis-Thymetalia scorpilii Akman et al. 1987 (cluster 1) from the Ilgaz Mountains in North-Western Turkey; Onobrychido armenae-Thymetalia leucostomi Akman et al. 1985 (cluster 3) described from Central Anatolia. Plots from Armenia did not fit into this group. The second group B (clusters 4–6) comprised plots from the high-mountain steppe vegetation. Cluster 4 consisted mainly of relevés from the northern slope of the Alborz Mountains in Iran, assigned to the association Alchemilletum plicatissimae Klein et Lacoste 1994. Plots from the Northern Caucasus region assigned to the alliance Artemisio chamaemelifoliae-Bromopsion variegatae Vynokurov in Vynokurov et al. 2021 were classified into cluster 5, together with some plots from Armenia. Cluster 6 comprised plots sampled near the Kısır Mountain in Turkey, in Eastern Anatolia. They were originally assigned to several associations of the class Astragalo-Brometea but without placement in alliances and orders (Ocakverdi et al. 2009). Thorn-cushion communities from Eastern Anatolia are combined in the group C (cluster 7). They were originally assigned to the order Festuco oreophilae-Veronicetalia orientalis Hamzaoğlu 2006 of the class Astragalo-Brometea, together with its type alliance Festuco oreophilae-Veronicion orientalis Hamzaoğlu 2006 and the respective association Astragalo-Onobrychidetum cornutae Gümüs 1992. Even though this cluster was not placed in the corresponding group A in the dendrogram (Figure 2), they seem closely related according to the ordination (Figure 3). Clusters 8–10 formed group D. It consisted exclusively of plots from Armenia. Among them, cluster 8 contained the most xeric communities sampled in the driest parts of Armenia, followed by cluster 10. Cluster 9 was transitional between the groups D and E. The group E (clusters 11–16) was formed by plots containing the more ‘typical’ Festuco-Brometea species, mostly from the region of the Northern Caucasus. Cluster 11 was comprised mainly of rocky grasslands belonging to the order Asphodelino tauricae-Euphorbietalia petrophilae Vynokurov in Vynokurov et al. 2021; cluster 12 contained grass steppes mostly of the Festucetalia valesiacae. Cluster 13 combined mountain steppes exclusively from Armenia. Clusters 14–16 comprised mostly meso-xeric communities of the order Brachypodietalia pinnati from the Northern Caucasus. A synoptic table with the five distinguished groups of clusters (A–E) is provided in Suppl. material 7. Figure 2. Results of the Modified TWINSPAN classification for the broad-scale comparison involving plots from the bordering countries (n = 598). The width of the bars is proportional to the number of included plots. The main groups (letters) and terminal clusters (numbers) are described in the text. Blue colour indicates clusters that predominantly or completely consisted of Armenian plots. Vegetation Classification and Survey 47 General overview of the Armenian plots In our 111 plots of 10 m2, we recorded a total of 739 vascular plant, 40 bryophyte and 13 lichen taxa (subspecies, species, aggregates and sections, further as ‘species’). The species richness per plot ranged from 21 to 85, with a mean of 47.3. On average there were 46.8 vascular plant, 0.4 bryophyte and 0.1 lichen species per plot. The most frequent vascular plant was Galium verum (in 72% of all plots), followed by Thymus kotschyanus (59%), Teucrium capitatum (58%), Poa bulbosa (55%), Dactylis glomerata (54%), Scutellaria orientalis aggr. (53%), Koeleria macrantha (51%), Stachys recta (50%) and Potentilla recta aggr. (50%). The most frequent bryophytes were Syntrichia ruralis (27%), Ptychostomum imbricatulum (19%) and Syntrichia montana (14%). Lichens were absent in most plots, with the most frequent one (Cladonia foliacea) reaching just 4%. Classification of the Armenian dataset The most meaningful modified TWINSPAN classification of the plots from Armenia resulted in the 12-cluster solution (Figure 4). The first cluster (X) had only a single relevé of scree vegetation recorded in the Vayots Dzor Province near Hermon. The rest of the clusters are interpreted at the community or association level. Clusters 1.1.1.1–1.1.3.2 consisted of the most xeric plots of the semi-desert, rocky and thorn-cushion vegetation in the lower elevations. Clusters 2.1.1.1 and 2.1.1.2 represent plots of mountain meadow steppes from the highest elevations. Clusters 2.2.1.1–2.2.1.4 consisted of plots of thorn-cushion and steppic grasslands of so-called mountain steppes. Ordination of the Armenian dataset Ordination of the plots with the assignment to one of these clusters revealed that the first axis of the DCA graph (Figure 5) corresponds to a gradient of moisture and temperature connected with the elevation range. The most mesic plots occupied higher altitudes (cluster 2.1.1.1), while the most xeric ones were distributed in the lower elevations (clusters 1.1.2.1–1.1.3.2). The climatic-elevation gradient and community parameters correlated with the differentiation of two higher classification units – Ziziphora tenuior-Stipa arabica grasslands and the class Festuco-Brometea. Syntaxonomic scheme Resulting from our analyses of the Armenian data and the comparison with the syntaxa of neighbouring territories, we propose the following syntaxonomic scheme Figure 3. DCA of the West Asian dataset (DCA with supplementary variables, eigenvalues/gradient lengths/cumulative explained variation of axis 1: 0.6313/5,37/4.02, axis 2: 0.4207/4.94/6.69). Vectors (environmental variables): BIO1: annual mean temperature; BIO7: temperature annual range; BIO12: annual precipitation; elevation: elevation (m a.s.l.); N species: vascular plant richness. Figure 4. Results of the Modified TWINSPAN classification of the Armenian plots (n = 111). The terminal clusters were interpreted as associations or, if represented by too few plots, as informal communities at association rank. The first cluster (X) consisted only of one plot of scree vegetation. Codes at the tips of the other clusters correspond to Table 1. Denys Vynokurov et al.: Dry grasslands of Armenia54 Syntaxon All 1 2 1.1 2.1 2.2 1.1.1 1.1.2 1.1.3 2.1.1 2.2.1 1.1.1.1 1.1.1.2 1.1.2.1 1.1.3.1 1.1.3.2 2.1.1.1 2.1.1.2 2.2.1.1 2.2.1.2 2.2.1.3 2.2.1.4 Number of plots 110 47 63 47 21 42 17 13 17 21 42 7 10 13 10 7 13 8 8 7 16 11 Mean species richness in 10 m2 (all) 47.4 45.0 49.2 45.0 50.1 48.8 43.5 45.0 46.6 50.1 48.8 42.4 44.2 45.0 45.3 48.4 56.8 39.3 49.1 52.3 42.3 55.8 Mean species richness in 10 m2 (vascular plants) 46.9 44.6 48.7 44.6 49.3 48.3 43.2 44.9 45.9 49.3 48.3 41.9 44.1 44.9 44.6 47.7 55.6 39.1 48.9 51.4 41.9 55.3 Mean species richness in 10 m2 (byophytes) 0.4 0.2 0.5 0.2 0.6 0.4 0.3 0.1 0.3 0.6 0.4 0.6 0.1 0.1 0.3 0.3 0.9 0.0 0.3 0.7 0.3 0.5 Mean species richness in 10 m2 (lichens) 0.1 0.1 0.1 0.1 0.2 0.0 0.0 0.0 0.4 0.2 0.0 0.0 0.0 0.0 0.4 0.4 0.2 0.1 0.0 0.1 0.1 0.0 Rhinanthus subulatus 2 . 3 . . 5 . . . . 5 . . . . . . . . 29 . . Lathyrus latifolius 2 . 3 . . 5 . . . . 5 . . . . . . . . 29 . . Valeriana officinalis aggr. 2 . 3 . . 5 . . . . 5 . . . . . . . . 29 . . Campanula bononiensis 2 . 3 . . 5 . . . . 5 . . . . . . . . 29 . . Melampyrum sp. 2 . 3 . . 5 . . . . 5 . . . . . . . . 29 . . Seseli libanotis 8 . 14 . 14 14 . . . 14 14 . . . . . 15 13 13 57 6 . Assoc. 2.2.1.4 Linum tenuifolium 15 . 25 . . 38 . . . . 38 . . . . . . . 13 . 44 73 Coronilla coronata 4 . 6 . . 10 . . . . 10 . . . . . . . . . . 36 Asphodeline taurica 4 . 6 . . 10 . . . . 10 . . . . . . . . . . 36 Androsace chamaejasme 4 . 6 . . 10 . . . . 10 . . . . . . . . . . 36 B Hypnum cupressiforme 13 4 19 4 14 21 . 8 6 14 21 . . 8 . 14 15 13 13 . . 73 Campanula sibirica 9 6 11 6 . 17 18 . . . 17 43 . . . . . . . . . 64 Sesleria phleoides 6 2 10 2 5 12 . 8 . 5 12 . . 8 . . 8 . . . . 45 Asperula affinis 5 . 8 . . 12 . . . . 12 . . . . . . . . . 6 36 Campanula rapunculoides 11 . 19 . 5 26 . . . 5 26 . . . . . 8 . 13 43 . 64 Pontechium maculatum 7 . 13 . 14 12 . . . 