Disertační Práce

PŘÍRODOVĚDECKÁ FAKULTA

Disertační práce

Tomáš Peterka

Brno 2019

PŘÍRODOVĚDECKÁ FAKULTA

Variability of fen vegetation on the European scale Ph.D. Dissertation

Tomáš Peterka

Vedoucí práce: doc. Mgr. Michal Hájek, Ph.D.

Ústav botaniky a zoologie

Brno 2019 Bibliographic Entry

Author: Mgr. Tomáš Peterka Faculty of Science, Masaryk University Department of Botany and Zoology

Title of Dissertation: Variability of fen vegetation on the European scale

Degree programme: Biology

Field of Study: Botany

Supervisor: doc. Mgr. Michal Hájek, Ph.D.

Academic Year: 2018/2019

Number of Pages: 37+181

Keywords: databases; phytosociological relevés; environmental gradients; classification; wetlands; endangered habitats; pH; mires; plant communities; vegetation

Bibliografický záznam

Autor: Mgr. Tomáš Peterka Přírodovědecká fakulta, Masarykova univerzita Ústav botaniky a zoologie

Název práce: Variabilita slatiništní vegetace na evropském měřítku

Studijní program: Biologie

Studijní obor: Botanika

Školitel: doc. Mgr. Michal Hájek, Ph.D.

Akademický rok: 2018/2019

Počet stran: 37+181

Klíčová slova: databáze; fytocenologické snímky; gradienty prostředí; klasifikace; mokřady; ohrožené biotopy; pH; rašeliniště; rostlinná společenstva; vegetace

Abstract

Fens (minerotrophic mires, plant communities of the Scheuchzerio palustris-Caricetea fuscae class) have great importance for biodiversity protection and belong among endangered habitats in many parts of Europe. Phytosociological classification of fen vegetation at the level of alliances (i.e. major vegetation units), however, differs among European countries, which complicates communication among scientist as well as habitat protection at the supra-national level. Several vegetation systems accent the stand physiognomy and hydrological factors as the main classification criteria for delimitation of alliances. Other systems, mostly following the Scandinavian tradition, distinguish alliances along the “poor- rich” gradient (complex gradient of pH and the total mineral richness). The aim of the thesis is hence to made an attempt on the harmonised classification of European fens at the alliance level. Relationship between vegetation and environmental data was investigated in well- preserved fens on the Bohemian Massif (Czech Republic). This study revealed the key role of poor-rich gradient for species turnover in fens and for delimitation of fen alliances. The macronutrient availability (fertility) coincided with the second gradient, independent on pH and mineral richness. As data collecting and filtering are prerequisites for vegetation syntheses, the thesis focuses also on these methodical aspects. The European Mire Vegetation Database was introduced. This database contains thousands of previously not digitized vegetation plots and fills in regional gaps in digital data on mire vegetation. Further, the thesis deals with the question whether plots of different sizes might be jointly used in broad-scale vegetation analyses. Since plots of 1–25 m2 shared comparable counts of fen specialists and produced coincident ordinations patterns, they might be safely merged in broad-scale syntheses and analyses of fen vegetation without introducing significant error, at least when compared with other possible sources of bias. The pan-European classification of fen vegetation was proposed and supported by analysis of vegetation-plot data. The presented classification scheme reflects differences in species composition driven by site chemistry and geographical (macroclimatic) variation. Formal definitions of alliances were created using the presence, absence and abundance of sociological species groups and indicator species. The following alliances were defined: Caricion viridulo-trinervis, Caricion davallianae, Caricion atrofusco-saxatilis, Stygio-Caricion limosae, Sphagno warnstorfii-Tomentypnion nitentis, Saxifrago-Tomentypnion, Narthecion scardici, Caricion stantis, Anagallido tenellae-Juncion bulbosi, Drepanocladion exannulati, Caricion fuscae, Sphagno-Caricion canescentis and Scheuchzerion palustris. Unsupervised classification and ordination supported the ecological meaningfulness of presented classification. In some cases, pan-European vegetation synthesis pointed to the occurrence of plant communities previously overlooked in particular national vegetation surveys. For example, the vegetation of several quaking brown-moss rich fens occurring in the Carpathians and adjacent territories were identified as the boreal rich fen community of Stygio-Caricion limosae alliance. This classification was confirmed by comparison of plots from Central and South-eastern Europe with phytosociogical material from Northern Europe (the locus classicus of Stygio-Caricion limosae). Macrofossil data suggested much more frequent occurrence of this vegetation type in Central Europe in glacial and Early Holocene times and hence its current relict status. During the fieldwork, new localities of relict bryophytes Meesia triquetra and Pseudocalliergon lycopodioides for individual countries were discovered, and one locality of endangered Eriophorum gracile was re-discovered in Czech Republic. The last section of the thesis hence focuses on updating knowledge on distributional ranges and vegetation affinity of these important fen specialists.

Abstrakt

Slatiniště (minerotrofní rašeliniště, rostlinná společentsva třídy Scheuchzerio palustris- Caricetea fuscae) mají velký význam pro ochranu biodiverzity a v mnoha oblastech Evropy patří k ohroženým biotopům. Fytocenocelogické klasifikace slatiništní vegetace se však mezi evropskými zeměmi liší, což komplikuje nejen komunikaci ve vědecké komunitě, ale také ochranu těchto stanovišť na mezinárodní úrovni. Některé vegetační systémy slatiništní vegetace rozlišují hlavní vegetační jednotky (tj. svazy) na základě fyziognomie porostů a hydrologických faktorů. Jiné systémy navazující na skandinávskou tradici vymezují svazy podél tzv. „poor-rich“ gradientu (komplexního gradientu pH a minerální bohatosti). Cílem této práce je vytvořit první pokus o jednotnou klasifikaci slatinišť v Evropě. Na posledních zachovalých slatiništích v Českém masivu (ČR) jsme se zabývali vztahem vegetace a proměnných prostředí. Tato studie potvrdila význam poor-rich gradientu pro změnu druhového složení slatiništní vegetace a pro vymezení jednotlivých svazů. Druhý vegetační gradient souvisel s dostupností makroprvků (fertilitou) a projevoval se nezávisle na pH a minerální bohatosti. Nezbytnými předpoklady vegetačních syntéz jsou sběr a třídění dat, proto se disertační práce zabývá rovněž těmito metodickými aspekty. Představena je databáze evropské rašelinné vegetace (European Mire Vegetation Database), která obhahuje několik tisíc dosud nedigitalizovaných fytocenologických snímků. Dále se zabýváme otázkou, jestli mohou být ve velkoškálových analýzách použity snímky o různé velikosti plochy. Zápisy o rozměrech 1–25 m2 obsahují srovnatelný počet specialistů a dávají obdobné výsledky v ordinančích analýzách. Proto mohou fyt. snímky o této velikosti být zařazeny do velkoplošných vegetačních syntéz, aniž by způsobily významné zkreslení výsledků (obzvláště přihlédneme-li k jiným možným zdrojů šumu ve fytocenologických datech). Navrhli jsme klasifikaci slatiništní vegetace v Evropě na základě analýzy fytocenologických snímků. Použité klasifikační schéma odráží vliv chemismu a geografických (makroklimatických) proměnných na druhové složení společenstev. Vytvořili jsme formální definice svazů Caricion viridulo-trinervis, Caricion davallianae, Caricion atrofusco-saxatilis, Stygio-Caricion limosae, Sphagno warnstorfii-Tomentypnion nitentis, Saxifrago- Tomentypnion, Narthecion scardici, Caricion stantis, Anagallido tenellae-Juncion bulbosi, Drepanocladion exannulati, Caricion fuscae, Sphagno-Caricion canescentis a Scheuchzerion palustris. Formální definice pracují s přítomností, absencí a pokryvností sociologických skupin a význačných druhů. Ekologickou smysluplnost navrženého systému podpořily ordinační metody a metody neřízené klasifikace. V některých případech celoevropská syntéza poukázala na přítomnost vegetačních typů, které nebyly dosud rozlišovány v národních vegetačních přehledech. Např. mírně vápnitá třasoviska s hnědými mechy vyskytující se vzácně v Karpatech a okolních oblastech byla identifikována jako společenstva boreálního svazu Stygio-Caricion limosae. Toto zařazení jsme ověřili srovnáním fytocenologických snímků s materiálem ze severní Evropy (locus classicus svazu Stygio-Caricion limosae). Makrozbytky naznačují běžný výskyt tohoto vegetačního typu ve střední Evropě v glaciálu a raném holocénu a tedy jeho reliktní povahu. Během terénního výzkumu byly zjištěny nové lokality reliktních mechů (Meesia triquetra, Pseudocalliergon lycopodioides) a potvrzena lokalita ohroženého suchopýru štíhlého (Eriophorum gracile). Poslední část disertace proto doppuje informace o rozšíření těchto významných slatiništních specialistů a o jejich vegetační vazbě.

Na tomto místě bych chtěl poděkovat všem těm, bez jejichž pomoci nebo podpory bych tuto práci nesepsal a kteří mi během uplynulých let pomáhali poznávat přírodu a snad i trochu pochopit některé její zákonitosti.

Hlavní dík náleží Michalu Hájkovi za to, že mi vůbec navrhl jít na doktorát. Za jeho přátelství, ochotu, rady, lidskost, trpělivost, toleranci a občas i za poněkud nekritickou důvěru. Za čas, který mi tolikrát věnoval, ačkoliv žádný čas neměl. Za to, že nad svým svérázným doktorandem s neméně svéráznými názory a pohledem na svět nezlomil hůl, ačkoliv k tomu měl nejednou pádný důvod.

Dalším důležitým člověkem, bez kterého by disertace zřejmě nevznikla, je Martin Jiroušek. Díky svým postřehům, nápadům, pomoci při shánění vegetačních dat a mnoha společným terénům, na nichž jsem si mohl rozšířit obzory na poli ekologie rašelinišť, se Martin vlastně stal mým neoficiálním konzultantem.

Též Peťa Hájková mnohokrát ochotně přispěla radami, postřehy, nápady, determinačními zkušenostmi a cennými daty.

Během doktorátu by se člověk těžko obešel bez pomoci Evky Hette Šmerdové, dobré duše pracovní skupiny pro výzkum rašelinišť. Evky, která vždy splní, co slíbí, zařídí, co je potřeba zařídit, a dovede člověku poradit snad v každé situaci.

Jsou lidé, kteří vám hodně pomohou jen tím, že prostě jsou, že vysílají do okolí pozitivní energii a že vám (nebo ve vás) věří. A právě takovým člověkem je Božka.

Děkuji Milanovi Chytrému za tvůrčí atmosféru na ÚBZ a za šanci, kterou jsem dostal.

Značnou zásluhu na vzniku článků, které jsou součástí disertace, mají Verča Kalníková, Ondra Knápek, Salza Palpurina, Zuzka Plesková, Anni Pyykönen, Paťa Singh, Víťa Syrovátka a Anička Šímová. Děkuji jim za spolupráci v terénu (= že to tam se mnou vydrželi), za poskytnutí vlastních dat, pomoc při analýzách i za trefné komentáře k rukopisům.

Honzovi Košnarovi jsem zavázán za nezištnou pomoc v mých botanických začátcích.

Stanislavu Adamcovi a Jarmilu Feltlovi děkuji za uvedení do světa biologie, nalezení botaniky a možnost se jí věnovat už během gymnaziálních let.

Monice Hrubanové vděčím za možnost zúčastnit se v červnu 2015 snímkování mokřadních luk v údolí Rožnovské Bečvy a díky tomuto osvěžujícímu terénu si připomenout, že botanika není jen o nekonečním zírání do monitoru počítače, ale také o pohybu v přírodě.