14 12 . . . . . 23 . . . . 45 Fritillaria caucasica 3 . 5 . . 7 . . . . 7 . . . . . . . . . . 27 Differential species in more than one association Arenaria dianthoides 5 . 8 . 14 5 . . . 14 5 . . . . . . 38 25 . . . Gagea caroli-kochii 5 . 8 . 10 7 . . . 10 7 . . . . . . 25 38 . . . Silene cephalantha 5 . 8 . 10 7 . . . 10 7 . . . . . . 25 38 . . . Companion species Thymus kotschyanus 60 66 56 66 48 60 59 69 71 48 60 43 70 69 60 86 31 75 50 71 56 64 Scutellaria orientalis aggr. 54 62 48 62 43 50 71 54 59 43 50 100 50 54 50 71 15 88 50 57 44 55 Koeleria macrantha 52 55 49 55 43 52 59 77 35 43 52 57 60 77 40 29 62 13 38 29 63 64 Elytrigia intermedia aggr. 47 49 46 49 19 60 65 46 35 19 60 71 60 46 50 14 8 38 88 57 44 64 Eryngium billardierei 38 43 35 43 5 50 24 46 59 5 50 14 30 46 60 57 . 13 50 43 69 27 Convolvulus lineatus 36 51 25 51 . 38 71 62 24 . 38 43 90 62 40 . . . 25 . 50 55 Medicago sativa 35 34 37 34 14 48 29 46 29 14 48 . 50 46 50 . 15 13 38 71 44 45 Alyssum alyssoides 35 43 29 43 19 33 41 38 47 19 33 43 40 38 40 57 23 13 50 . 56 9 Festuca valesiaca aggr. 35 23 43 23 48 40 41 8 18 48 40 14 60 8 20 14 54 38 75 . 38 45 Euphorbia seguieriana 32 40 25 40 10 33 24 62 41 10 33 43 10 62 50 29 8 13 13 29 56 18 Centaurea ovina aggr. 27 38 19 38 . 29 47 46 24 . 29 . 80 46 30 14 . . . 14 56 18 B Syntrichia ruralis 27 21 32 21 24 36 6 31 29 24 36 . 10 31 50 . 23 25 38 57 19 45 Dianthus cretaceus 26 21 30 21 52 19 29 31 6 52 19 14 40 31 . 14 54 50 13 29 6 36 Medicago x varia 26 26 27 26 29 26 41 23 12 29 26 43 40 23 20 . 8 63 13 . 44 27 Ziziphora clinopodioides 25 19 30 19 14 38 24 . 29 14 38 . 40 . 40 14 . 38 13 43 56 27 Hypericum scabrum 25 32 19 32 . 29 41 .47 . 29 29 50 . 60 29 . . 13 . 56 18 Leontodon asperrimus 25 13 33 13 33 33 29 . 6 33 33 14 40 . 10 . 23 50 25 14 38 45 Thesium ramosum 22 36 11 36 5 14 53 23 29 5 14 43 60 23 30 29 8 . 13 29 6 18 Plantago lanceolata 22 15 27 15 10 36 35 . 6 10 36 14 50 . 10 . 8 13 63 43 25 27 Onobrychis cornuta 21 19 22 19 . 33 29 8 18 . 33 . 50 8 20 14 . . . 29 44 45 Falcaria vulgaris 20 17 22 17 10 29 18 8 24 10 29 14 20 8 40 . 8 13 13 57 38 9 Vegetation Classification and Survey 55 Potential class 1: Ziziphora tenuior-Stipa arabica grasslands – Western Asian dry grasslands and xeric thorn cushion communities Diagnostic species: Achillea arabica, Aegilops cylindrica, Agropyron cristatum, Allium pseudoflavum, Alyssum turkestanicum, Androsace maxima, Anisantha tectorum, Arabis auriculata aggr., Arenaria serpyllifolia aggr., Artemisia fragrans, Asperula arvensis, Bromus danthoniae, B. japonicus, B. squarrosus, Centaurea aggregata, Chardinia orientalis, Crepis sancta, Crupina vulgaris, Dianthus orientalis, Helianthemum ledifolium, Helichrysum plicatum, Holosteum marginatum, H. umbellatum, Marrubium parviflorum, Meniocus linifolius, Minuartia hamata, Noaea mucronata, Poa bulbosa, Sideritis montana, Stachys inflata, S. lavandulifolia, Stipa arabica, S. capillata, S. holosericea, Taeniatherum caput-medusae subsp. crinitum, Tanacetum aureum, Teucrium capitatum, Thymelaea passerina, Xeranthemum squarrosum, Ziziphora tenuior. Ecology and distribution. Communities of the potential new class occur in the lower elevations in dry conditions and include semi-desert vegetation, xeric thorn-cushion communities, and xeric grasslands. Within Armenia, it is represented by one order and three alliances. Order 1.1: Cousinio brachypterae-Stipetalia arabicae – Western Asian dry grasslands and xeric thorn cushion communities Diagnostic species: Identical with those of the monotypic class. Ecology and distribution. Communities of this order are distributed in Transcaucasia and possibly even broader within Western Asia. We expect them to occur throughout Western Asia in dry conditions (semi-deserts, dry steppe-like communities, low-elevation thorn-cushion communities). Also, according to our analysis, communities of this order may occur even in higher elevations on rocky substrates. Alliance 1.1.1: Onobrychido michauxii-Stipion capillatae – Transcaucasian rocky dry grasslands Diagnostic species: Astracantha condensata, Onobrychis michauxii, Salvia aethiopis, Stachys lavandulifolia, Teucrium capitatum, Veronica multifida (mainly negatively differentiated central alliance). Ecology and distribution. Communities of this alliance are distributed in higher elevations than those of the other two alliances included in this order. This alliance is a transitional unit between this order and the order Onobrychido transcaucasicae-Stipetalia pulcherrimae, comprising Transcaucasian mountain steppes (see below). 1.1.1.1: Stachys lavandulifolia-Astracantha condensata community (Figure 7B) Diagnostic species: Asperula arvensis, Astracantha condensata, Centaurea phrygia subsp. abbreviata, Crepis ciliata, Euphorbia orientalis, Gypsophila elegans, Herniaria hirsuta, Leptunis trichodes, Melica taurica, Nepeta racemosa, Onobrychis michauxii, Onosma setosa, Salvia aethiopis, Sempervivum transcaucasicum, Stachys lavandulifolia, Tanacetum aureum, Teucrium orientale, Tragopogon dubius, Viola occulta, Zosima absinthiifolia. Structure, ecology and distribution. We sampled this vegetation type in the Gegharkunik (vicinity of the town of Sevan, Shorja) and Vayots Dzor (Hermon, vicinity of Gnishik and Khachik) provinces. These communities were located at the most south-facing rocky slopes with shallow soil and low humus content. The herb layer was sparse and with a high representation of Irano-Turanian species, e.g. Astracantha condensata, A. microcephala, Stachys lavandulifolia, Teucrium orientale. 1.1.1.2: Marrubio parviflorae-Stipetum capillatae (Figure 7C) Diagnostic species: Allium cardiostemon, Centaurea ovina aggr., Euphorbia condylocarpa, Marrubium parviflorum, Stipa capillata. Structure, ecology and distribution: These communities were sampled on slopes with shallow rocky substrates in Gegharkunik (Ardanish), Lori (near Shirakamut) and Vayots Dzor (vicinity of Gnishik and Khachik) provinces. The association differed by a higher herb layer cover compared to the previous community and the highest participation of hemicryptophytes among all associations of the class. The dominant species were Festuca valesiaca aggr., Onobrychis cornuta and Teucrium capitatum. Alliance 1.1.2: Artemision fragrantis – Transcaucasian wormwood semi-deserts Diagnostic species: Agropyron cristatum, Allium pseudoflavum, Alyssum turkestanicum, Androsace albana, Arenaria serpyllifolia aggr., Artemisia fragrans, Astragalus hyalolepis, Bromopsis riparia, Ceratocephala falcata, Cousinia brachyptera, Crupina vulgaris, Cuscuta araratica, Consolida hispanica, Didymodon tophaceus (B), Meniocus linifolius, Minuartia hamata, Noaea mucronata, Peganum harmala, Polygala hohenackeriana, Sclerocaryopsis spinocarpos, Syntrichia caninervis (B). Ecology and distribution: Artemisia fragrans semi-deserts in Armenia are distributed in the lowest elevations in the country. We did not sample other semi-desert types, but we can expect that Armenian loamy and sandy semi-deserts will also be probably included in this unit. In our dataset, this alliance is represented by a single association. 1.1.2.1: Noaeo mucronatae-Artemisietum fragrantis (Figure 7D) Diagnostic species: identical with those of the monotypic alliance. Structure, ecology and distribution: This association is typical for the Aragatsotn province (vicinity of Dashtadem and Tatool). The sampled plots were distributed at the lowest elevations with the highest mean annual temperature and lowest mean annual precipitation compared to the other studied associations. The communities were dominated by Artemisia fragrans, Poa bulbosa, and Taeni- Denys Vynokurov et al.: Dry grasslands of Armenia56 atherum caput-medusae subsp. crinitum. The herb layer is relatively sparse and with a high representation of therophytes (e.g. Alyssum turkestanicum, Bromus squarrosus, Ceratocephala falcata, Crupina vulgaris, Sclerocaryopsis spinocarpos) and characteristic chamaephytes (Artemisia fragrans, Noaea mucronata). Alliance 1.1.3: Acantholimono caryophyllacei-Stipion holosericeae – Transcaucasian dry grasslands and xeric thorn-cushion communities Diagnostic species: Acantholimon vedicum, Aegilops cylindrica, Aethionema carneum, Arabis auriculata aggr., Bromus japonicus, Bunium microcarpum, Chardinia orientalis, Crepis sancta, Crupina vulgaris, Ephedra procera, Galium verticillatum, Gaudiniopsis macra, Helianthemum ledifolium, Noccaea perfoliata, Papaver minus, Roemeria hybrida, Stachys inflata, Stipa arabica, S. holosericea, Taeniatherum caput-medusae subsp. crinitum, Petrorhagia cretica, Ziziphora tenuior. Ecology and distribution: This unit comprises vegetation traditionally known as ‘highland xerophytic vegetation’. It includes dry grasslands and xeric tragacanth communities distributed above the semi-desert belt and below the mountain steppe altitudinal belt. We distinguish one association and one community within this alliance. 1.1.3.1: Acantholimono caryophyllacei-Stipetum holosericeae (Figure 7E) Diagnostic species: Acantholimon caryophyllaceum, Aegilops triuncialis, Carduus hamulosus, Crepis sancta, Geranium lucidum, Lomelosia rotata, Noccaea annua, Stipa zalesskii subsp. pontica, Torilis arvensis, Xeranthemum squarrosum. Structure, ecology and distribution: We sampled this association mainly in the Vayots Dzor province (Hermon, vicinities of Areni, Gnishik and Khachik), and also in one locality in Aragatsotn province (near Tatool). The communities were distributed on shallow soils, but with higher humus content and lower gravel cover compared to the other associations of this class. Acantholimon caryophyllaceum, Stipa holosericea, and Taeniatherum caput-medusae subsp. crinitum are the dominant species in this association. Among other species, Irano-Turanian elements often occur, such as Achillea arabica, Eryngium billardierei, Hypericum scabrum, and Thymus kotschyanus. 1.1.3.2: Stachys inflata-Acantholimon vedicum community (Figure 7F) Diagnostic species: Acantholimon vedicum, Aegilops biuncialis, A. cylindrica, Aethionema carneum, Androsace maxima, Arabis auriculata aggr., Artemisia fragrans, Aspicilia hispida (L), Astragalus ornithopodioides, Bunium microcarpum, Callipeltis cucullaria, Camelina laxa, Chardinia orientalis, Cousinia daralaghezica, Crossidium squamiferum (B), Crucianella exasperata, Crupina vulgaris, Cuscuta pedicellata, Ephedra procera, Galium verticillatum, Gaudiniopsis macra, Helianthemum ledifolium, Holosteum marginatum, Lactuca tuberosa, Lamium amplexicaule, Onobrychis atropatana, Papaver minus, Petrorhagia cretica, Stachys inflata, Stipa arabica, Tanacetum aureum, Trinia glauca, Valerianella coronata, Ziziphora tenuior. Structure, ecology and distribution: We sampled this vegetation type in Ararat (vicinity of Tigranashen) and Vayots Dzor (vicinity of Gnishik) provinces. Communities were distributed at lower elevations with high mean annual temperature and low mean annual precipitation. The substrate differed by the most alkaline soil reaction (mean pH: 8). The herb layer was sparse and with a high representation of Irano-Turanian and Mediterranean therophyte species (Aegilops spp., Crupina vulgaris, Petrorhagia cretica, Ziziphora tenuior), while the cover of hemicryptophytes was the lowest among all studied communities. These communities did not have clear dominants, but Chardinia orientalis, Stachys inflata, Stipa arabica, and S. sareptana subsp. anisotricha occurred with higher cover than the other species. The species richness of vascular plants, bryophytes and lichens was higher compared to the other associations of the class. Class 2: Festuco-Brometea – Mesoxeric and xeric basiphilous grasslands of temperate Europe and adjacent regions Diagnostic species: Abietinella abietina (B), Achillea millefolium aggr., Artemisia absinthium, Bupleurum falcatum aggr., Campanula glomerata aggr., C. rapunculoides, C. stevenii, Cirsium leucocephalum, Dactylis glomerata, Galium verum, Koeleria albovii, Leontodon hispidus, Linum nervosum, L. tenuifolium, Lotus corniculatus, Onobrychis transcaucasica, Origanum vulgare, Phleum phleoides, Pimpinella saxifraga aggr., Plantago atrata, Poa pratensis aggr., Polygala anatolica, Polygonum cognatum, Potentilla recta aggr., Scabiosa bipinnata, Securigera varia, Stachys macrostachys, Taraxacum sect. Taraxacum, Thalictrum minus, Trifolium alpestre, T. ambiguum, T. trichocephalum, Trisetum flavescens. Ecology and distribution: Within Armenia, this class comprises meso-xeric grasslands and mountain steppes at higher elevations. We distinguish two orders representing different altitudinal belts. Order 2.1: Plantagini atratae-Bromopsietalia variegatae – High-mountain meso-xeric grasslands of the Caucasus Diagnostic species: Achillea millefolium aggr., Ajuga orientalis, Alchemilla sericea, Arenaria blepharophylla aggr., A. gypsophiloides, Aster alpinus, Brachypodium pinnatum, Bromopsis variegata, Campanula collina, C. stevenii, Cirsium leucocephalum, Festuca ovina aggr., Schedonorus pratensis, Festuca rubra aggr., Filipendula vulgaris, Gagea glacialis, Galium cordatum, Gentiana septemfida, Huynhia pulchra, Koeleria albovii, Lathyrus digitatus, Lomelosia caucasica, Lotus corniculatus, Luzula multiflora, Medicago papillosa, Muscari armeniacum, Myosotis alpestris, Ornithogalum sigmoideum, Papaver orientale, Pedicularis condensata, Phleum alpinum, Pilosella officinarum aggr., Pimpinella saxifraga aggr., Plantago atrata, Poa pratensis Vegetation Classification and Survey 57 aggr., Pohlia nutans (B), Polygonum cognatum, Potentilla argentea, Psephellus xanthocephalus, Pseudoleskella tectorum (B), Pulsatilla albana, Ranunculus caucasicus, Rumex acetosella, Senecio pseudo-orientalis, Stachys macrantha, Stipa tirsa, Taraxacum sect. Taraxacum, Thymus collinus, Tortula acaulon (B), Trifolium ambiguum, T. spadiceum, T. trichocephalum, Verbascum speciosum, Veronica denudata, V. gentianoides. Ecology and distribution: Communities of this order occupy the highest sampled elevations in Armenia: upper subalpine and lower alpine belts. They form a particular unit recognized in the dominant approach typology: mountain meadow steppes. Beyond this elevation belt, they are replaced by the alpine grasslands which possibly belong to the class Juncetea trifidi Hadač in Klika et Hadač 1944 (Festucetalia woronowii Tsepkova 1987). Alliance 2.1.1: Artemisio chamaemelifoliae-Bromopsion variegatae – Caucasian subalpine and lower-alpine meso-xeric grasslands Diagnostic species: Achillea millefolium aggr., Alchemilla sericea, Arenaria gypsophiloides, Bromopsis variegata, Campanula collina, C. stevenii, Festuca ovina aggr., F. rubra aggr., Huynhia pulchra, Koeleria albovii, Lathyrus digitatus, Lomelosia caucasica, Lotus corniculatus, Myosotis alpestris, Plantago atrata, Polygonum cognatum, Potentilla argentea, Ranunculus caucasicus, Taraxacum sect. Taraxacum, Trifolium ambiguum, T. trichocephalum, Veronica denudata, V. gentianoides. Ecology and distribution: This unit was described from the Main Range of the North Caucasus (Vynokurov et al. 2021) in the elevations of 1,800–2,200 m a.s.l. In Armenia, it occurs mainly higher than 2,000 m a.s.l., and shares multiple species with the Northern Caucasus unit. Thus, we are classifying mountain meadow steppes of Armenia within the same alliance. Here we distinguish one association and one informal community within it. 2.1.1.1: Ranunculo caucasici-Bromopsietum variegatae (Figure 8A) Diagnostic species: Alchemilla sericea, Arenaria blepharophylla aggr., Artemisia chamaemelifolia, Aster alpinus, Avenula pubescens, Brachypodium pinnatum, Bromopsis variegata, Campanula collina, C. stevenii, Carex caryophyllea, Cirsium leucocephalum, Schedonorus pratensis, Galium cordatum, Huynhia pulchra, Lomelosia caucasica, Luzula multiflora, Medicago papillosa, Muscari armeniacum, Myosotis alpestris, Pedicularis condensata, Phascum cuspidatum (B), Phleum alpinum, Pilosella officinarum aggr., Pimpinella saxifraga aggr., Plantago atrata, Polygonum cognatum, Psephellus xanthocephalus, Pulsatilla albana, Ranunculus caucasicus, Rumex acetosella, Stachys macrantha, Stipa tirsa, Tragopogon graminifolius, Trifolium trichocephalum, Veronica gentianoides. Structure, ecology and distribution: We sampled this vegetation type at the steep north-facing slopes at elevations around 2,100 m, mainly in the Shirak province (vicinities of Amasia and Zorakert), and also in Gegharkunik province (Ardanish). The localities were characterised by a high mean annual precipitation (around 700–900 mm). The soil reaction was slightly acidic (mean pH: 6.5). The association differed by high species richness and the highest participation of Caucasian species, e.g. Dianthus cretaceus, Lomelosia caucasica, Stachys macrantha, and Trifolium trichocephalum, among all studied communities. Graminoids were dominant, particularly Brachypodium pinnatum, Bromopsis variegata, Carex humilis, and Phleum alpinum. 