I am very indebted to mire ecologists and vegetation scientists all over Europe for providing phytososciological data for vegetation synthesis, for useful comments to fen classification and particular vegetation types, for their willingness to cooperate and for their friendly attiude. I give my thanks namely to Liene Aunina, Dano Dítě, Ljuba Felbaba- Klushyna, Tatiana Ivchenko, Natalia Koroleva, Elena Lapshina, Predrag Lazarevid, Asbjørn Moen, Maxim Napreenko, Pawel Pawlikowski, Aaron Pérez-Haase, Lucia Sekulová, Viktor Smagin and Teemu Tahvanainen. Further, I would like to express my gratitude also to custodians of national or regional vegetation databases for providing vegetation plots and for comments to the manuscript of fen synthesis (paper 4). Although there was sometimes disagreement between me and a custodian (concerning e.g. selected methodical steps or validity of some fen alliances), I am glad for fruitful democratic discussion and reaching the final consensus. This group of vegetation scientists includes: Borja Jiménez-Alfaro, Ariel Bergamini, Claudia Biţă-Nicolae, Idoia Biurrun, Henri Brisse, Renata Dušterevska, Els De Bie, Jörg Ewald, Úna FitzPatrick, Xavier Font, Ulrich Graf, U. Jandt, Florian Jansen, Zygmunt Kącki, Anna Kuzemko, Flavia Landucci, Jesper Moeslund, Valerijus Rašomavičius, John Rodwell, Joop Schamiée, Urban Šilc, Zvjezdana Stančid and Annett Thiele. Vegetation plots were also kindly provided by Milan Valachovič and Volfgang Willner, although they refused to be co-authors of paper. Díky. Paldies. Vďaka. Дякуємо. Спасибо. Хвала. Takk. Dzięki. Gracias. Kiitos. Danke. Mulțumesc. Merci. Благодарам. Go raibh maith agat. Grazie. Takk. Ačiū. Thanks. Bedankt. Hvala. Дзякуй.

V neposlední řadě děkuji své rodině, zejména pak mamce. Za to, že jsem na světě. Za její vnitřní sílu. Za veškerou podporu, kterou jsem měl od dětských let až do současnoti. Za filosofii, že je potřeba v životě dělat to, co člověk považuje za správné, a nikoliv to, co od něj očekává „normální“ společnost.

Dále děkuji: Lubošovi Tichému za superprogram JUICE, bez kterého bych se neobešel. Petru Burešovi za tipy na lokality během bakalářského a magisterského studia a za podnětné rozhovory o tom, jak to vypadalo, vypadá a časem zřejmě bude vypadat s přírodou na Českomoravské vrchovině. Svatce Kubešové a Evě Mikuláškové za vydatnou a ochotnou pomoc při určování mechorostů. Jirkovi Danihelkovi a Vítu Grulichovi za mnohé přeurčení mých přibližných determinací cévnatých rostlin. Pavlu Dřevojanovi, Pavlu Novákovi, Lucce Hradilová a Janě Beneschové za dobrou atmosféru v našem ročníku během Bc. a Mgr. studia. Řadě dalších (nad)regionálních přírodovědců, s nimiž jsem měl tu možnost a čest spolupracovat, potkat se v terénu nebo s nimi alespop neformálně konzultovat některé otázky. Kromě výše uvedených do této skupiny patří mj. Babča, Tomáš Blažek, Jindřiška Bojková, Antonín Cedzo, Luděk Čech, Miloš Dudycha, Ester Ekrtová, Karej Fajmon, Kjell- Ivar Flatberg, Eva Holá, Michal Horsák, Verča Horsáková, Jana Glombová, Kamila & Jirka Juřičkovi, Monika Kolényová, Jirka Košnar, Berenika Lukášková, Katka Marečková, Dáša Papáčková, Martina Poláková, Vendula Polášková, Rolda, Terka Růžičková, Jana Schenková, Jakub Šmerda, Vanda Šorfová, Táňa & Milan Štechovi a Jitka Šterbová. Iloně Knollové a Daně Holubové za výpisy z České národní fytocenologické databáze i Evropského vegetační archivu. Knihovnicím Lucce Jarošové a Petře Šolcové za pomoc s hledáním knižních klenotů. Všem ostatním kamarádům a známým z ÚBZ. Přátelům z gymplu, pilířům mého života Jakubu Paulíčkovi a Davidu Schafferovi, jakož i Katce a Ivaně za to, že tu jsou a že jsou pořád stejní. Pavlíkovým z Poličky a Fajmonovým z Pusté Rybné. Oběma svým autům Zelené krasavici (†) a Fialovému broukovi. Železniční dopravě na trase (Svitavy–)Polička–Borová(–Žďárec u Skutče) bez jejíž existence bych těžko realizoval své první výzkumy mokřadů a vysokoškolské studium vůbec. Trollům, vládcům divoké skandinávské přírody, kteří se pravděpodobně na dálku zasloužili o vznik článku o celoevropské klasifikaci slatin (paper 4). Bez jisté nadpřirozené podpory by tento text byl jen těžko publikován. Skupinám Pink Floyd, Queen, Led Zeppelin, Scorpions, Iron Maiden, Nazareth, Uriah Heep, Whitesnake, Omega, Dire Straits, Framus Five, Etc..., Blue effect, Olympic a Progres 2 za jejich hudbu, která mě jako věrná společnice provázela po všechny ty roky. Některým rašeliništním lokalitám, jejichž prostá existence a krása udržovala plamínek zájmu v několika kritických obdobích, kdy jsem se už už chtěl na všechno vykašlat. Jedná se zejména o lokality Damašek, Dářko, Louky v Jeníkově, Meandry Svratky u Milov, Návesník, Radostínské rašeliniště, Ratajské rybníky, Řeka, Volákův kopec a Zlámanec, ale i další rašeliniště na Českomoravské vrchovině i jinde ve světě. Doufám, že tyto lokality přežijí 21. století a že zde pak bude ještě někdo, kdo se bude moci radovat z jejich krásy – v bezpečné a demokratické České republice, kde slušní lidé mohou beze strachu a obav z budoucnosti pracovat, vycházet na ulici a vychovávat své děti.

Preface

Classification of fens and fen vegetation has a long tradition in Europe, lasting for more than one century. Large number of different systems has been established: Cajanderian typology, phytosociological systems of Uppsala school as well as variety of classification schemes using the hierarchy of syntaxa and following the Braun-Blanquet's approach. Authors of all these systems obviously tried to create the best possible classification according to their expert experience. It is obvious that the development of such systems is a long lasting process which cannot be finished within a few years. This is all the more important for broad-scale classifications spanning national boundaries. Hence, this thesis does not have an arrogant ambition to create one perfect ultimate phytosociological system but only to make a contribution to the classification of fen communities and related issues for further evaluation and discussion.

Author contributions to the papers in the thesis

Peterka T., Plesková Z., Jiroušek M. & Hájek M. (2014): Testing floristic and environmental differentiation of rich fens on the Bohemian Massif. – Preslia 86: 337–366.

TP and MH conceived the ideas; ZP, TP and MJ sampled field data; TP and MH analysed the data and led manuscript writing; all authors discussed the ideas and commented on the manuscript.

Peterka T., Jiroušek M., Hájek M. & Jiménez-Alfaro B. (2015): European Mire Vegetation Database: a gap-oriented database for European fens and bogs. – Phytocoenologia 45: 291–297.

TP, MH and BJA conceived the idea; TP, MJ and MH organised the database building; TP led manuscript writing; all the authors commented on the manuscript.

Peterka T., Syrovátka V., Dítě D., Hájková P., Hrubanová M., Jiroušek M., Plesková Z., Singh P., Šímová A., Šmerdová E. & Hájek M.: Is variable plot size a serious constraint in broad- scale vegetation studies? A case study on fens. (manuscript)

TP, PH, MJ and MHá conceived the ideas; TP, DD, PH, MHr, MJ, ZP, PS, AŠ, EŠ and MHá sampled the field data; TP and VS analysed the data; TP and MHá led the manuscript writing; VS, PH, MJ, PS, AŠ and EŠ commented on the manuscript.

Peterka T., Hájek M., Jiroušek M., Jiménez-Alfaro B., Aunina L., Bergamini A., Dítě D., Felbaba-Klushyna L., Graf U., Hájková P., Hettenbergerová E., Ivchenko T.G., Jansen F., Koroleva N.E., Lapshina E.D., Lazarevid P.M., Moen A., Napreenko M.G., Pawlikowski P., Plesková Z., Sekulová L., Smagin V.A., Tahvanainen T., Thiele A., Biţæ-Nicolae C., Biurrun I., Brisse H., Dušterevska R., De Bie E., Ewald J., FitzPatrick Ú., Font X., Jandt U., Kącki Z., Kuzemko A., Landucci F., Moeslund J.E., Pérez-Haase A., Rašomavičius V., Rodwell J.S., Schaminée J.H.J., Šilc U., Stančid Z. & Chytrý M. (2017): Formalized classification of European fen vegetation at the alliance level. – Applied Vegetation Science 20: 124–142.

TP, MH, MJ and BJA conceived the ideas; MH, MJ, BJA, LA, AB, DD, LF-K, UG, PH, EH, TGI, NEK, EDL, PML, AM, MGN, PP, ZP, LS, VAS, TT, AT and APH provided own phytosociological data from private vegetation databases; FJ, CB-N, IB, HB, RD, EDB, JE, ÚFP, XF, UJ, ZK, AK, FL, JEM, VR, JSR, JHJS, UŠ, ZS and MC provided phytosociological data from national or regional vegetation databases; TP and MH analysed the data and led the manuscript writing; DD, BJA, PH, EDL, PML, AM, PP, VAS and TT provided valuable remarks to fen classification and validity of individual alliances; all the authors commented on the manuscript and discussed the ideas.

Peterka T., Hájek M., Dítě D., Hájková P., Palpurina S., Goia I., Grulich V., Kalníková V., Plesková Z., Šímová A. & Štechová T. (2018): Relict occurrences of boreal brown-moss quaking rich fens in the Carpathians and adjacent territories. – Folia Geobotanica 53: 265–276.

TP and MH conceived the ideas and led the manuscript writing; TP, MH, DD, PH, SP, IG, VK, ZP and AŠ sampled phytosociological data; DD and VG provided data on distribution of selected vascular plants; TŠ and PH provided data on distribution of selected bryophytes; PH provided data from macrofossil database; TP did the analyses; all the authors commented on the manuscript and discussed the ideas.

Peterka T., Plesková Z., Palpurina S., Kalníková V., Lazarevid P. & Hájek M. (2016): Meesia triquetra, new relict moss for the Republic of Macedonia. – Herzogia 29: 66–71.

TP analysed the data and led writing; all the authors sampled field data and commented on the manuscript.

Peterka T., Kalníková V. & Plesková Z. (2017): Pseudocalliergon lycopodioides, a new bryophyte species for Montenegro. – Herzogia 30: 496–500.

TP analysed the data and led writing; all the authors sampled field data and commented on the manuscript.

Peterka T., Dítě D., Hájková P. & Hájek M. (2016): Ověření výskytu suchopýru štíhlého (Eriophorum gracile) ve Žďárských vrších. – Východočeský sborník přírodovědný, Práce a studie 23: 47–56.

DD rediscovered target species at the studied locality; TP led manuscript writing and analysed the data; all the authors participated in the fieldwork and commented on the manuscript.

Vytržen a smýkán sebe sám se zříkám mám tě ve svých vráskách sladký ráji poslední Zůstaň... zůstaň...

Progres 2: Ztracený ráj

Snad každý autor, který se setkal s potřebou klasifikovat rašelinnou vegetaci, vytvořil svůj systém nebo alespoň adaptoval třídění, které mu nejlépe vyhovovalo... Se současným stavem klasifikace rašelinných společenstev nemůžeme tedy být spokojeni.

Rybníček (1981)

Content

1. Introduction ...... 15 1.1 Object of study ...... 15 1.1.1 What are fens? ...... 15 1.1.2 Distribution in Europe ...... 15 1.1.3 Importance, threats and protection ...... 16 1.2 Main factors determining species composition and variability of fen vegetation ...... 16 1.2.1 pH and mineral richness ...... 16 1.2.2 Availability of macronutrients ...... 17 1.2.3 Hydrological gradients ...... 18 1.2.4 Other gradients ...... 18 1.2.5 Biogeographical factors ...... 18 1.3 Broad-scale vegetation surveys ...... 19 1.4 Different classification systems of fen vegetation in Europe ...... 20 1.4.1 Classification of brown-moss quaking rich fens in temperate Europe ...... 20

2. Aims of the thesis ...... 22

3. Methodical aspects of fen vegetation syntheses – a brief overview ...... 24 3.1 Data collection ...... 24 3.2 Nomenclature ...... 24 3.3. Data filtering and stratification ...... 24 3.3.1 Filtering according to plot sizes ...... 24 3.4 Unsupervised versus formalised classification ...... 25

4. Main results of the thesis ...... 26 4.1 The importance of the poor-rich gradient for fen classifications ...... 26 4.2 European Mire Vegetation Database ...... 27 4.3 Effect of different plot sizes on vegetation classification ...... 28 4.4 Alliances of fen vegetation in Europe ...... 28 4.5 Relict occurrences of boreal brown-moss fens in Central and South-eastern Europe ...... 29 4.6 Remarks to ecology and occurrence of selected fen specialists ...... 29

5. References ...... 30

Paper 1 ...... 38 Paper 2 ...... 76 Paper 3 ...... 87 Paper 4 ...... 122 Paper 5 ...... 162 Paper 6 ...... 193 Paper 7 ...... 200 Paper 8 ...... 205

Curriculum vitae ...... 216

1. Introduction

1.1 Object of study

1.1.1 What are fens? Fens (minerotrophic mires) can be defined as low productive groundwater-fed1 wetlands characterised by accumulation of peat and low availability of macronutrients. The herb layer is mostly composed of the Cyperaceae family; the bryophyte layer is usually well-developed and consists of Sphagnum species or “brown-mosses”2 or both. From a syntaxonomical point of view, European fens are traditionally classified as the Scheuchzerio palustris-Caricetea fuscae Tüxen 1937 class (Mucina et al. 2016). This definition is relatively broad and involves on one hand large fens developed on deep peat layer and on the other hand young fen meadows, arctic fens or spring fens having only shallow peat layer.