2.1.1.2: Tragopogon reticulatus-Astracantha aurea community (Figure 8B) Diagnostic species: Arenaria dianthoides, A. gypsophiloides, Astracantha aurea, Campanula stevenii, Elytrigia repens, Gagea glacialis, Koeleria albovii, Lathyrus digitatus, Papaver orientale, Plantago atrata, Senecio pseudo-orientalis, Tragopogon reticulatus, Trifolium ambiguum, T. spadiceum, Trisetum flavescens, Verbascum speciosum. Structure, ecology and distribution: We sampled this community mainly in the Gegharkunik province (Selim pass), and also in the Aragatsotn province (near the fortress of Amberd). Most of the localities were situated at 2,300–2,400 m a.s.l. and represented the highest elevations among all studied sites. The soil reaction was slightly acidic. These communities are characterised by low species richness with a high participation of Transcaucasian and Caucasian species, e.g. Arenaria dianthoides, Astracantha aurea, Koeleria albovii. Festuca ovina aggr. and Plantago atrata were the dominant species. Order 2.2: Onobrychido transcaucasicae-Stipetalia pulcherrimae – Transcaucasian mountain steppes Diagnostic species: Artemisia absinthium, Bupleurum falcatum aggr., Campanula glomerata aggr., C. rapunculoides, Cerinthe minor, Dactylis glomerata, Euphrasia pectinata, Galium verum, Globularia trichosantha, Helictochloa armeniaca, Hypericum perforatum, Klasea radiata, Linum nervosum, L. tenuifolium, Lotus corniculatus, Nepeta nuda, Onobrychis transcaucasica, Origanum vulgare, Phlomis tuberosa, Polygala anatolica, Rosa spinosissima, Salvia verticillata, Scabiosa bipinnata, Securigera varia, Stachys macrostachys, S. recta, Stipa pennata, S. pulcherrima, Teucrium chamaedrys, Thalictrum minus, Vicia canescens subsp. variegata, Viola ambigua. Ecology and distribution: Mountain steppes in the Transcaucasus form a distinct altitudinal belt above highland xerophyte vegetation (Cousinio brachypterae-Stipetalia arabicae) and below mountain meadow steppes (Plantagini atratae-Bromopsietalia variegatae). In our dataset, the order is represented by one alliance. Alliance 2.2.1: Onobrychido transcaucasicae-Stipion pulcherrimae – Transcaucasian mountain steppes Diagnostic species: Campanula rapunculoides, Cerinthe minor, Dactylis glomerata, Linum nervosum, L. tenuifolium, Onobrychis transcaucasica, Origanum vulgare, Scabiosa bipinnata, Securigera varia, Stachys macrostachys, S. recta. Denys Vynokurov et al.: Dry grasslands of Armenia58 Ecology and distribution: Despite its high floristic heterogeneity, we unite all mountain steppes into one alliance. We distinguish two informal communities and two associations. 2.2.1.1: Trisetum flavescens-Stachys macrostachys community (Figure 8C) Diagnostic species: Arenaria graminea, Artemisia absinthium, Chaerophyllum roseum, Gagea caroli-kochii, Silene cephalantha, Stachys macrostachys, Trisetum flavescens, Verbascum cheiranthifolium. Structure, ecology and distribution: We sampled this vegetation type at the elevations 1,950–2,300 m a.s.l. in Aragatsotn (near Amberd fortress), Gegharkunik (Selim pass, Shorja) and Shirak (vicinity of Amasia) provinces. These communities develop on soils with high humus content. The herb layer is relatively dense with dominance of grasses (Elytrigia intermedia aggr., Phleum nodosum, Poa pratensis aggr., Trisetum flavescens) and legumes (Securigera varia, Trifolium alpestre, Vicia tenuifolia subsp. variabilis, V. canescens subsp. variegata). Figure 7. Dry grassland and thorn-cushion communities of Armenia that do not belong to the Festuco-Brometea. A. Euphorbia orientalis-Melilotus officinalis scree community near Hermon (Vayots Dzor Province); B. Stachys lavandulifolia-Astracantha condensata community (alliance 1.1.1); C. Marrubio parviflorae-Stipetum capillatae (alliance 1.1.1); D. Semi-deserts of the association Noaeo mucronatae-Artemisietum fragrantis (alliance 1.1.2); E. Highland xerophytic vegetation of the Acantholimono caryophyllacei-Stipetum holosericeae (alliance 1.1.3); F. Stachys inflata-Acantholimon vedicum community (alliance 1.1.3) (Photos: A, C, E: Jürgen Dengler; B: Thomas Becker; D: Denys Vynokurov; F: Dariia Borovyk). Vegetation Classification and Survey 59 2.2.1.2: Onobrychis transcaucasica-Vicia canescens subsp. variegata community (Figure 8D) Diagnostic species: Arabis hirsuta, Campanula bononiensis, Campyliadelphus chrysophyllus (B), Chaerophyllum aureum, Daphne oleoides subsp. kurdica, Helictochloa armeniaca, Klasea radiata, Lathyrus latifolius, Linum nervosum, Nepeta nuda, Onobrychis transcaucasica, Origanum vulgare, Phlomis tuberosa, Primula veris subsp. macrocalyx, Rhinanthus subulatus, Salvia nemorosa, Securigera varia, Seseli libanotis, Stachys macrostachys, Stipa zalesskii subsp. canescens, Thalictrum minus, Valeriana officinalis aggr., Vicia canescens subsp. variegata, Viola ambigua. Structure, ecology and distribution: We recorded relevés of this community only in the Vayots Dzor province, at north-facing slopes (inclination 10–40°) in the vicinity of Gnishik and between Khachik and Areni. Communities differed by closed herb layer and high litter cover. The species composition is characterised by a high participation of forbs with European distribution, such as Campanula bononiensis, Klasea radiata, and Securigera varia. The dominant species is Vicia canescens subsp. variegata. 2.2.1.3: Globulario trichosanthae-Stipetum pulcherrimae (Figure 8E) Diagnostic species: central association (no diagnostic species) Structure, ecology and distribution: We sampled this vegetation type at elevations around 1,900–2,200 m a.s.l. in the provinces of Gegharkunik (Ardanish and vicinity of the town of Sevan), Shirak (Jajur pass) and Vayots Dzor (vicinities of Gnishik, Khachik and Areni). In the species composition, prevailing groups of species were European (Potentilla recta aggr., Stachys recta, Stipa pulcherrima) and Irano-Turanian (Eryngium billardierei, Onobrychis michauxii, Thymus kotschyanus, Ziziphora clinopodioides), followed by Caucasian endemics (Astragalus cancellatus, Centaurea pseudoscabiosa, Scabiosa bipinnata). The dominant species of the association was Stipa pulcherrima. 2.2.1.4: Seslerio phleoidis-Onobrychidetum cornutae (Figure 8F) Diagnostic species: Abietinella abietina (B), Adonis volgensis, Androsace chamaejasme, Asperula affinis, Asphodeline taurica, Briza media, Campanula rapunculoides, C. sibirica, Coronilla coronata, Euphorbia esula aggr., Euphrasia pectinata, Fritillaria caucasica, Homalothecium lutescens (B), Hypnum cupressiforme (B), Leucanthemum vulgare, Linum tenuifolium, Pimpinella saxifraga aggr., Pinus sylvestris, Pontechium maculatum, Psephellus karabaghensis, Sesleria phleoides, Spiraea crenata, Thalictrum foetidum, Viola alba. Structure, ecology and distribution: We sampled this association at elevations 1,940–2,070 m a.s.l. in Gegharkunik (Ardanish, Shorja, vicinity of the town of Sevan) and Shirak (Jajur pass) provinces. The association differed by the highest mean total species richness and richness of bryophytes across all studied communities. The species composition is represented by a high participation of European species of grasses and forbs, such as Briza media, Campanula rapunculoides, Galium verum, Pimpinella saxifraga aggr., and Stachys recta. Carex humilis, Onobrychis cornuta and Teucrium chamaedrys are the dominant species. Differentiation of the syntaxa with respect to ecology, structure and biodiversity Topography, climate and soil Communities within the Festuco-Brometea predominantly thrived at higher elevations, especially those of Plantagini atratae-Bromopsietalia variegatae, reaching up to 2,400 m a.s.l. (Figure 9A). Generally, they occurred within the elevation range of 1,900–2,400 m a.s.l., with some exceptions, like in the case of the Onobrychis transcaucasica-Vicia canescens subsp. variegata community, which were found between 1,700 to 2,100 m a.s.l. These low elevation occurrences were compensated by local topographic preferences, particularly avoidance of drier slopes with a southerly aspect. This community exhibited the lowest southerness index among all syntaxa (Figure 9B). In contrast, communities of the Ziziphora tenuior-Stipa arabica grasslands with its order Cousinio brachypterae-Stipetalia arabicae occurred at lower elevations, primarily below 2,000 m a.s.l. Among its three alliances, Artemision fragrantis, which comprises wormwood semi-deserts, thrived at the lowest altitudes, ranging from 1,300 to 1,600m a.s.l. Additionally, the Stachys inflata-Acantholimon vedicum community, which belongs to the alliance Acantholimono caryophyllacei–Stipion holosericeae, also occupied comparably low elevations, at about 1,600m a.s.l. Other units of this order generally occupy elevations not exceeding 2,000 m a.s.l. The altitudinal zonation reflected the climatic preferences of the communities. The Cousinio brachypterae-Stipetalia arabicae communities tended to prefer warmer and drier conditions, while the class Festuco-Brometea thrived in more mesic and cooler environments (Figure 9C, D). Regarding soil characteristics, most communities preferred neutral or slightly alkaline soils, except for the order Plantagini atratae-Bromopsietalia variegatae, which occurred in slightly acidic soil conditions with a pH around 6.5 (Figure 10A). This variation aligns with the altitudinal zonation, reflecting different soil compositions and vegetation types across elevation levels. Lower and medium elevations (montane and lower subalpine belt) were characterised by kastanozems and chernozems, which hosted semi-desert and steppe vegetation, and had neutral or slightly alkaline reaction. We classified them within the orders Cousinio brachypterae-Stipetalia arabicae and Onobrychido transcaucasicae-Stipetalia pulcherrimae respectively. Upper subalpine and lower alpine belts were home to meadow-steppe soils and mountain-meadow soils, which were characterised by slightly acidic to acidic reaction. Here, mountain meadow steppe vegetation thrived, which we summarised in the order Plantagini atratae-Bromopsietalia variegatae. At even higher altitudes, mountain steppes were replaced by acidophilous alpine grasslands, which were not covered in our study. Denys Vynokurov et al.: Dry grasslands of Armenia60 In terms of skeleton content, communities of the Ziziphora tenuior-Stipa arabica grasslands generally tended to occur on sites with shallower soils and higher skeleton content (Figure 10B, C). Among them, Stachys lavandulifolia-Astracantha condensata community and Stachys inflata-Acantholimon vedicum community represent rocky grasslands, and occurred on sites with the highest proportion of gravel in the upper soil level, reaching up to 80% and 90% respectively. Among the communities classified into the class Festuco-Brometea, we observed higher heterogeneity. Both units of the order Plantagini atratae-Bromopsietalia variegaFigure 8. Dry grasslands of Armenia classified within the Festuco-Brometea. A. Mountain meadow steppes in the Lake Arpi National Park with the association Ranunculo caucasici-Bromopsietum variegatae (alliance 2.1.1); B. Tragopogon reticulatus-Astracantha aurea community (alliance 2.1.1); C. Trisetum flavescens-Stachys macrostachys community (alliance 2.2.1); D. Onobrychis transcaucasica-Vicia canescens subsp. variegata community (alliance 2.2.1); E. Globulario trichosanthae-Stipetum pulcherrimae (alliance 2.2.1); F. mountain steppes near the Sevan Lake with the association Seslerio phleoidis-Onobrychidetum cornutae (alliance 2.2.1). (Photos: A: Philipp Kirschner; B, C: Dariia Borovyk; D, F: Jürgen Dengler; E: Denys Vynokurov). Vegetation Classification and Survey 61 tae occurred on sites with low gravel content. Among them, Ranunculo caucasici-Bromopsietum variegatae was distributed on deeper soils. Mountain steppes of the order Onobrychido transcaucasicae-Stipetalia pulcherrimae seemed to occur on sites with varying skeleton proportion. While the Trisetum flavescens-Stachys macrostachys and Onobrychis transcaucasica-Vicia canescens subsp. variegata communities occurred on sites with low gravel proportions, the associations Globulario trichosanthae-Stipetum pulcherrimae and Seslerio phleoidis-Onobrychidetum cornutae were distributed mainly on rocky sites. All communities of the class Festuco-Brometea tended to occur on sites with high humus content, with mean values within a narrow range of 7.5–8.1%. The tentative Ziziphora tenuior-Stipa arabica grasslands occurred on sites with significantly lower humus content, with mean values of 3.9–4.5% (Figure 10D). Figure 9. Site characteristics of the grassland types at different syntaxonomic levels. Box plots (median, interquartile range, range and outliers) as well as arithmetic means (black points) are shown. For the codes of syntaxa, see Table 1. Different letters within one syntaxonomic level indicate significant differences at p < 0.05 according to Tukey’s test following a significant ANOVA. For orders, the two-digit codes after the slashes and the letters below the boxplots apply. Figure 10. Soil characteristics of the grassland types at different syntaxonomic levels. Box plots (median, interquartile range, range and outliers) as well as arithmetic means (black points) are shown. For the codes of syntaxa, see Table 1. Different letters within one syntaxonomic level indicate significant differences at p < 0.05 according to Tukey’s test following a significant ANOVA. For orders, the two-digit codes after the slashes and the letters below the boxplots apply. Denys Vynokurov et al.: Dry grasslands of Armenia62 Structure and species composition Herb layer cover and litter cover, proxies for ecosystem productivity, were notably higher in the communities belonging to the class Festuco-Brometea (Figure 11A, B). Regarding the dominant life forms, we observed a strong differentiation between the two classes. Communities belonging to the class Festuco-Brometea had a significantly higher proportion of hemicryptophytes, which was highest in the case of the Ranunculo caucasici-Bromopsietum variegatae association (Figure 11D). On the contrary, communities of the tentative new class ‘Ziziphora tenuior-Stipa arabica grasslands’ had much lower proportions of hemicryptophytes, but a much higher proportion of therophytes (Figure 11C). The proportion of therophytes was particularly high in the case of the alliances Artemision fragrantis and Acantholimono caryophyllacei-Stipion holosericeae, which represent wormwood desert steppes and xerophytic thorn-cushions respectively. The proportion of chamaephytes was also significantly higher in the case of the latter class (not shown). The proportion of species’ range types also showed a strong differentiation between the two classes (Figure 12). The class Festuco-Brometea was characterised by a significantly higher proportion of species with European distribution (Figure 12A). This suggests that the class Festuco-Brometea comprises exclusively Euro-Siberian steppe vegetation and that its distribution range reaches a limit in Armenia. This is also well reflected by the co-occurrence of two biogeographic regions: the Euro-Siberian and the Irano-Turanian. On the contrary, the new tentative class ‘Ziziphora tenuior-Stipa arabica grasslands’ represents communities with a lower presence of species with European distribution, and significantly higher proportions of species with Mediterranean and Irano-Turanian distribution (Figure 12B, E). The presence of endemic species in the studied communities was also remarkable. While there were no significant differences detected between the two classes concerning narrow Transcaucasian endemics (Figure 12D), a distinct pattern emerged for the broader Caucasian endemics. These species, ranging across the Caucasus region, including the North Caucasus, were more prevalent in the Festuco-Brometea compared to the Ziziphora tenuior-Stipa arabica grasslands (Figure 12C). Among the two recognized orders of the Festuco-Brometea, the proportion of Caucasian endemics was higher in the Plantagini atratae-Bromopsietalia variegatae, which comprises mountain meadow steppes of higher elevations. Overall, we observed an increasing proportion of endemic species alongside elevation. Interestingly, the proportion of species with other distribution ranges, namely, broader than the listed above, was also significantly higher in the case of the Festuco-Brometea plant communities (Figure 12F). Plot-scale species richness Total species richness in 10 m2 did not differ significantly among the higher syntaxa (Figure 13). At the association level, there was no strong differentiation either, with only the Ranunculo caucasici-Bromopsietum variegatae and the Seslerio phleoidis-Onobrychidetum cornutae being above average and Tragopogon reticulatus-Astracantha aurea community below. For vascular plant species richness, the pattern largely matched that of total species richness, while for lichens there were no significant differences at any level. Only bryophytes showed a weak richness pattern at the order and alliance level, with the monotypic order Plantagini atratae-Bromopsietalia variegatae being the richest. Figure 11. Structure and dominant life forms of the grassland types at different syntaxonomic levels. Box plots (median, interquartile range, range and outliers) as well as arithmetic means (black points) are shown. For the codes of syntaxa, see Table 1. Different letters within one syntaxonomic level indicate significant differences at p < 0.05 according to Tukey’s test following a significant ANOVA. For orders, the two-digit codes after the slashes and the letters below the boxplots apply. Vegetation Classification and Survey 63 Figure 12. Proportions of range types of the grassland types at different syntaxonomic levels. Box plots (median, interquartile range, range and outliers) as well as arithmetic means (black points) are shown. For the codes of syntaxa, see Table 1. Different letters within one syntaxonomic level indicate significant differences at p < 0.05 according to Tukey’s test following a significant ANOVA. For orders, the two-digit codes after the slashes and the letters below the boxplots apply. Figure 13. Species richness in 10 m2-plots for different taxonomic groups compared at different syntaxonomic levels. Box plots (median, interquartile range, range and outliers) as well as arithmetic means (black points) are shown. For the codes of syntaxa, see Table 1. Different letters within one syntaxonomic level indicate significant differences at p < 0.05 according to Tukey’s test following a significant ANOVA. For orders, the two-digit codes after the slashes and the letters below the boxplots apply. Denys Vynokurov et al.: Dry grasslands of Armenia70 Quinn GP, Keough MJ (2002) Experimental design and data analysis for biologists. Cambridge University Press, Cambridge, UK, 537 pp. https://doi.org/10.1017/CBO9780511806384 Roleček J, Tichý L, Zelený D, Chytrý M (2009) Modified TWINSPAN classification in which the hierarchy respects cluster heterogeneity. Journal of Vegetation Science 20: 596–602. https://doi.org/10.1111/ j.1654-1103.2009.01062.x Sabatini FM, Jiménez-Alfaro B, Jandt U, Chytrý M, Field R, Kessler M, Lenoir J, Schrodt F, Wiser SK, … Bruelheide H (2022) Global patterns of vascular plant alpha diversity. Nature Communications 13: e4683. https://doi.org/10.1038/s41467-022-32063-z Schaminée JHJ, Stortelder AHF, Westhoff V (1995) (Eds) De vegetatie van Nederland – Deel 1. Inleiding tot de plantensociologie – grondslagen, methoden en toepassingen [The vegetation of the Netherlands – Part 1. Introduction to plant sociology – foundations, methods and applications]. Opulus, Uppsala, SE, 296 pp. [In Dutch] Stojilkovič V, Záveská E, Frajman B (2022) From Western Asia to the Mediterranean Basin: diversification of the widespread Euphorbia nicaeensis alliance (Euphorbiaceae). Frontiers in Plant Science 13: e815379. https://doi.org/10.3389/fpls.2022.815379 Sukachev VN (1928) Rastitelnye soobschestva (Vvedenie v fitosotsiologiiu) [Vegetation communities (Introduction to phytosociology)]. Kniga, Leningrad, RU. [In Russian] Takhtajan A (1941) Botanico-geographicheskij ocherk Armenii [Phytogeographic survey of Armenia]. Proc. Inst. Bot. Armen. branch USSR Acad. Sci 2: 3–180. Takhtajan A (1986) Floristic regions of the world. University of California Press, Berkley and Los Angeles, CA, US, 522 pp. Tatlı A (1991) Phytosociological investigation of vegetation of wind-eroded areas of Eastern Anatolia. Ecologia Mediterranea 17: 161–168. https://doi.org/10.3406/ecmed.1991.1697 ter Braak JFC, Šmilauer P (2012) Canoco 5. Canoco reference manual and user´s guide. Software for ordination (version 5). Biometrics, Wageningen, Česke Budějovice, NL, CZ, 496 pp. Theurillat J, Willner W, Fernández‐González F, Bültmann H, Čarni A, Gigante D, Mucina L, Weber H (2021) International Code of Phytosociological Nomenclature. 4th edition. Applied Vegetation Science 24: e12491. https://doi.org/10.1111/avsc.12491 Tichý L (2002) JUICE, software for vegetation classification. Journal of Vegetation Science 13: 451–453. https://doi. org/10.1111/j.1654-1103.2002.tb02069.x Tichý L, Chytrý M (2006) Statistical determination of diagnostic species for site groups of unequal size. Journal of Vegetation Science 17: 809–818. https://doi.org/10.1111/j.1654-1103.2006.tb02504.x Török P, Dengler J (2018) Palaearctic grasslands in transition: Overarching patterns and future prospects. In: Squires VR, Dengler J, Feng H, Limin H (Eds) Grasslands of the World: Diversity, Management and Conservation. CRC Press, Boca Raton, FL, US, 15–26. Tsepkova NL (1987) O novyih assotsiatsiyah travyanistoy rastitelnosti gornyih regionov Tsentral-nogo Kavkaza [About new associations of grassland vegetation in the mountain regions of the Central Caucasus]. Redkollegiya zhurnala «Biologicheskie nauki». Dep. v VINITI 11.05.87, № 3361-В87, Moscow, RU, 15 pp. [in Russian] Tsepkova NL (2005) K sintaxonomii vysokogornykh lugovykn stepey Tsentralnogo Kavkaza (Rossia) [Syntaxonomy of the high-mountain meadow steppes of the Central Caucasus (Russia)]. Vegetation of Russia 7: 93–96. https://doi.org/10.31111/vegrus/2005.07.93 [In Russian] Večeřa M, Axmanová I, Chytrý M, Divíšek J, Ndiribe C, Velasco Mones G, Čeplová N, Aćić S, Bahn M, … Lososová Z (2023) Decoupled phylogenetic and functional diversity in European grasslands. Preslia 95: 413–445. https://doi.org/10.23855/preslia.2023.413 Vynokurov D, Didukh Y, Krasova O, Lysenko H, Goncharenko I, Dmytrash-Vatseba I, Chusova O, Shyriaieva D, Kolomiychuk V, Moysiyenko I (2020) Eastern European Steppe Database. Vegetation Classification and Survey 1: 149–150. https://doi.org/10.3897/ VCS/2020/60520 Vynokurov D, Lysenko T, Dutova Z, Shylnikov D, Doroshina G, Urbanavichene I, Urbanavichus G, Tsepkova N (2021) The dry grasslands (Festuco-Brometea) of the North Caucasus: first data on numerical classification and biodiversity patterns. Tuexenia 41: 175–201. https://doi.org/10.14471/2021.41.004 Willner W, Faber-Langendoen D (2021) Braun-Blanquet meets EcoVeg: a formation and division level classification of European phytosociological units. Vegetation Classification and Survey 2: 275–291. https://doi.org/10.3897/VCS/2021/71299 Willner W, Kuzemko A, Dengler J, Chytrý M, Bauer N, Becker T, Biţă-Nicolae C, Botta-Dukát Z, Čarni A, … Janišová M (2017) A higher-level classification of the Pannonian and western Pontic steppe grasslands (Central and Eastern Europe). Applied Vegetation Science 20: 143–158. https://doi.org/10.1111/avsc.12265 Willner W, Roleček J, Korolyuk A, Dengler J, Chytrý M, Janišová M, Lengyel A, Aćić S, Becker T, … Yamalov S (2019) Formalized classification of semi-dry grasslands in central and eastern Europe. Preslia 91: 25–49. https://doi.org/10.23855/preslia.2019.025 Ziroyan AN (1989) Fitotsenoticheskaya kharakteristika i produktivnost osnovnykh tipov rastitelnosti Armenii [Phytocoenotic characteristics and productivity of the main vegetation types in Armenia]. Summary of the Dr. habil. thesis. Academy of Sciences of Georgian SSR, N.N. Ketshoveli Institute of Botany, Tbilisi, GE, 43 pp. [In Russian] Zhang J, Gillet F, Bartha S, Alatalo JM, Biurrun I, Dembicz I, Grytnes J, Jaunatre R, Pielech R, … Dengler J (2021) Scale dependence of species–area relationships is widespread but generally weak in Palaearctic grasslands. Journal of Vegetation Science 32: e13044. https://doi. org/10.1111/jvs.13044 Zohary M (1973) Geobotanical foundations of the Middle East. Volume 2. G. Fischer, Stuttgart, DE, 739 pp. Appendix 1 Formal descriptions of the new syntaxa according to the ICPN For the diagnostic species we refer to the main text as well as Table 2 and Suppl. material 2. 1.1 Cousinio brachypterae-Stipetalia arabicae ord. nov. hoc loco Holotypus hoc loco: Artemision fragrantis Vynokurov et al. 2024 (this paper) 2.1 Plantagini atratae-Bromopsietalia variegatae ord nov. hoc loco Holotypus hoc loco: Artemisio chamaemelifoliae-Bromopsion variegatae Vynokurov in Vynokurov et al. 2021 (page 186) Vegetation Classification and Survey 71 2.2 Onobrychido transcaucasicae-Stipetalia pulcherrimae ord. nov. hoc loco Holotypus hoc loco: Onobrychido transcaucasicae-Stipion pulcherrimae Vynokurov et al. 2024 (this paper) 1.1.1 Onobrychido michauxii-Stipion capillatae all. nov. hoc loco Holotypus hoc loco: Marrubio parviflorae-Stipetum capillatae Vynokurov et al. 2024 (this paper) 1.1.2 Artemision fragrantis all. nov. hoc loco Holotypus hoc loco: Noaeo mucronatae-Artemisietum fragrantis Vynokurov et al. 2024 (this paper) 1.1.3 Acantholimono caryophyllacei-Stipion holosericeae all. nov. hoc loco Holotypus hoc loco: Acantholimono caryophyllacei-Stipetum holosericeae Vynokurov et al. 2024 (this paper) 2.2.1 Onobrychido transcaucasicae-Stipion pulcherrimae all. nov. hoc loco Holotypus hoc loco: Seslerio phleoidis-Onobrychidetum cornutae Vynokurov et al. 2024 (this paper) 1.1.1.2 Marrubio parviflorae-Stipetum capillatae ass. nov. hoc loco Holotypus hoc loco: plot ID 81 in Suppl. material 2 (original code AMR008); Armenia, Lori Province, vicinity of Shirakamut, 40.85801°N, 44.19223°E, altitude: 1,662 m a.s.l., aspect: 210°, inclination: 40°, 27 June 2019, authors of the relevé: Dieter Frank, Salza Palpurina, Denys Vynokurov. Floristic composition (species sorted by their cover in percent): Elytrigia intermedia aggr. 10, Marrubium parviflorum 10, Achillea arabica 5, Teucrium polium 5, Onobrychis michauxii 4, Artemisia austriaca 3, Thymus sipyleus 3, Medicago x varia 2, Cota melanoloma 1, Festuca valesiaca aggr. 1, Iris pumila 1, Isatis steveniana 1, Psephellus zuvandicus 1, Xeranthemum longepapposum 1, Ajuga chamaepitys subsp. chia 0.5, Asperula arvensis 0.5, Dactylis glomerata 0.5, Odontarrhena muralis 0.5, Stachys recta 0.5, Viola ambigua 0.5, Holosteum umbellatum 0.3, Nonea pulla 0.3, Scleranthus annuus 0.3, Convolvulus lineatus 0.2, Euphorbia condylocarpa 0.2, Stipa capillata 0.2, Zeravschania pauciradiata 0.2, Centaurea ovina aggr. 0.1, Salvia nemorosa 0.1, Stipa arabica 0.1, Thesium arvense 0.1, Falcaria vulgaris 0.01, Meniocus linifolius 0.01, Nepeta racemosa 0.01, Noccaea perfoliata 0.01, Reseda lutea 0.01, Veronica multifida 0.01, Viola arvensis aggr. 0.01. 1.1.2.1 Noaeo mucronatae-Artemisietum fragrantis ass. nov. hoc loco Holotypus hoc loco: plot ID 65 in Suppl. material 2 (original code AM11NE); Armenia, Aragatsotn Province, vicinity of Dashtadem, 40.32428°N, 43.85075°E, altitude: 1,357 m a.s.l., aspect: 265°, inclination: 7°, 29 June 2019, authors of the relevé: Alla Aleksanyan, Jürgen Dengler, Denys Vynokurov. Floristic composition (species sorted by their cover in percent): Poa bulbosa 20, Artemisia fragrans 17, Peganum harmala 10, Taeniatherum caput-medusae subsp. crinitum 10, Stipa arabica 1, Androsace albana 0.5, Koelpinia linearis 0.5, Stipa x kolakovskyi 0.5, Allium pseudoflavum 0.2, Crupina vulgaris 0.2, Minuartia hamata 0.2, Noaea mucronata 0.2, Arenaria serpyllifolia aggr. 0.1, Alyssum turkestanicum 0.1, Astragalus hyalolepis 0.1, Cousinia brachyptera 0.1, Erysimum gelidum 0.1, Odontarrhena tortuosa 0.1, Sclerocaryopsis spinocarpos 0.1, Syntrichia caninervis 0.1, Xeranthemum squarrosum 0.1, Aegilops cylindrica 0.01, Ceratocephala falcata 0.01, Cota triumfettii 0.01, Dianthus crinitus 0.01, Draba verna 0.01, Euphorbia seguieriana 0.01, Helichrysum plicatum 0.01, Hohenackeria exscapa 0.01, Holosteum umbellatum 0.01, Linaria simplex 0.01, Medicago medicaginoides 0.01, Meniocus linifolius 0.01, Minuartia meyeri 0.01, Polygala hohenackeriana 0.01, Thesium szovitsii 0.01, Thymus collinus 0.01. 1.1.3.1 Acantholimono caryophyllacei-Stipetum holosericeae ass. nov. hoc loco Holotypus hoc loco: plot ID 71 in Suppl. material 2 (original code AMR035); Armenia, Vayots Dzor Province, vicinity of Khachik, 39.659529°N, 45.201325°E, altitude: 2,017 m a.s.l., aspect: 315°, inclination: 9°, 4 July 2019, authors of the relevé: Alla Aleksanyan, Idoia Biurrun, Dariia Borovyk. Floristic composition (species sorted by their cover in percent): Chardinia orientalis 35, Acantholimon caryophyllaceum 25, Taeniatherum caput-medusae subsp. crinitum 20, Poa bulbosa 15, Anisantha tectorum 10, Syntrichia ruralis 6, Eryngium billardierei 5, Onobrychis cornuta 3, Medicago sativa 2, Stipa ehrenbergiana 0.5, Stipa holosericea 2, Teucrium polium 2, Achillea arabica 0.5, Convolvulus lineatus 0.5, Gagea germainae 0.5, Xeranthemum squarrosum 0.3, Ziziphora capitata 0.3, Centaurea virgata 0.2, Hypericum scabrum 0.2, Rochelia disperma 0.2, Alyssum turkestanicum 0.1, Asperula arvensis 0.1, Bromus danthoniae 0.1, Bromus japonicus 0.1, Ceratocephala falcata 0.1, Crepis sancta 0.1, Euphorbia iberica 0.1, Hordeum bulbosum 0.1, Marrubium parviflorum 0.1, Odontarrhena tortuosa 0.1, Potentilla recta aggr. 0.1, Thymus kotschyanus 0.1, Alyssum alyssoides 0.01, Arenaria serpyllifolia aggr. 0.01, Caucalis platycarpos 0.01, Holosteum umbellatum 0.01, Minuartia meyeri 0.01, Noccaea perfoliata 0.01, Veronica denudata 0.01. 