1.1.2 Distribution in Europe The development and distribution of fens is influenced by climate, local hydrological regime and terrain configuration as well as by past and recent human activities (Sádlo 2000, Joosten & Clarke 2002, Grootjans et al. 2006, Rydin & Jeglum 2006). Due to favourable climate (i.e. cold and humid with low evapotranspiration), stable hydrological conditions and still relatively low human impact, minerotrophic mires occur frequently in the arctic and boreal zones of Europe, whereas they become much scattered towards the south. Hence, the current centres of fen distribution in Europe are Fennoscandia and European Russia (e.g. Sjörs et al. 1965, Botch & Masing 1983, Pakarinen 1995). In temperate Europe, fens are mostly confined to mountain and highland areas, such the Bohemian-Moravian Highlands, the Carpathians, the Alps or the Massif Central. This pattern is caused partly by more suitable climatic conditions at higher altitudes and partly by more intensive level of human impact in lower altitudes (Joosten et al. 2017; see also chapter 1.1.3). In southern Europe, fen vegetation is very rare and mostly restricted to the isolated localities in the highest mountains (e.g. Hájková et al. 2006, Jiménez-Alfaro et al. 2012). The isolation of fens in southern European mountain ranges probably resulted in higher level of endemism and smaller species pool of fen specialists, which partly complicates phytosociological classification (Joosten et al. 2017; see also chapter 1.2.5). Fens, and actually all the mires, are usually assigned to azonal vegetation formations (e.g. Pott 2005, Mucina et al. 2016). When consider common occurrence of fens in boreal and subarctic zone of Europe and gradually decreasing towards the south, they can be also considered as extrazonal formations. I mostly agree with the opinion of Rybníček (2005), who regarded mires as “semizonal”.

1 Concerning the origin of water, the term “fens” used in this thesis corresponds with the most frequently applied concept introduced by Scandinavian authors (Sjörs 1948, Malmer 1962, Økland et al. 2001) and followed by other ecologist (Bedford & Godwin 2003, Hájek et al. 2006). This concept distinguishes two main types of mires: fens (minerotrophic mires, saturated by both groundwater and precipitation) and bogs (ombrotrophic mires, supplied exclusively by precipitation). This division contradicts proposing of Wheeler & Proctor (2000) who divided mires into bogs of pH <5.0 and fens having pH >5.5. 2 The term “brown-mosses” refers to the non-sphagnaceous weft-forming bryophytes, mainly of the Amblystegiaceae family (the genera Calliergon s. lat., Campylium, Drepanocladus s. lat.). For further information see e.g. Rybníček (1964) or Hedenäs (2003).

1.1.3 Importance, threats and protection Together with other mires and peatlands, fens act as an effective carbon sinks (Armentano & Menges 1986, Nykänen et al. 1995) and natural water reservoirs stabilizing water regime in the landscape (Fojt 1994). The drainage of peatlands causes increase of water outflow as well as substantial emissions of carbon dioxide and other greenhouse gases, and thus contributes significantly to global warming (Flessa et al. 1998, van Diggelen et al. 2006). Fens further act as a model system in general ecology and island biogeography (Nekola 1999); they are important natural archives, storing biological material for millennia (Hájková et al. 2018). Last but not least, fens are very important in terms of community and biodiversity protection, especially in the agricultural landscape in Central and Western Europe. Minerotrophic mires harbour wide array of sensitive and endangered organisms, which cannot survive elsewhere (e.g. Holmen et al. 1967, Rybníček 1974, Grootjans et al. 2005, Bergamini et al. 2009, Juutinen 2011). European mires have been largely destroyed since 18th century (Joosten et al. 2017). The intensity of destruction nevertheless accelerated in the second half of 20th century together with the technological development. The retreat or deterioration of mires (including fens) have been recently reported across various European countries (Růžička 1989, Harding 1993, Lindholm & Heikkilä 2006, Topid & Stančid 2006, Mälson et al. 2008, Koch & Jurasinski 2014, Hájek et al. 2015, Rion et al. 2018). The primary cause of the destruction was intentional transformation of fens for agriculture and forestry, i.e. drainage of mires and conversion into agricultural fields, forests etc., as well as peat extraction for fuel, fertiliser and growing media (Joosten & Clarke 2002, Lamers et al. 2015). Recent deterioration of fens is further connected with general disruption of water regime in the landscape and subsequent water shortage. The lowering of water level in fens converts anoxic conditions in the soil into oxic environment, which causes peat mineralization (Grootjans et al. 1986) and subsequent increase of available macronutrients. The analogous effect has also external macronutrient input due to atmospheric deposition and nutrient influx from agricultural areas (Lamers et al. 2002, Lohila et al. 2010). Shifts in nutrient limitation together with absence of adequate management cause that low productive sedge- moss vegetation is overgrown by competitively stronger herbs and graminoids, previously limited by low nutrient availability, or by trees (Middleton et al. 2006, Billeter et al. 2007). These negative successional changes might be, to a considerable extent, blocked by regular mowing (Moen et al. 1999, Hájková et al. 2009). It holds true especially for fens and fen meadows that had been traditionally mown. In the current European Red List of Habitats, Janssen et al. (2016) identified 85% of mire habitats to be threatened in the European Union. Two habitat types of base-rich (calcareous) fens were even categorized as endangered, falling amongst 10% of the most threatened European habitats.

1.2 Main factors determining species composition and variability of fen vegetation

1.2.1 pH and mineral richness The vegetation composition of the Scheuchzerio palustris-Caricetea fuscae class varies predominantly along so-called “poor-rich gradient” (Fig. 1). The poor-rich gradient was originally formulated by Scandinavian authors (Du Rietz 1949, Sjörs 1950a) to mirror differences in species richness, though later studies emphasized the complexity of the gradient involving water chemistry connected to pH, calcium concentration and the total

16 mineral richness (Malmer 1986, Sjörs & Gunnarsson 2002, Hájek et al. 2006). The key role of poor-rich gradient for species turnover in fen vegetation was reported by studies from Fennoscandia (Persson 1962, Mörnsjö 1969, Fransson 1972, Malmer 1986, Heikkilä 1987, Singsaas 1989, Sjörs & Gunnarsson 2002, Tahvanainen 2004), Central Europe (Hájek et al. 2002, Navrátilová et al. 2006), the Alps (Gerdol 1995) or the Balkan peninsula (Hájková et al. 2006, Hájek et al. 2008). Although recent studies proved that pH and base richness are largely independent from gradient of macronutrient availability (i.e. fertility; Waughmann 1980, Wheeler & Proctor 2000, Bragazza & Gerdol 2002, Rozbrojová & Hájek 2008, Kooijman & Hedenäs 2009), some parts of the poor-rich gradient coincide also with increasing availability or deficiency of particular macronutrients or their forms. For example, the most calcium-rich fens are limited by phosphorus that is immobilized by calcium into forms unavailable to plants (Boyer & Wheeler 1989, Bedford et al. 1999). The ratio of nitrogen forms (ammonium versus nitrate) also differ along this gradient (Paulissen et al. 2004, Kooijman & Hedenäs 2009). Therefore, the mutual relationships between macronutrient availability, mineral richness and species composition of fen vegetation are not definitively resolved and studies from other than traditionally explored regions are needed. Additionally, some other chemical factors may contribute to forming the poor-rich gradient, e.g. iron or aluminium toxicity (Zohlen & Tyler 2000, Rozbrojová & Hájek 2008, Aggenbach et al. 2013).

Fig. 1. Scheme of poor-rich gradient in mires. Adopted from Rydin et al. (1999).

1.2.2. Availability of macronutrients Fen vegetation is generally characterized by low macronutrient availability. Nitrogen and phosphorus seems to be the main limiting elements, whereas potassium is probably of minor importance (Øien et al. 2018, but see de Mars et al. 1996). Besides impact on co- forming the poor-rich gradient (see previous section), macronutrient availability is mainly connected with the gradient of fertility or productivity, i.e. with increasing cover of nutrient- demanding species at the expense of sensitive fen specialists (Boyer & Wheeler 1989, Hájek et al. 2006, Kotowski et al. 2006). In other words, macronutrient availability is responsible for gradient from fens to meadows (Molinio-Arrhenatheretea) or tall reeds (Phragmito-

Magnocaricetea). The successional changes from fen communities towards more productive ones might be caused by external macronutrient supply, by water level dynamics or its decrease (see chapter 1.1.3). Recent studies suggest that species composition is influenced by nutrient concentration ratios rather than by absolute concentration of particular macroelements (Wassen et al. 1995, Koerselman & Meuleman 1996, Øien & Moen 2001, Güsewell 2005, Rozbrojová & Hájek 2008). Although overall macronutrient availability is very low in all minerotrophic mires, the nutrient ratios can account also for the compositional differences among fen vegetation types. For example, interesting phenomenon was found by Pawlikowski et al. (2013), who revealed that the N:P ratio is the best ecological predictor for the differences in species composition of the two brown-moss rich fen types in North- eastern Poland. The Caricion davallianae fens were usually P-limited, while the fens lacking typical species of Caricion davallianae were N-limited.

1.2.3 Hydrological gradients All the fens are characterised by the high and mostly stable water level and soil moisture. These factors determine axonic (reducing) conditions (Sjörs 1950a) that changes chemical processes, prevent or mitigate the decomposition and thus enable peat formation. As mentioned in the previous chapters, decreasing or fluctuating water level is connected with improved macronutrient availability. The gradient of water table depth is hence frequently related to the fertility gradient. Even species composition of pure fens is, however, shaped by differences in moisture, water table depth and its fluctuations. Especially in Scandinavian literature, the importance of the Hummock-mud-bottom gradient is stressed (Sjörs 1950b, 1990, Malmer et al. 1994). This gradient stretches from strongly waterlogged or even regularly inundated sites (“mud-bottom”, “carpets”) to wet or moderately wet sites (“lawns”, “hummocks”). The hydrological gradients are, however, apparent mostly in local or regional studies (Bragazza & Gerdol 1996, Jabłooska et al. 2011, Moeslund et al. 2013, Schenková et al. 2014, Horsáková et al. 2015, Pérez-Haase & Ninot 2015).

1.2.4 Other gradients Fennoscandian authors (e.g. Økland 1990, Laitinen et al. 2017) further distinguish the Mire expanse–mire margin gradient that combines more ecological factors, such as mineral richness, nutrient accessibility, soil organic content and groundwater fluctuation versus stability (Bragazza et al. 2005). Mire expanse is characterized by deep peat, stable water level and nutrient inaccessibility, whereas mire margin is defined by shallow peat, fluctuating water table and relatively good access to nutrients. Species of mire margin frequently occur also in forest and grassland vegetation, e.g. Filipendula ulmaria, Picea abies, Salix spp. (Joosten et al. 2017).

1.2.5 Biogeographical factors At broad geographic scales, the differences in species composition are driven also by macroclimate, even within relatively homogeneous vegetation (Jiménez-Alfaro et al. 2014). Macroclimate again represents complex gradient including ecophysiological requirements of individual species (Dahl 1988, Prentice et al. 1992) as well as competition intensity. For instance, Vicherová et al. (2017) detected that tolerance of Sphagnum species to higher calcium concentration is facilitated by higher precipitation. Therefore, Sphagnum expansion to alkaline fens is more common in humid regions, such as in the atlantic Europe.

Climatic patterns are further deepened by macroecological processes determining regional and local species pools, i.e. speciation, extinction, migration and dispersal events (Götzenberger et al. 2012). In Europe, the representation of fen specialists generally decreases towards the south (Horsáková et al. 2018). This pattern might be related either to spatial mass effect (Shmida & Ellner 1984), if scattered occurrence of fens in southern areas is considered, or to climate fluctuations in the past. Jiménez-Alfaro et al. (2012) concluded that fen vegetation in Iberian peninsula might be established under much more suitable climatic conditions and may experience long-time persistence in climatically sub-optimal mountain refugia, but related plant specialists may be sensitive to climatic changes and extinct in local populations. On the other hand, the species richness in isolated oromediterranean and oro-submediterranean areas is enhanced by higher level of endemism. Fen vegetation in high Balkan mountains can serve as a good example. These communities harbour wide array of endemic taxa, presumed geographic vicariants of fen species common in other parts of Europe (e.g. Narthecium scardicum, Pinguicula balcanica, Gymnadenia frivaldii, Primula farinosa subsp. exigua; Lakušid & Grgid 1971, Lakušid 1973). Dispersal abilities of individual species may play a role in forming of spatial differences in vegetation as well. Hájek et al. (2011) proved that several vascular plants are significantly linked to ancient fens at the millennial scale in the Western Carpathians. It suggests that dispersal limits of fen specialists from old to young localities within the same habitat (~ vegetation type) may contribute to spatial differences in plant communities, even at a relatively fine scale.

1.3 Broad-scale vegetation surveys Vegetation is one of the fundamental components of the ecosystems, and therefore vegetation survey is important for answering general ecological or biogeographical questions, as well as for conservation monitoring, management and delimitation of sites of conservational interest. However, the effective protection of selected habitats (i.e. vegetation types in this context) on the scale of the entire Europe is possible only on basis of harmonised classification systems with clearly defined units. Only such consistent vegetation systems may facilitate interpretation of the habitats of European conservation concern within the Natura 2000 network or the habitat classification system of EUNIS – European Nature Information System (Schaminée et al. 2016). Several typological studies of various vegetation types in Europe were published in last decades, but they were always restricted to one country or few neighbouring countries, i.e. to a part of the range of the target vegetation type (e.g. Botta-Dukát et al. 2005, Dúbravková et al. 2010, Michl et al. 2010, Rozbrojová et al. 2010, Sekulová et al. 2011). However, the recent need for consistent classification systems in Europe has recently driven vegetation scientists to elaborate broad-scale vegetation syntheses integrating national classification systems (De Cáceres et al. 2015). One of the first steps is an expert-based synopsis of nomenclaturally valid high-rank syntaxa in Europe (EuroVegChecklist; Mucina et al. 2016), which has been, however, elaborated without comparative analysis using primary data. Alongside, several vegetation syntheses and analyses based on large sets of vegetation plots were created (e.g. Eliáš et al. 2013, Douda et al. 2016, Willner et al. 2017a, b, Marcenò et al. 2018).

1.4 Different classification systems of fen vegetation in Europe Even phytosociological systems of fen communities in various European countries are different. The discrepancy among systems has led to different delimitation and acceptance of vegetation units in European countries, to the overall confusing situation in fen syntaxonomy as well as to frequent misunderstanding among mire ecologists. The inconsistency of fen classification results from diverse classification concepts applied (Rybníček 1981, Malmer 1985). Although there are some regional differences, the two main classification approaches are used. The vegetation systems defining particular vegetation types on the basis of hydrological conditions and stand physiognomy were introduced by Vanden Bergen (in Lebrun et al. 1949), Oberdorfer (1957), Oberdorfer et al. (1998) and Dierssen (1982, 1996) and more or less accepted in some other national vegetation surveys (e.g. Steiner 1992, Martinčič 1995, Coldea et al. 1997, Gerdol & Tomaselli 1997, Lájer 1998, Jermacãne & Laivirš 2001, Lawesson 2004, Matuszkiewicz 2007). These classification systems distinguish (i) topogenic waterlogged fens, usually called Caricion lasiocarpae and Rhynchosporion albae and (ii) spring fens and fen meadows, usually called Caricion davallianae and Caricion fuscae. The dominance of several vascular plants (Carex davalliana, C. lasiocarpa, C. limosa, C. nigra, C. rostrata, Rhynchospora alba) is frequently used as one of the key criteria for delimiting an alliance. The alliances, and even associations, delimited following this concept hence span habitats of different ecological features. For example, the Rhynchosporion albae alliance in these systems comprises both calcium-rich fens and dystrophic (bog) hollows. The parallel classification approach is based on the “poor-rich” gradient (see chapter 1.2.1), along which individual alliances are delimited, and emphasises equally bryophytes and vascular plants. Such classification schemes were introduced by botanists in Fennoscandia (Nordhagen 1943, Dahl 1956, Ruuhijärvi 1960, Persson 1961, Heikkilä 1987, Moen 1990) and, in many modifications, followed by other vegetation surveys (e.g. Passarge 1964, Succow 1974, Rybníček et al. 1984, Valachovič 2001, Kuznetsov 2003, Dítě et al. 2007, Tzonev et al. 2009, Lapshina 2010, Chytrý 2011). The identical (“ecological”) view on fen classification (with biogeographical factors as the additional criterion) is adopted also in the current overview of fen alliances within the European high-rank syntaxa checklist (Mucina et al. 2016). Besides the two contradict concepts, there are also transitional vegetation systems delimiting units along the poor-rich gradient, though still keeping broadly-defined alliances (Rhynchosporion albae or Caricion lasiocarpae), characterized by dominance of selected vascular plants. Such system were presented e.g. by Kojid et al. (1998), Koska & Timmermann (2004), Felbaba-Klushyna (2010) and Rivas-Martínez (2011).

1.4.1 Classification of brown-moss quaking rich fens in temperate Europe As an example of ambiguous classification issues, we can mention brown-moss quaking rich fens with boreal elements (Calliergon trifarium, Carex chordorrhiza, C. lasiocarpa, C. limosa, Scorpidium scorpioides) occurring rarely southward of boreal zone. In German-Austrian or similar vegetation systems based on physiognomy and hydrological gradients (Dierssen 1982, Steiner 1992, Coldea et al. 1997), these communities were assigned to broadly defined Caricion lasiocarpae. Other systems classified this vegetation either to the Caricion davallianae or Sphagno warnstorfii-Tomentypnion nitentis alliances (e.g. Dítě et al. 2007). This solution, however, does not match traditional concept of both alliances, since target vegetation mostly lacks Sphagnum species typical for Sphagno warnstorfii-Tomentypnion

20 nitentis as well as calcicole species characteristic for the Caricion davallianae alliance. In Scandinavia similar vegetation has been classified either as the Caricion lasiocarpae (Dierssen 1996) or the Stygio-Caricion limosae alliance (Nordhagen 1943, Dahl 1956, Moen et al. 2012).

2. Aims of the thesis

As mentioned above, the classification of fen vegetation in Europe is complicated and the national vegetation systems are different (chapter 1.4), which hampers effective communication among scientists as well as habitat protection on the supra-national level. Therefore, the main aim of the thesis was to made the first attempt on harmonised classification of European fens on the basis of a large set of vegetation plots and support the integration of national vegetation systems. Since the proposed classification scheme followed largely the poor-rich gradient as the main compositional driver for fen communities, the first step was to test the validity of delimitation of alliances by partitioning this gradient using original data on vegetation and site chemistry. For this purpose, the case study was conducted in fens on the Bohemian Massif (Czech Republic), i.e. the region where both soligenous and topogenous fens are rather common and where groundwater contains generally more potassium, iron and phosphorus as compared to most other investigated European regions. This study also deals with relationships between the pH/mineral richness gradient and macronutrient availability. Preliminary analysis of accessibility of fen vegetation data revealed significant gaps in Northern and South-eastern Europe. The baseline for classification of European fens hence comprised computerization of plots that had not been previously stored in electronic databases (see chapter 3.1) and subsequent building a new database. Broad-scale vegetation syntheses are inherently connected with a series of methodical steps and decisions concerning data stratification and filtering. As one of the expected limitations of continental-scale vegetation surveys is different size of included plots (see chapter 3.3.1), the thesis partially concerns with this issue. After dealing with these methodological aspects I focused on the principal aim of the thesis, i.e. on the delimitation of fen alliances in Europe using formalized classification approach. Additional studies focused on specific vegetation types were designed if the continental vegetation synthesis would recognize a new vegetation types for some territory, or would change the classification concept used in a region. In was the case of brown-moss quaking rich fens occurring rarely in Central and South-eastern Europe, having ambiguous syntaxonomical status here, which were classified in the Stygion-Caricion limosae alliance (boreal rich fen community). Therefore, another aim was to evaluate the correctness of this classification by comparison of relevés from temperate Europe with phytosociological material from Northern Europe (locus classicus of Stygio-Caricion limosae). At the same time, the thesis was focused on current and historical distribution of these exceptional fens in Central and South-eastern Europe (the Carpathians and adjacent areas). Although there is a long history of botanical research on fens in the majority of European countries, the knowledge about distribution and vegetation affinity of several fen specialists, important in terms of nature conservation or indicator value, is still incomplete. Therefore, a part of the thesis deals also with an update of knowledge of the occurrence and ecological demands of species, whose biogeographically important localities were found in the course of the field research.

In summary, the principal aims of the thesis are:

1) to test the floristic and environmental differentiation of fen alliances delimited along the poor-rich gradient (paper 1),

2) to introduce the European Mire Vegetation Database, new database that arose to fill the gap in data availability for continental syntheses or mire vegetation (paper 2),

3) to assess which plot sizes provide mutually consistent results in vegetation classifications in fens, regarding both the number of habitat specialists (i.e. diagnostic species) and the capturing main compositional gradients (paper 3),

4) to delimit formally fen alliances in Europe, to identify diagnostic species and distribution patterns of individual alliances and to test the robustness of the presented supervised classification by comparing it with unsupervised classifications of regional datasets and unconstrained gradient analysis (paper 4),

5) to assess whether the vegetation of brown-moss quaking rich fens in the Carpathians and adjacent areas differs from vegetation of the Stygio-Caricion limosae alliance in Northern Europe and map its current and potential historical distribution in Central Europe (paper 5),

6) update knowledge on distributional patterns and vegetation affinity of selected fen endangered species (papers 6–8).

3. Methodical aspects of fen vegetation syntheses – a brief overview

3.1 Data collection In the last decades, development of modern technologies, especially the Turboveg database management software (Hennekens & Schaminée 2001), led to the creation of national and regional vegetation databases (Schaminée et al. 2009, Dengler et al. 2011). Recently, majority of these databases have been integrated within the European Vegetation Archive (EVA; Chytrý et al. 2016). In spite of the progress in data digitizing on the national level and their sharing on the international level, some European regions nevertheless still remained insufficiently covered by relevés or lack digitized vegetation plots at all. These geographical gaps concerns mostly regions of Northern and South-eastern Europe (Schaminée et al. 2009, Chytrý et al. 2016). Hence, the phase of data collection for synthesis of European fen vegetation was divided into following steps: (i) to request data from national or regional vegetation databases stored in EVA, (ii) to ask regional experts on mire ecology and vegetation for private unpublished data and (iii) to computerize vegetation plots from individual “empty” regions which are scattered in monographs, manuscripts or local journals (paper 2).

3.2 Nomenclature The potential source of bias in vegetation analyses can, among others, result from different taxonomic concepts applied in particular countries or time periods (Jansen & Dengler 2010) as well as from misidentification within taxonomically-problematic species complexes. Although even closely related species frequently differ in ecological demands and indicator value (e.g. Kooijman & Hedenäs 1991, Hassel et al. 2018), it is necessary to merge selected taxa into species groups or aggregates to avoid potential taxonomical bias even at the cost of loss of information in broad-scale vegetation surveys.

3.3 Data filtering and stratification In mires, bryophytes are an extremely important group of organisms (Bergamini et al. 2001, Udd et al. 2015) acting as ecosystem engineers (Jones et al. 1994). Hence, all plots with no or insufficiently identified bryophytes (e.g. with non-identified species of sphagna) must be excluded from analyses of fen (mire) vegetation. Vegetation plots are not equally distributed across the European continent. While some regions are relatively rich in phytosociological data, there is a shortage of suitable vegetation plots from other areas (see Schaminée et al. 2009, Chytrý et al. 2016). To reduce the effect of oversampling, datasets of vegetation plots should be subjected to geographic stratification (Knollová et al. 2005). Before the geographical stratification, the duplicate plots should be removed from analysed dataset. Although management of vegetation database has been recently centralised by the development of EVA, the content of individual vegetation databases is fully dependent on custodians' choice. Since geographical scope of two or more concurrently existing databases may overlap, there is also a high chance that same vegetation plot might have been independently digitized more times.