2.1.1.1 Ranunculo caucasici-Bromopsietum variegatae ass. nov. hoc loco Holotypus hoc loco: plot ID 18 in Suppl. material 2 (original code AM06NW); Armenia, Shirak Province, vicinity of Zorakert, 41.10135°N, 43.66933°E, altitude: 2,114 m a.s.l., aspect: 255°, inclination: 35°, 28 June 2019, authors of the relevé: Elena Belonovskaya, Jürgen Dengler, Denys Vynokurov. Floristic composition (species sorted by their cover in percent): Thymus sipyleus 25, Achillea millefolium aggr. 15, Phleum alpinum 15, Koeleria macrantha 12, Festuca ovina aggr. 10, Astragalus incertus 8, Carex humilis 8, Trifolium ambiguum 7, Bromopsis variegata 5, Hypnum cupressiforme 5, Lotus corniculatus 5, Medicago falcata 5, Abietinella abietina 2, Myosotis alpestris 2, Poa pratensis aggr. 2, Bryum caespiticium 1, Plantago atrata 1, Securigera varia 1, Weissia brachycarpa 1, Avenula pubescens 0.5, Streblotrichum convolutum 0.5, Cirsium leucocephalum 0.5, Gelasia rigida 0.5, Huynhia pulchra 0.5, Lomelosia caucasica 0.5, Rumex acetosella 0.5, Scabiosa bip- Denys Vynokurov et al.: Dry grasslands of Armenia72 innata 0.5, Stachys recta 0.5, Carex caryophyllea 0.3, Hypericum linarioides 0.3, Poa badensis 0.3, Potentilla argentea 0.3, Schedonorus pratensis 0.3, Taraxacum sect. Taraxacum 0.3, Teucrium orientale 0.3, Veronica denudata 0.3, Galium verum 0.2, Papaver monanthum 0.2, Pimpinella saxifraga aggr. 0.2, Potentilla recta aggr. 0.2, Ajuga orientalis 0.1, Alyssum alyssoides 0.1, Arenaria blepharophylla aggr. 0.1, Arenaria gypsophiloides 0.1, Artemisia absinthium 0.1, Bupleurum falcatum aggr. 0.1, Campanula stevenii 0.1, Cirsium obvallatum 0.1, Colchicum trigynum 0.1, Galium cordatum 0.1, Linaria schelkownikowii 0.1, Pedicularis condensata 0.1, Phascum cuspidatum 0.1, Ranunculus caucasicus 0.1, Scleranthus perennis 0.1, Silene bupleuroides 0.1, Silene chlorantha 0.1, Syntrichia montana 0.1, Bryum argenteum 0.01, Crepis alpina 0.01, Draba nemorosa 0.01, Encalypta vulgaris 0.01, Herniaria incana 0.01, Muscari armeniacum 0.01, Polygala alpicola 0.01, Polygonum cognatum 0.01, Sedum acre 0.01. 2.2.1.3 Globulario trichosanthae-Stipetum pulcherrimae ass. nov. hoc loco Holotypus hoc loco: plot ID 93 in Suppl. material 2 (original code AM42SE); Armenia, Vayots Dzor Province, vicinity of Gnishik, 39.675269°N, 45.30616°E, altitude: 2,136 m a.s.l., aspect: 80°, inclination: 23°, 3 July 2019, authors of the relevé: Asun Berastegi, Idoia Biurrun, Denys Vynokurov. Floristic composition (species sorted by their cover in percent): Ziziphora clinopodioides 15, Koeleria macrantha 7, Stipa capillata 7, Dactylis glomerata 5, Stipa pulcherrima 5, Hypericum scabrum 4, Onobrychis cornuta 4, Helichrysum graveolens 3, Lotus corniculatus 2, Odontarrhena tortuosa 2, Teucrium chamaedrys 2, Teucrium polium 2, Centaurea ovina aggr. 1, Daphne oleoides subsp. kurdica 1, Eryngium billardierei 1, Galium verticillatum 1, Galium verum 1, Linum tenuifolium 1, Medicago sativa 1, Stipa pennata 1, Cichorium intybus 0.5, Cruciata laevipes 0.5, Cuscuta epithymum 0.5, Euphorbia seguieriana 0.5, Hypericum linarioides 0.5, Plantago lanceolata 0.5, Scabiosa bipinnata 0.5, Securigera varia 0.5, Stipa tirsa 0.5, Tanacetum polycephalum subsp. argyrophyllum 0.5, Globularia trichosantha 0.3, Leontodon hispidus 0.3, Stachys lavandulifolia 0.3, Stachys recta 0.3, Ajuga chamaepitys subsp. chia 0.2, Cerinthe minor 0.2, Leontodon asperrimus 0.2, Linum nervosum 0.2, Phleum phleoides 0.2, Poa bulbosa 0.2, Tragopogon sosnowskyi 0.2, Verbascum chaixii subsp. austriacum 0.2, Astragalus cancellatus 0.1, Bupleurum falcatum aggr. 0.1, Carlina vulgaris 0.1, Sanguisorba minor 0.1, Silene bupleuroides 0.1, Thesium arvense 0.1, Trinia glauca 0.1, Veronica microcarpa 0.1, Achillea millefolium aggr. 0.01, Agrimonia eupatoria 0.01, Crepis pulchra 0.01, Euphrasia pectinata 0.01, Odontites aucheri 0.01, Polygala alpicola 0.01. 2.2.1.4 Seslerio phleoidis-Onobrychidetum cornutae ass. nov. hoc loco Holotypus hoc loco: plot ID 13 in Suppl. material 2 (original code AMR076); Armenia, Gegharkunik Province, vicinity of Shoghakat, 40.49547°N, 45.29575°E, altitude: 1,959 m a.s.l., aspect: 305°, inclination: 30°, 1 July 2019, authors of the relevé: George Fayvush, Salza Palpurina, Iulia Vasheniak. Floristic composition (species sorted by their cover in percent): Carex humilis 20, Onobrychis cornuta 15, Elytrigia intermedia aggr. 8, Primula veris subsp. macrocalyx 7, Thymus kotschyanus 7, Stipa pulcherrima 6, Onobrychis transcaucasica 5, Scutellaria orientalis aggr. 5, Galium verum 4, Inula aspera 4, Helianthemum nummularium 2, Sesleria phleoides 2, Teucrium chamaedrys 2, Abietinella abietina 1, Bromus scoparius 1, Campanula sibirica 1, Euphorbia iberica 1, Homalothecium lutescens 1, Hypnum cupressiforme 1, Linum tenuifolium 1, Lotus corniculatus 1, Medicago sativa 1, Syntrichia ruralis 1, Ziziphora clinopodioides 1, Achillea millefolium aggr. 0.5, Astracantha stenonychioides 0.5, Campanula rapunculoides 0.5, Dianthus cretaceus 0.5, Filipendula vulgaris 0.5, Linum nervosum 0.5, Plantago media 0.5, Polygala anatolica 0.5, Scabiosa bipinnata 0.5, Stachys recta 0.5, Thalictrum foetidum 0.5, Jurinea squarrosa 0.3, Viola ambigua 0.3, Phleum phleoides 0.2, Poa badensis 0.2, Psephellus karabaghensis 0.2, Tanacetum aureum 0.2, Convolvulus lineatus 0.1, Festuca valesiaca aggr. 0.1, Pontechium maculatum 0.1, Tragopogon reticulatus 0.1, Xanthoparmelia camtschadalis 0.1, Asperula prostrata 0.01, Carum caucasicum 0.01, Cetraria ericetorum 0.01, Euphrasia sevanensis 0.01, Hypericum scabrum 0.01, Orobanche alsatica 0.01, Teucrium orientale 0.01. E-mail and ORCID Denys Vynokurov ([email protected]), ORCID: https://orcid.org/0000-0001-7003-6680 Alla Aleksanyan (alla.alexan[email protected]m), ORCID: https://orcid.org/0000-0003-4073-1812 Thomas Becker ([email protected]), ORCID: https://orcid.org/0000-0002-7942-5575 Idoia Biurrun (idoia.b[email protected]), ORCID: https://orcid.org/0000-0002-1454-0433 Dariia Borovyk (dariiaborov[email protected]), ORCID: https://orcid.org/0000-0001-7140-7201 George Fayvush ([email protected]), ORCID: https://orcid.org/0000-0002-9710-2200 Itziar García-Mijangos ([email protected]), ORCID: https://orcid.org/0000-0002-6642-7782 Martin Magnes (ma[email protected]), ORCID: https://orcid.org/0000-0002-5453-5735 Salza Palpurina (salza.palpur[email protected]m), ORCID: https://orcid.org/0000-0003-0416-5622 Ute Becker ([email protected]) Asun Berastegi ([email protected]), ORCID: https://orcid.org/0000-0003-0456-3305 Beata Cykowska-Marzencka (beatacyko[email protected]), ORCID: https://orcid.org/0000-0002-5468-4909 Iwona Dembicz ([email protected]), ORCID: https://orcid.org/0000-0002-6162-1519 Dieter Frank ([email protected]), ORCID: https://orcid.org/0000-0002-8602-9570 Vegetation Classification and Survey 73 Andreas Hilpold (andreas.hil[email protected]), ORCID: https://orcid.org/0000-0003-0262-8080 Philipp Kirschner (philipp[email protected]), ORCID: https://orcid.org/0000-0002-9263-5112 Helmut Mayrhofer ([email protected]), ORCID: https://orcid.org/0000-0002-9433-3847 Marine Oganesian ([email protected]), ORCID: https://orcid.org/0000-0002-3297-8519 Iuliia Vasheniak (arrhenat[email protected]), ORCID: https://orcid.org/0000-0003-1020-3007 Jürgen Dengler (Corresponding author, [email protected]), ORCID: https://orcid.org/0000-0003-3221-660X Supplementary material Supplementary material 1 Complete header data of the Armenian plots (*.xlsx) Link: https://doi.org/10.3897/VCS.119253.suppl1 Supplementary material 2 Complete synoptic table of 110 classified Armenian plots with percent constancies and phi values for the syntaxa of all levels as well as the individual plots (*.xlsx) Link: https://doi.org/10.3897/VCS.119253.suppl2 Supplementary material 3 Definition of additional species aggregates (*.pdf) Link: https://doi.org/10.3897/VCS.119253.suppl3 Supplementary material 4 Value distribution of all recorded and analysed numerical environmental, structural and biodiversity variables (*.pdf) Link: https://doi.org/10.3897/VCS.119253.suppl4 Supplementary material 5 Data sources of the West Asian and Caucasian dataset (*.pdf) Link: https://doi.org/10.3897/VCS.119253.suppl5 Supplementary material 6 List of the vascular plants from the Armenian dataset with assignment of life forms and distribution ranges (*.pdf) Link: https://doi.org/10.3897/VCS.119253.suppl6 Supplementary material 7 Synoptic table of the West Asian and Caucasian dataset with the results of the broad-scale comparison with five distinguished groups of clusters (A–E) (*.pdf) Link: https://doi.org/10.3897/VCS.119253.suppl7 Supplementary material 8 Planned applications to the EuroVegChecklist Committee (EVCC) (*.pdf) Link: https://doi.org/10.3897/VCS.119253.suppl8