3.3.1 Filtering according to plot sizes Another problem related to broad-scale vegetation syntheses, which is also reflected and partly discussed in this thesis (paper 3), is different size of analysed plots. The question whether to include plots of different sizes into analyses is connected with the fact that the

24 larger areas (i.e. larger vegetation plots in this context) inevitably harbour more species than the smaller ones (e.g. Arrhenius 1921, Storch 2016). Hence, vegetation scientists argue that the joint use of plots of different sizes may affect the results of classification (Podani 1984, Dengler et al. 2009). The dependence of community delimitation on the plot size is clear when different spatial and structural organization levels are captured at different measuring scales (Chytrý & Otýpková 2003). Less attention has been, however, paid to the question whether plot sizes affects classification within one magnitude of plot sizes. Chytrý & Otýpková (2003) suggested that plots falling into a certain range of the most frequently used sizes might be analysed together in a single dataset after excluding outliers. On the other hand, Dengler et al. (2009) recommended to analyse only plots within relatively narrow size ranges (even-sized plots) when using old data from databases and to apply uniform (standardised) plot sizes for all vegetation types that will by classified jointly in future surveys. In mire vegetation, variation of plot sizes, however, seems to limit continental-scale syntheses, because different plot sizes have been traditionally used in different regions (see overview in paper 3).

3.4 Unsupervised versus formalised classification Two fundamental plot-based vegetation classification approaches, both having obvious advantages and disadvantages, supporters and opponents, are generally applied. Unsupervised methods (clustering, numerical methods) group plots on the basis of their mutual resemblance and without any a priori information on plot membership (De Cáceres et al. 2015). Unsupervised methods demonstrate main patterns and variability within the dataset and are considered to be less subjective than supervised ones (see below). However, the resulting groups of plots are influenced by classification algorithm and its properties, data transformation, number of clusters and the overall content of the entire dataset, including diversity and frequency of particular “vegetation types” (Eliáš et al. 2013). Thus, the subjectivity is not fully eliminated but shifted towards the preparatory phase of the analysis (Moravec 1989). The supervised (expert-based) methods apply existing classification (a priori) criteria on existing data set and are therefore data-set-independent (Bruelheide & Chytrý 2000). Among supervised approaches, the COCKTAIL method (Bruelheide 2000) has become quite popular in the last decades. This method uses so-called sociological groups consisting of species with a strong statistical tendency to occur together in vegetation plots (Kočí et al. 2003). In some recent vegetation surveys (e.g. Dítě et al. 2007, Chytrý 2011, Douda et al. 2016, paper 4), the sociological groups, supplemented by covers of selected species, are combined by logical operators (AND, OR, NOT) into formal definitions (Bruelheide 1997, Kočí et al. 2003). The formal definitions allow unequivocal assignment of vegetation plots to defined vegetation unit. Instead of sociological groups, the functional species groups (Landucci et al. 2015) can be used. Expert systems involving formal definitions can be, however, criticised for authors´ subjectivism. The possible solution how to evaluate the quality of supervised classification is comparison of supervised classification with the results of unsupervised classification and ordination techniques (cf. Grabherr et al. 2003, Roleček 2007).

4. Main results of the thesis

4.1 The importance of the poor-rich gradient for fen classifications If the assumption of the key role of main environmental gradients for differentiation of alliances within a class is accepted, then the poor-rich gradient (Persson 1962, Malmer 1986, Hájek et al. 2006) should be considered as the main alliance-delimiting criterion in fens. The differentiation of extremely rich fens (Caricion davallianae), rich fens (Sphagno warnstorfii- Tomentypnion nitentis), moderately rich fens (Caricion fuscae = Caricion canescenti-nigrae) and poor fens (Sphagno-Caricion canescentis) along this gradient was tested using vegetation plots and environmental variables sampled on the Bohemian Massif (paper 1). Particular vegetation types were nearly completely differentiated in the PCA of environmental data, and they all differed significantly with respect to pH which (together with calcium) correlated with the main vegetation gradient (Fig. 2, 3). The second gradient coincided with nitrate and potassium concentration and thus corresponded to fertility (fen- to-meadow) gradient. Additionally, the N:P ratio in bryophyte biomass suggested a similar level of nutrient limitation across the vegetation types (Fig. 3). In other words, pH and calcium rather than nutrient availability differentiate causally major fen vegetation types (alliances). The presented study thus corroborates results of analogous studies conducted in Fennoscandia (Malmer 1986, Tahvanainen 2004), the Western Carpathians (Hájek et al. 2002), other parts of the Bohemian Massif (Navrátilová et al. 2006) and Bulgaria (Hájková et al. 2006, Hájek et al. 2008).

1.0 NH4

NO3 K WTD

PO4 Fe

Mg Ca cond

pH -1.0 -1.0 1.0

Fig. 2. PCA ordination of samples based only on environmental variables. Eigenvalues of the first two axes are 0.376 and 0.180. Plots of different vegetation types are indicated by different symbols: ○ Caricion davallianae (extremely rich fens), ■ Sphagno warnstorfii-Tomentypnion (rich fens), × Caricion fuscae (Caricion canescenti- nigrae); moderately rich fens), □ Sphagno-Caricion canescentis (poor fens).

8 a 120 60 b c d a b bc c n.s.

100 50 7

80 40

]

6 -1

60 30

[mg.l

5 2+

Ca 40 20

4 20 10

N:P (ratio biomass) in bryophyte

3 0 0 1 2 3 4 1 2 3 4 1 (n = 18) 2 (n = 79) 3 (n = 38) 4 (n = 35) most frequent species sampled w ithin vegetation type: Campylium stellatum Sphagnum warnstorfii Sphagnum teres Sphagnum fallax Fig. 3. Comparison of selected measured environmental variables and the N:P ratio in bryophyte biomass among the four vegetation types. 1: Caricion davallianae, 2: Sphagno warnstorfii-Tomentypnion, 3: Caricion fuscae (Caricion canescenti-nigrae), 4: Sphagno-Caricion canescentis. Medians are indicated by horizontal lines. Significant differences between groups (p > 0.05, one-way ANOVA, the post-hoc test) are indicated by different letters, n.s. = no significant differences.

In the analysis of pan-European dataset of fen plots (paper 4), the main variation in the species composition likewise mirrored site chemistry (pH, mineral richness) and sorted plots from calcareous and extremely rich fens, through rich and moderately rich fens, to poor fens and dystrophic hollows (Fig. 4). Hence, the concept of broadly-delimited fen alliances spanning different ecological fen types (e.g. Caricion lasiocarpae) and refuting some other alliances (e.g. Sphagno warnstorfii-Tomentypnion nitentis) should be abandoned.

Fig. 4. Detrended correspondence analysis of plots formally assigned to alliances with centroids of particular alliances along first three axes. Eigenvalues: 1st axis (DCA1) 0.595, 2nd axis (DCA2) 0.430, 3rd axis (DCA 3) 0.378. CvT = Caricion viridulo-trinervis, Cd = Caricion davallianae, CaS = Caricion atrofusco-saxatilis, SCl = Stygio- Caricion limosae, SwT = Sphagno warnstorfii-Tomentypnion nitentis, SaT = Saxifrago-Tomentypnion, Ns = Narthecion scardici, Cs = Caricion stantis, AJ = Anagallido tenellae-Juncion bulbosi, De = Drepanocladion exannulati, Cf = Caricion fuscae (Caricion canescenti-nigrae), SCc = Sphagno-Caricion canescentis, Sp = Scheuchzerion palustris.

4.2 European Mire Vegetation Database The effort to produce broad-scale vegetation synthesis on the basis of vegetation plots (i.e. primary data) has led to the creation of the European Mire Vegetation Database (paper 2). The database currently contains 10 000< relevés of the classes Scheuchzerio palustris- Caricetea fuscae and Oxycocco-Sphagnetea published in various monographs, manuscripts

27 or local journals, but not stored in any other national or regional electronic vegetation databases. Most of the digitized data come from the territories of Northern and South- eastern Europe (Fig. 5). Since 2015 the database has been included in the European Vegetation Archive (Chytrý et al. 2016).

Fig. 5. Distribution of relevés stored in the European Mire Vegetation Database (data accessed on 16 May 2018).

4.3 Effect of different plot sizes on vegetation classification Vegetation plots stored in databases are quite heterogeneous concerning plot sizes (Chytrý & Otýpková 2003, Dengler et al. 2009, papers 2, 3 and 4). This fact must be taken into account before any analysis and the range of included plot sizes should then depend on studied questions (Kenkel et al. 1989, Jalonen et al. 1998). Using species-area curves, we detected that plots of 1 m2 and 16–25 m2 capture very comparable counts of fen specialist species (i.e. presumed diagnostic species essential for vegetation classification; paper 3). This pattern was consistent across four fen vegetation types and two independent geographical regions. Further, separate ordinations using plots of 1 m2, 4 m2 and 16 m2 detected identical environmental gradients. Even clusters of plots produced by K-means classification reflected different vegetation types or geographic regions rather than different plot sizes. Hence, we can conclude that plots of different sizes (at least within the range of 1–25 m2) might be used jointly in broad-scale studies of fen vegetation without introducing significant error. The potential bias introduced by different plot sizes is of relatively low importance when compared with all other possible sources of uncertainty. Plots smaller than 1 m2 seem to be inconvenient for classifications of plant communities due to the considerably lower representation of specialists and less tight relation to the main environmental gradients.

4.4 Alliances of fen vegetation in Europe Formalised vegetation classification of European fens at the alliance level was proposed (paper 4). The distribution of individual alliances in Europe (plus western Siberia and south- eastern Greenland) was mapped and their diagnostic species were identified. As mentioned above, the main variation in species composition reflected pH and mineral richness.

Geographic (macroclimatic) variation was reflected in the second most important gradient (fens in atlantic Europe in one end of the gradient versus arctic, alpine and arcto-alpine fens in the second one; Fig. 4). The following vegetation units were formally defined and characterized: Caricion viridulo-trinervis (sub-halophytic Atlantic dune-slack fens), Caricion davallianae (temperate calcareous fens), Caricion atrofusco-saxatilis (arcto-alpine calcareous fens), Stygio-Caricion limosae (boreal topogenic brown-moss fens), Sphagno warnstorfii-Tomentypnion nitentis (Sphagnum-brown-moss rich fens), Saxifrago-Tomentypnion (continental to boreo- continental nitrogen-limited brown-moss rich fens), Narthecion scardici (alpine fens with Balkan endemics), Caricion stantis (arctic brown-moss rich fens), Anagallido tenellae-Juncion bulbosi (Ibero-Atlantic moderately rich fens), Drepanocladion exannulati (arcto-boreal-alpine non-calcareous fens), Caricion fuscae (Caricion canescenti-nigrae, temperate moderately rich fens), Sphagno-Caricion canescentis (poor fens) and Scheuchzerion palustris (dystrophic hollows3). The ecological meaningfulness of this classification scheme was supported by the results of a set of regional unsupervised classifications. The proposed classification can serve as a state-of-the-art baseline for further development of the pan-European fen typologies on various hierarchical levels.

4.5 Relict occurrences of boreal brown-moss fens in Central and South-eastern Europe Phytosociological classification of brown-moss quaking rich fens with boreal elements in the Carpathians and adjacent territories (the Bohemian Massif, the Dinaric Alps) has not been sufficiently solved yet. The European synthesis of fen vegetation (paper 4) suggests that these fens belong to the Stygio-Caricion limosae alliance (boreal rich fen community). NMDS and cluster analysis further revealed that species composition of these fens corresponds well with that in Northern Europe, confirming their assignment to Stygio-Caricion limosae (paper 5). Macrofossil data suggest that this vegetation type had been much more common in Central Europe in the late glacial and early post-glacial times. The main causes of present- day rarity of these fens are forest spreading in Middle Holocene and recent disruption of water regime in the landscape and subsequent successional shifts.

4.6 Remarks to ecology and occurrence of selected fen specialists During the field research, new localities of Meesia triquetra and Pseudocalliergon lycopodioides were found in the Balkan peninsula (papers 6, 7). Further, the occurrence of Eriophorum gracile was re-discovered in one fen in the Bohemian Massif (paper 8). All the above-mentioned taxa are regarded as glacial or post-glacial relicts in temperate Europe and southern areas (Frahm 2012, Dítě et al. 2018). They further belong among fen specialists (Mucina et al. 2016), rare and endangered taxa in particular regions. Deeper knowledge of their local distribution and vegetation affinity might support the design of appropriate management plans for individual species as well as for their habitats.

3 The Scheuchzerion palustris alliance is frequently characterized as the vegetation of bog hollows. However, the same vegetation may cover large areas without any contact with truly bog habitats in the boreal zone of Europe. Therefore, we suggest replacing definitely the term bog hollows by more appropriate term of dystrophic hollows.

5. References

Aggenbach C.J.S., Backx H., Emsens W.J., Grootjans A.P., Lamers L.P.M., Smolders A.J.P., Stuyfzand P.J., Wołejko L. & Van Diggelen R. (2013): Do high iron concentrations in rewetted rich fens hamper restoration? – Preslia 85: 405–420. Armentano T.V. & Menges E.S. (1986): Patterns of change in the carbon balance of organic soil-wetlands of the temperate zone. – J. Ecol. 74: 755–774. Arrhenius O. (1921): Species and area. – J. Ecol. 9: 95–99. Bedford B.L. & Godwin K.S. (2003): Fens of the United States: distribution, characteristics, and scientific connection versus legal isolation. – Wetlands 23: 608–629. Bedford B.L., Walbridge M.R. & Aldus A. (1999): Patterns in nutrient availability and plant diversity of temperate North American wetlands. – Ecology 8: 2151–2169. Bergamini A., Pauli D., Peintinger M. & Schmid B. (2001): Relationships between productivity, number of shoots and number of species in bryophytes and vascular plants. – J. Ecol. 89: 920–929. Bergamini A., Peintinger M., Fakheran S., Moradi H., Schmid B. & Joshi J. (2009): Loss of habitat specialist despite conservation management in fen remnants 1995–2006. – Persp. Pl. Ecol. Evol. Syst. 11: 65–79. Billeter R., Peintinger M. & Diemer M. (2007): Restoration of montane fen meadows by mowing remains possible after 4–35 years of abandonment. – Bot. Helv. 117: 1–13. Botch M.S. & Masing V.V. (1983): Mire ecosystems in the U.S.S.R. – In: Gore A.J.P. (ed.), Mires: Swamp, Bog, Fen and Moor, B. Regional Studies, pp. 95–152, Elsevier Scientific Publishing Company, Amsterdam. Botta-Dukát Z., Chytrý M., Hájková P. & Havlová M. (2005): Vegetation of lowland wet meadows along a climatic continentality gradient in Central Europe. – Preslia 77: 89–111. Boyer M.L.H. & Wheeler B.D. (1989): Vegetation patterns in spring-fed calcareous fens: calcite precipitation and constrains on fertility. – J. Ecol. 77: 597–609. Bragazza L. & Gerdol R. (1996): Response surfaces of plant species along water-table depth and water pH gradients in a poor mire on the Southern Alps. – Ann. Bot. Fenn. 33: 11–20. Bragazza L. & Gerdol R. (2002): Are nutrient availability and acidity-alkalinity gradients related in Sphagnum- dominated peatlands? – J. Veg. Sci. 13: 473–482. Bragazza L., Rydin H. & Gerdol R. (2005): Multiple gradients in mire vegetation: a comparison of a Swedish and an Italian bog. – Pl. Ecol. 177: 223–236. Bruelheide H. & Chytrý M. (2000): Towards unification of national vegetation classifications: A comparison of two methods for analysis of large data sets. – J. Veg. Sci. 11: 295–306. Bruelheide H. (1997): Using formal logic to classify vegetation. – Folia Geobot. 32: 41–46. Bruelheide H. (2000): A new measure of fidelity and its application to defining species groups. – J. Veg. Sci. 11: 167–178. Chytrý M. & Otýpková Z. (2003): Plot sizes used for phytosociological sampling of European vegetation. – J. Veg. Sci. 14: 563–570. Chytrý M. (ed.) (2011): Vegetace České republiky 3. Vodní a mokřadní vegetace.– Academia, Praha. Chytrý M., Hennekens S.M., Jiménez-Alfaro B., Knollová I., Dengler J., Jansen F., Landucci F., Schaminée J.H.J., Adid S., Agrillo E., Ambarlı D., Angelini P., Apostolova I., Attorre F., Berg C., Bergmeier E., Biurrun I., Botta- Dukát Z., Brisse H., Campos J.A., Carlón L., Čarni A., Casella L., Csiky J., Dušterevska R., Dajid Stevanovid Z., Danihelka J., De Bie E., de Ruffray P., De Sanctis M., Dickoré W.B., Dimopoulos P., Dubyna D., Dziuba T., Ejrnæs R., Ermakov N., Ewald J., Fanelli G., Fernández-González F., FitzPatrick Ú., Font X., García- Mijangos I., Gavilán R.G., Golub V., Guarino R., Haveman R., Indreica A., Işık Gürsoy D., Jandt U., Janssen J.A.M., Jiroušek M., Kącki Z., Kavgacı A., Kleikamp M., Kolomiychuk V., Krstivojevid Duk M., Krstonošid D., Kuzemko A., Lenoir J., Lysenko T., Marcenò C., Martynenko V., Michalcová D., Moeslund J.E., Onyshchenko V., Pedashenko H., Pérez-Haase A., Peterka T., Prokhorov V., Rašomavičius V., Rodríguez- Rojo M.P., Rodwell J.S., Rogova T., Ruprecht E., Rūsira S., Seidler G., Šibík J., Šilc U., Škvorc Ž., Sopotlieva D., Stančid Z., Svenning J.-C., Swacha G., Tsiripidis I., Turtureanu P.D., Uğurlu E., Uogintas D., Valachovič M., Vashenyak Y., Vassilev K., Venanzoni R., Virtanen R., Weekes L., Willner W., Wohlgemuth T. & Yamalov S. (2016): European Vegetation Archive (EVA): an integrated database of European vegetation plots. – Appl. Veg. Sci. 19: 173–180. Coldea G. (ed.) (1997): Les associations végétales de Roumanie. Tome 1. Les associations herbacées naturelles. – Presses Universitaires, Cluj. Dahl E. (1956): Rondane. Mountain vegetation in south Norway and its relation to the environment. – Skr. Norske Vidensk.-Akad. Oslo, Mat.-Naturvidensk. Kl. 3: 1–374.

Dahl E. (1988): The phytogeography of northern Europe: British Isles, Fennoscandia and adjacent areas. – Cambridge University Press, Cambridge. De Cáceres M., Chytrý M., Agrillo E., Attorre F., Botta-Dukát Z., Capelo J., Czúcz B., Dengler J., Ewald J., Faber- Langendoen D., Feoli E., Franklin S.B., Gavilán R., Gillet F., Jansen F., Jiménez-Alfaro B., Krestov P., Landucci F., Lengyel A., Loidi J., Mucina L., Peet R.K., Roberts D.W., Roleček J., Schaminée J.H.J., Schmidtlein S., Theurillat J.-P., Tichý L., Walker D.A., Wildi O., Willner W. & Wiser S.K. (2015): A comparative framework for broad-scale plot-based vegetation classification. – Appl. Veg. Sci. 18: 543– 560. de Mars H., Wassen M.J. & Peeters W.H.M. (1996): The effect of drainage and management on peat chemistry and nutrient deficiency in the former Jegrznia-floodplain (Ne-Poland). – Vegetatio 126: 59–72. Dengler J., Jansen F., Glöckler F., Peet R.K., De Cáceres M., Chytrý M., Ewald J., Oldeland J., Lopez-Gonzalez G., Finckh M., Mucina L., Rodwell J.S., Schaminée J.H.J. & Spencer N. (2011): The Global Index of Vegetation- Plot Databases (GIVD): a new resource for vegetation science. – J. Veg. Sci. 22: 582–597. Dengler J., Löbel S. & Dolnik C. (2009): Species constancy depends on plot size – a problem for vegetation classification and how it can be solved. – J. Veg. Sci. 20: 754–766. Dierssen K. (1982): Die wichtigsten Pflanzengesellschaften der Moore NW-Europas. – Conservatoire et Jardin botaniques Genéve, Genf. Dierssen K. (1996) : Vegetation Nordeuropas. – Verlag Eugen Ulmer, Stuttgart. Dítě D., Hájek M. & Hájková P. (2007): Formal definitions of Slovakian mire plant associations and their application in regional research. – Biologia 62: 400–408. Dítě D., Hájek M., Svitková I., Košuthová A., Šoltés R. & Kliment J. (2018): Glacial-relict symptoms in the Western Carpathian flora. – Folia Geobot. 53: 277–300. Douda J., Boublík K., Slezák M., Biurrun I., Nociar J., Havrdová A., Doudová J., Adid S., Brisse H., Brunet J., Chytrý M., Claessens H., Csiky J., Didukh Ya., Dimopoulos P., Dullinger S., FitzPatrick Ú., Guisan A., Horchler P.J., Hrivnák R., Jandt U., Kącki Z., Kevey B., Landucci F., Lecomte H., Lenoir J., Paal J., Paternoster D., Pauli H., Pielech R., Rodwell J.S., Roelandt B., Svenning J.-C., Šibík J., Šilc U., Škvorc Ž., Tsiripidis I., Tzonev R.T., Wohlgemuth T. & Zimmermann N.E. (2016): Vegetation classification and biogeography of European floodplain forests and alder carrs. – Appl. Veg. Sci. 19: 147–163. du Rietz G.E. (1949): Huvudenheter och huvudgränser i svensk myrvegetation. – Svensk Bot. Tidskr. 43: 279– 304. Dúbravková D., Chytrý M., Willner W., Illyés E., Janišová M. & Kállayné Szerényi J. (2010): Dry grasslands in the Western Carpathians and the northern Pannonian Basin: a numerical classification. – Preslia 82: 165– 221. Eliáš P. Jr., Sopotlieva D., Dítě D., Hájková P., Apostolova I., Senko D., Melečková Z. & Hájek M. (2013): Vegetation diversity of salt-rich grasslands in Southeast Europe. – Appl. Veg. Sci. 16: 521–537. Felbaba-Klushina L. (2010): Prodromus syntaksonib roslynnosti bolit i cholodnych džerel Ukrains'kych Karpat (Klasy Scheuchzerio-Caricetea fuscae Tx. 1937, Oxycocco-Sphagnetea Br.-Bl. et Tx. ex Westhoff et al. 1946, Montio-Cardaminetea Br .-Bl. et Tx. ex Klika et Hadač 1944). – Nauk. Visn. Užhor. Un., Ser. Biolohija 28: 73–82. Flessa H., Wild U., Klemisch M. & Pfadenhauer J. (1998): Nitrous oxide and methane fluxes from organic soils under agriculture. – Eur. J. Soil Sci. 49: 327–335. Fojt W. (1994): Dehydration and the threat to East Anglian fens, England. – Biol. Conserv. 69: 163–175. Frahm J.P. (2012): The phytogeography of European bryophytes. – Bot. Serbica 36: 23–36. Fransson S. (1972): Myrvegetation i sydvästra Värmland. – Acta Phytogeogr. Suec. 57: 1–133. Gerdol R. & Tomaselli M. (1997): Vegetation of wetlands in the Dolomites. – Diss. Bot. 281: 1–197. Gerdol R. (1995): Community and species-performance patterns along an alpine poor-rich mire gradient. – J. Veg. Sci. 6: 175–182. Götzenberger L., de Bello F., Brçthen K.A., Davison J., Dubuis A., Guisan A., Lepš J., Lindborg R., Moora M., Pärtel M., Pellissier L., Pottier J., Vittoz P., Zobel K. & Zobel M. (2012): Ecological assembly rules in plant communities-approaches, patterns and prospects. – Biol. Reviews 87: 111–127. Grabherr G., Reiter K. & Willner W. (2003): Towards objectivity in vegetation classification: the example of the Austrian forests. – Pl. Ecol. 169: 21–34. Grootjans A.P., Alserda A., Bekker C. W., Janáková M., Madaras M., Stanová V., Ripka J., van Delft B. & Wolejko L. (2005): Calcareous spring mires in Slovakia; jewels in the crown of the mire kingdom. – Stapfia 85: 97– 116. Grootjans A.P., Adema E.B., Bleuten W., Joosten H., Madaras M. & Janáková M. (2006): Hydrological landscape settings of base-rich fen mires and fen meadows: an overview. – Appl. Veg. Sci. 9: 175–184.

Grootjans A.P., Schipper P.C. & van der Windt H.J. (1986): Influence of drainage on N-mineralization and vegetation response in wet meadows. 2. Cirsio-Molinietum stands. – Acta Oecol. 7: 3–14. Güsewell S. (2005): Nutrient resorption of wetland graminoids is related to the type of nutrient limitation. – Functional Ecol. 19: 344–354. Hájek M., Hájková P. & Apostolova I. (2008): New plant associations from Bulgarian mires. – Phytol. Balcan. 14: 377–399. Hájek M., Hekera P. & Hájková P. (2002): Spring fen vegetation and water chemistry in the Western Carpathian flysch zone. – Folia Geobot. 37: 205–224. Hájek M., Horsák M., Hájková P. & Dítě D. (2006): Habitat diversity of central European fens in relation to environmental gradients and an effort to standardise fen terminology in ecological studies. – Persp. Pl. Ecol. Evol. Syst. 8: 97–114 Hájek M., Horsák M., Tichý L., Hájková P., Dítě D. & Jamrichová E. (2011): Testing a relict distributional pattern of fen plant and terrestrial snail species at the Holocene scale: a null model approach. – J. Biog. 38: 742– 755. Hájek M., Jiroušek M., Navrátilová J., Horodyská E., Peterka T., Plesková Z., Navrátil J., Hájková P. & Hájek T. (2015): Changes in the moss layer in Czech fens indicate early succession triggered by nutrient enrichment. – Preslia 87: 279–301. Hájková P., Hájek M. & Apostolova I. (2006): Diversity of wetland vegetation in the Bulgarian high mountains, main gradients and context-dependence of the pH role. – Pl. Ecol. 184: 111–130. Hájková P., Hájek M. & Kintrová K. (2009): How can we effectively restore species richness and natural composition of a Molinia-invaded fen? – J. Appl. Ecol. 46: 417–425. Hájková P., Štechová T., Šoltés R., Šmerdová E., Plesková Z., Dítě D., Bradáčová J., Mútpanová M., Singh P. & Hájek M. (2018): Using a new database of plant macrofossils of the Czech and Slovak Republics to compare past and present distribution of hypothetically relict fen mosses. – Preslia 90: 367–386. Harding M. (1993): Redgrave and lopham fens, East Anglia, England: a case study of change in flora and fauna due to groundwater abstraction. – Biol. Conserv. 66: 35–45. Hassel K., Kyrkjeeide M.O., Yousefi N., Prestø T., Stenøien H.K., Shaw J.A. & Flatberg K.I. (2018) Sphagnum divinum (sp. nov.) and S. medium Limpr. and their relationship to S. magellanicum Brid. – J. Bryol. 40: 197–222. Hedenäs L. (2003): The European species of the Calliergon-Scorpidium-Drepanocladus complex, including some related or similar species. – Meylania 28: 1–116. Heikkilä H. (1987): The vegetation and ecology of mesotrophic and eutrophic fens western Finland. – Ann. Bot. Fenn. 24: 155–175. Hennekens S.M. & Schaminée J.H.J. (2001): TURBOVEG, a comprehensive data base management system for vegetation data. – J. Veg. Sci. 12: 589–591. Holmen H., Johnels A., Malmer N., Perrson Å. & Sjörs H. (1967). Peatland and peatland conservation in Sweden. – Aquilo, Ser. Botanica 6: 120–136. Horsáková V., Hájek M., Hájková P., Dítě D. & Horsák M. (2018): Principal factors controlling the species richness of European fens differ between habitat specialists and matrix-derived species. – Divers. Distribut. 24: 742–754. Horsáková V., Horsák M., Hájek M., Hájková P. & Dítě D. (2015): Mollusc assemblages of Scandinavian fens: species composition in relation to environmental gradients and vegetation. – Ann. Zool. Fenn. 52: 1–16. Jabłooska E., Pawlikowski P., Jarzombkowski F., Chormaoski J., Okruszko T. & Kłosowski S. (2011): Importance of water level dynamics for vegetation patterns in a natural percolation mire (Rospuda fen, NE Poland). – Hydrobiologia 674: 105–117. Jalonen J., Vanha-Majamaa I. & Tonteri T. (1998): Optimal sample and plot size for inventory of field and ground layer vegetation in a mature Myrtillus-type boreal spruce forest. – Ann. Bot. Fenn. 35: 191–196. Jansen F. & Dengler J. (2010): Plant names in vegetation databases – a neglected source of bias. – J. Veg. Sci. 21: 1179–1186. Janssen J.A.M., Rodwell J.S., García Criado M., Gubbay S., Haynes T., Nieto A., Sanders N., Landucci F., Loidi J., Ssymank A., Tahvanainen T., Valderrabano M., Acosta A., Aronsson M., Arts G., Attorre F., Bergmeier E., Bijlsma R.-J., Bioret F., Biţæ-Nicolae C., Biurrun I., Calix M., Capelo J., Čarni A., Chytrý M., Dengler J., Dimopoulos P., Essl F., Gardfjell H., Gigante D., Giusso del Galdo G., Hájek M., Jansen F., Jansen J., Kapfer J., Mickolajczak A., Molina J.A., Molnár Z., Paternoster D., Piernik A., Poulin B., Renaux B., Schaminée J.H.J., Šumberová K., Toivonen H., Tonteri T., Tsiripidis I., Tzonev R. & Valachovič M. (2016): European Red List of Habitats - Part 2. Terrestrial and freshwater habitats. – Publications Office of the European Union, Luxembourg.

Jermacãne S. & Laivirš M. (2001): Latvijå aprakstīto augu sabiedrību sintaksonu saraksts. – Latvijas Veģetåcija 4: 115–132. Jiménez-Alfaro B., Fernández Pascual E., Díaz González T. E., Pérez Haase A. & Ninot J.M. (2012): Diversity of fen vegetation and related plant specialists in mountain refugia of the Iberian Peninsula. – Folia Geobot. 47: 403–419. Jiménez-Alfaro B., Hájek M., Ejrnaes R., Rodwell J., Pawlikowski P., Weeda E.J., Laitinen J., Moen A., Bergamini A., Aunina L., Sekulová L., Tahvanainen T., Gillet F., Jandt U., Dítě D., Hájková P., Corriol G., Kondelin H. & Díaz T.E. (2014): Biogeographic patterns of base-rich fen vegetation across Europe. – Appl. Veg. Sci. 17: 367–380. Jones C.G., Lawton J.H. & Shachak M. (1994): Organisms as ecosystem engineers. – Oikos 69: 373–386. Joosten H. & Clarke D. (2002): Wise Use of Peatlands. – International Mire Conservation Group and International Peat Society, Jyväskylä. Joosten H., Tanneberger F. & Moen A. (eds.) (2017): Mires and peatlands of Europe. Status, distribution and conservation. – Schweizerbart Science Publishers, Stuttgart. Juutinen S. (2011): The decrease of rich fen bryophytes in springs as a consequence of large-scale environmental loss. A 50-year re-sampling study. – Lindbergia 34: 2–8. Kenkel N.C., Juhász-Nagy P. & Podani J. (1989): On sampling procedures in population and community ecology. – Vegetatio 83: 195–207. Knollová I., Chytrý M., Tichý L. & Hájek O. (2005): Stratified resampling of phytosociological databases: some strategies for obtaining more representative data sets for classification studies. – J. Veg. Sci. 16: 479– 486. Koch M. & Jurasinski G. (2014): Four decades of vegetation development in a percolation mire complex following intensive drainage and abandonment. – Pl. Ecol. Divers 8: 49–60. Kočí M., Chytrý M. & Tichý L. (2003): Formalised reproduction of an expert-based phytosociological classification: A case study of subalpine tall-forb vegetation. – J. Veg. Sci. 14: 601–610. Koerselman W. & Meuleman A.F.M. (1996): The vegetation N:P ratio: a new tool to detect the nature of nutrient limitation. – J. Appl. Ecol. 33: 1441–1450. Kojid M., Popovid R. & Karažid B. (1998): Sintaksonomski pregled vegetacije Srbije. – Inst. Biol. Res. „Siniša Stankovič“, Beograd. Kooijman A.M. & Hedenäs L. (1991): Differentiation in habitat requirements within the genus Scorpidium, especially between S. revolvens and S. cossonii. – J. Bryol. 16: 619–627. Kooijman A.M. & Hedenäs L. (2009): Changes in nutrient availability from calcareous to acid wetland habitats with closely related brown moss species: increase instead of decrease in N and P. – Pl. Soil 324: 267–278. Koska I. & Timmermann T. (2004): Parvo-Caricetea den Held & Westhoff in Westhoff & den Held 1969 nom. cons. propos. – Riede und Röhrichte mäßig nährstoffarmer Niedermoore und Ufer. In: Berg C., Dengler J., Abdank A. & Isermann M. (eds.), Die Pflanzengesellschaften Mecklenburg-Vorpommerns und ihre Gefährdung, pp. 163–195, Wiessdorn Verlag, Jena. Kotowski W., Thörig W., van Diggelen R. & Wassen M.J. (2006): Competition as a factor structuring species zonation in riparian fens – a transplantation experiment. – Appl. Veg. Sci. 9: 231–240. Kuznetsov O. (2003): Topological-ecological classification of mire vegetation in the Republic of Karelia (Russia). – The Finn. Environ. 485: 117–123. Laitinen J., Oksanen J., Kaakinen E., Parviainen M., Küttim M. & Ruuhijärvi R. (2017): Regional and vegetation- ecological in northern boreal flark fens of Finnish Lapland: analysis from a classic material. – Ann. Bot. Fenn. 54: 179–195. Lájer K. (1998): Bevezetés a magyarországi Lápok vegetáció-ökológiájába. – Tilia 6: 84–238. Lakušid R. & Grgid P. (1971): Ekologija i rasprostiranje endemičnih vrsta Narthecium scardicum Koš., Pinguicula balcanica Cas., Gymnadenia friwaldii Hampe i Silene asterias Grsb. – Ekologija 6: 337–350. Lakušid R. 1973. Narthecietalia ordo novus der Scheuchzerio-Caricetea fuscae Nordh. 1936 in den südeuropäischen Gebirgen. – Veröff. Geobot. Institut. ETH 51: 158–162. Lamers L.P.M., Smolders A.J.P. & Roelofs J.G.M. (2002): The restoration of fens in the Netherlands. – Hydrobiol. 478: 107–130. Lamers L.P.M., Vile M.A., Grootjans A.P., Acreman M.C., van Diggelen R., Evans M.G., Richardson C.J., Rochefort L., Kooijman A.M., Roelofs J.G. M. & Smolders A.J.P. (2015): Ecological restoration of rich fens in Europe and North America: from trial and error to an evidence-based approach. – Biol. Rev. 90: 182–203. Landucci F., Tichý L., Šumberová K. & Chytrý M. (2015): Formalized classification of species-poor vegetation: a proposal of a consistent protocol for aquatic vegetation. – J. Veg. Sci. 26: 791–803. Lapshina E. (2010): Rastitěl'nost' bolot jugo–vostoka Zapadnoj Sibiri. – Izd–vo NGU, Novosibirsk.

Lawesson J.E. (2004): A tentative annotated checklist of Danish syntaxa. – Folia Geobot. 39: 73–95. Lebrun J., Noirfalise A., Heinemann P. & Vanden Berghen C. (1949): Les Associations végétales de Belgique. – Bull. Soc. Royal. Bot. Belg. 82: 105–199. Lindholm T. & Heikkilä R. (2006): Destruction of mires in Finland. In: Lindholm, T. & Heikkilä, R. (eds.) Finland – land of mires, pp. 179–192, Finnish Environment Institute, Helsinki. Lohila A., Aurela M., Hatakka J., Pihlatie M., Minkkinen K., Penttilä T. & Laurila T. (2010): Responses of N2O fluxes to temperature, water table and N deposition in a Northern boreal fen. – Eur. J. Soil Sci. 61: 651– 661. Malmer N. (1962): Studies on mire vegetation in the Archean area of southwestern Götland (south Sweden). I. Vegetation and habitat conditions on the Åkhult mire. – Opera Bot. 7: 1–309. Malmer N. (1985): Remarks to the classification of mires and mire vegetation – Scandinavian arguments. – Aquilo Ser. Bot. 21: 9–17. Malmer N. (1986): Vegetation gradients in relation to environmental conditions in northwestern European mires. – Canad. J. Bot. 64: 375–383. Malmer N., Svensson B.M. & Wallén B. (1994): Interactions between Sphagnum mosses and field layer vascular plants in the development of peat-forming systems. – Folia Geobot. Phytotaxon. 29: 483–496. Mälson K., Backéus I. & Rydin H. (2008): Long-term effects of drainage and initial effects of hydrological restoration on rich fen vegetation. – Appl. Veg. Sci. 11: 99–106. Marcenò C., Guarino R., Loidi J., Herrera M., Isermann M., Knollová I., Tichý L., Tzonev R.T., Acosta A.T.R., FitzPatrick Ú., Iakushenko D., Janssen J.A.M., Jiménez-Alfaro B., Kącki Z., Keizer-Sedláková I., Kolomiychuk V., Rodwell J.S., Schaminée J.H.J., Šilc U. & Chytrý M. (2018): Classification of European and Mediterranean coastal dune vegetation. – Appl. Veg. Sci. 21: 533–559. Martinčič A. (1995): Vegetacija razreda Scheuchzerio-Caricetea fuscae (Nordh. 36) R. Tx. 37 v Sloveniji. – Biol. vestnik 40: 101–111. Matuszkiewicz W. (2007): Przewodnik do oznaczania zbiorowisk roślinnych Polski. 3rd ed. – Paostwove Wydawnictwo Naukowe, Warszawa. Michl T., Dengler J. & Huck S. (2010): Montane–subalpine tallherb vegetation (Mulgedio-Aconitetea) in central Europe: large-scale synthesis and comparison with northern Europe. – Phytocoenologia 40: 117–154. Middleton B.A., Holsten B. & van Diggelen R. (2006): Biodiversity management of fens and fen meadows by grazing, cutting and burning. – Appl. Veg. Sci. 9: 307–316. Moen A. (1990): The plant cover of the boreal uplands of Central Norway. I. Vegetation ecology of Solendet nature reserve; haymaking fens and birch woodlands. – Gunneria 63: 1–451. Moen A., Lyngstad A. & Øien D.-I. (2012): Boreal rich fen vegetation formerly used for haymaking. – Nord. J. Bot. 30: 226–240. Moen A., Nilsen L.S., Øien D.-I. & Arnesen T. (1999): Outlying haymaking lands at Sølendet, central Norway; effects of scything and grazing. – Norsk Geogr. Tidsskr. 53: 93–102. Moeslund J.E., Arge L., Bøcher P.K., Dalgaard T., Odgaard M.V., Nygaard B. & Svenning J.-C. (2013): Topographically controlled soil moisture is the primary driver of local vegetation patterns across a lowland region. – Ecosphere 4: 1–26. Moravec J. (1989): Influences of the individualistic concept of vegetation on syntaxonomy. – Vegetatio 81: 29– 39. Mörnsjö T. (1969): Studies on vegetation and development of a peatland in Scania, South Sweden. – Oper. Bot. 24: 1–187. Mucina L., Bültmann H., Dierßen K., Theurillat J.-P., Raus T., Čarni A., Šumberová K., Willner W., Dengler J., Gavilán García R., Chytrý M., Hájek M., Di Pietro R., Iakushenko D., Pallas J., Daniëls F.J.A., Bergmeier E., Santos Guerra A., Ermakov N., Valachovič M., Schaminée J.H.J., Lysenko T., Didukh Ya.P., Pignatti S., Rodwell J.S., Capelo J., Weber H.E., Solomeshch A., Dimopoulos P., Aguiar C., Hennekens S.M. & Tichý L. (2016): Vegetation of Europe: Hierarchical floristic classification system of vascular plant, bryophyte, lichen, and algal communities. – Appl. Veg. Sci. 19, Suppl. 1: 3–264. Navrátilová J., Navrátil J. & Hájek M. (2006): Relationships Between Environmental Factors and Vegetation in Nutrient-Enriched Fens at Fishpond Margins. – Folia Geobot. 41: 353–376. Nekola J.C. (1999): Paleorefugia and neorefugia: the influence of colonization history on community pattern and process. – Ecology 80: 2459–2473. Nordhagen R. (1943): Sikilsdalen og Norges fjellbeiter. En plantesosiologisk monografi. – Bergens Mus. Skr. 22: 1–607. Nykänen H., Alm J., Lçng K., Silvola J. & Martikainen P.J. (1995): Emissions of CH4 , N2O and CO2 from a virgin fen and a fen drained for grassland in Finland. – J. Biog. 22: 351–357.

Oberdorfer E. (1957): Süddeutsche Pflanzengesellschaften. – Pflanzensoziologie 10: 1–564. Oberdorfer E. (ed.) (1998): Süddeutsche Pflanzengesellschaften. Teil I: Fels- und Mauergesellschaften, alpine Fluren, Wasser-, Verlandungs- und Moorgesellschaften. Ed. 4. – Gustav Fischer Verlag, Jena/Stuttgart/Lübeck/Ulm. Øien D.-I., Pedersen B., Kozub Ł., Goldstein K. & Wilk M. (2018): Long‐term effects of nutrient enrichment controlling plant species and functional composition in a boreal rich fen. – Appl. Veg. Sci. 29: 907–920. Øien D-I. & Moen A. (2001): Nutrients limitation in boreal plant communities and species influenced by scything. – Appl. Veg. Sci. 4: 197–206. Økland R.H. (1990a): A phytoecological study of the mire Northern Kisselbergmosen, SE Norway. II. Identification of gradients by detrended (canonical) correspondence analysis. – Nord. J. Bot. 10: 79–108. Økland R.H., Økland T. & Rydgren K. (2001): A Scandinavian perspective on ecological gradients in north-west European mires: reply to Wheeler and Proctor. – J. Ecol. 89: 481–486. Pakarinen P. (1995): Classification of boreal mires in Finland and Scandinavia: a review. – Vegetatio 118: 29–38. Passarge H. (1964): Pflanzengesellschaften des nordostdeutschen Flachlandes I. – Pflanzensoziologie 13: 1–324. Paulissen M.P.C.P., van Der Ven P.J., Dees A.J. & Bobbink R. (2004): Differential effects of nitrate and ammonium on three fen bryophyte species in relation to pollutant nitrogen input. – New Phytol. 164: 451–458. Pawlikowski P., Abramczyk K., Szczepaniuk A. & Kozub Ł. (2013): Nitrogen: phosphorus ratio as the main ecological determinant of the differences in the species composition of brown-moss rich fens in north- eastern Poland. – Preslia 85: 349–367. Pérez-Haase A. & Ninot J.M. (2017): Hydrological heterogeneity rather than water chemistry explains the high plant diversity and uniqueness of a Pyrenean mixed mire. – Folia Geobot. 52: 143–160. Persson Å. (1961): Mire and spring vegetation in an area north of lake Torneträsk, Torne Lappmark, Sweden. I. Description of the vegetation. – Opera Bot. 6/1: 1–187. Persson Å. (1962): Mire and spring vegetation in an area north of lake Torneträsk, Torne Lappmark, Sweden. II. Habitat conditions. – Opera Bot. 6/3: 1–100. Podani J. (1984): Spatial processes in the analysis of vegetation: theory and review. – Acta Bot. Hungarica 30: 75–118. Pott R. (2005): Allgemeine Geobotanik. Biogeosysteme und Biodiversität. – Springer, Berlin. Prentice C., Cramer W., Harrison S.P., Leemans R., Monserud R.A. & Solomon A.M. (1992): A global biome model based on plant physiology and dominance, soil properties and climate. – J. Biog. 19: 117–134. Rion V., Gallandat J.-D., Gobat J.-M. & Vittoz P. (2018): Recent changes in the plant composition of wetlands in the Jura Mountains. – Appl. Veg. Sci. 21: 121–131. Rivas-Martínez S. (ed.) (2011): Mapa de series, geoseries y geopermaseries de vegetación de Espaqa. Parte II. – Itinera Geobot. 18: 5–424. Roleček J. (2007): Formalized classification of thermophilous oak forests in the Czech Republic: what brings the Cocktail method? – Preslia 79: 1–21. Rozbrojová Z. & Hájek M. (2008): Changes in nutrient limitation of spring fen vegetation across environmental gradients in the West Carpathians. – J. Veg. Sci. 19: 613–620. Rozbrojová Z., Hájek M. & Hájek O. (2010): Vegetation diversity of mesic meadows and pastures in the West Carpathians. – Preslia 82: 307–332. Ruuhijärvi R. (1960): Über die regionale Einteilung der Nordfinnischen Moore. – Ann. Bot. Soc. Zool. Bot. Fenn. 'Vanamo' 31/1: 1–360. Růžička I. (1989): Výsledky záchranného výzkumu ohrožené květeny mizejících rašelinišť a rašelinných luk na Jihlavsku. – Vlastiv. Sborn. Vysočiny, Odd. Věd Přír. 9: 135–176. Rybníček K. (1964): Die Braunmoorgesellschaften der Böhmisch-mährischen Höhe (Tschechoslowakei) und die Problematik ihrer Klassifikation. – Preslia 36: 403–415. Rybníček K. (1974): Die Vegetation der Moore im südlichen Teil der Böhmisch-Mährischen Höhe. – Vegetace ČSSR A6, Academia, Praha. Rybníček K. (1981): Problematika klasifikace rašelinných společenstev. – Zpr. Čs. Bot. Spol., Mater. 2: 67–70. Rybníček K. (2005): Regional mire complex types in Europe. In: Bohn U., Hettwer C. & Gollub G. (eds), Application and analysis of the map of the natural vegetation of Europe, pp. 143–149, Bundesamt für Naturschutz, Bonn. [cit. sec. Joosten et al. 2017] Rybníček K., Balátová-Tuláčková E. & Neuhäusl R. (1984): Přehled rostlinných společenstev rašelinišť a mokřadních luk Československa. – Stud. ČSAV 1984/8: 1–124. Rydin H. & Jeglum J. (2006): The biology of peatlands. – Oxford University Press, Oxford. Rydin H., Sjörs H. & Löfroth M. (1999): Mires. – Acta Phytogeogr. Suec. 84: 91–112.

Sádlo J. (2000): Původ travinné vegetace slatin v Čechách: sukcese kontra cenogeneze. – Preslia 72: 495–506. Schaminée J.H.J., Chytrý M., Hennekens S.M., Janssen J.A.M., Jiménez-Alfaro B., Knollová I., Marceno C., Mucina L., Rodwell J.S., Tichý L. & data-providers (2016): Review of grassland habitats and development of distribution maps of heathland, scrub and tundra habitats of EUNIS habitats classification. Report EEA/NSV/15/005. – European Environment Agency, Copenhagen. Schaminée J.H.J., Hennekens S.M., Chytrý M. & Rodwell J.S. (2009): Vegetation-plot data and databases in Europe: an overview. – Preslia 81: 173–185. Schenková V., Horsák M., Hájek M., Plesková Z., Dítě D. & Pawlikowski P. (2014): Mollusc and plant assemblages controlled by different ecological gradients at Eastern European fens. – Acta Oecol. 56: 66– 73. Sekulová L., Hájek M., Hájková P., Mikulášková E. & Rozbrojová Z. (2011): Alpine wetlands in the West Carpathians: vegetation survey and vegetation-environment relationships. – Preslia 83: 1–24. Shmida A. & Ellner S. (1984): Coexistence of plant species with similar niches. – Vegetatio 58: 29–55. Singsaas S. (1989): Classification and ordination of the mire vegetation of Stormyra near Tynset, S Norway. – Nordic. J. Bot. 9: 413–423. Sjörs H. & Gunnarsson U. (2002): Calcium and pH in north and central Swedish mire waters. – J. Ecol. 90: 650– 657. Sjörs H. (1948): Myrvegetation i Bergslagen. – Acta Phytogeogr. Suec. 21: 1–340. Sjörs H. (1950a): On the relation between vegetation and electrolytes in north Swedish mire waters. – Oikos 2: 241–258. Sjörs H. (1950b): Regional studies in North Sweish mire vegetation. – Bot. Notis. 103: 173–222. Sjörs H. (1990): Divergent succession in mires, a comparative study. – Aquilo, Ser. Botanica 28: 67–77. Sjörs H., Björkbäck F. & Nordqvist Y. (1965): Northern mires. – Acta Phytogeogr. Suec. 50: 180–197. Steiner G.M. (1992): Österreichischer Moorschutzkatalog. Ed. 4. – Verlag Ulrich Moser, Graz/Wien. Storch D. (2016): The theory of the nested species-area relationship: geometric foundations of biodiversity scaling. – J. Veg. Sci. 27: 880–891. Succow M. (1974): Vorschlag einer systematischen Neugliederung der mineralbodenwasserbeeinflussten wachsenden Moorvegetation Mitteleuropas unter Ausklammerung des Gebirgsraumes. – Fedd. Repertorium 85: 57–113. Tahvanainen T. (2004): Water chemistry of mires in relation to the poor-rich vegetation gradient and contrasting geochemical zones of north-eastern Fennoscandian shield. – Folia Geobot. 39: 353–369. Topid J. & Stančid Z. (2006): Extinction of fen and bog plants and their habitats in Croatia. – Biod. Conserv. 15: 3371–3381. Tzonev R.T., Dimitrov M.A. & Roussakova V.H. (2009): Syntaxa according tо the Braun-Blanquet approach in Bulgaria. – Phytol. Balcan. 15: 209–233. Udd D., Mälson K., Sundberg S. & Rydin H. (2015): Explaining species distributions by traits of bryophytes and vascular plants in a patchy landscape. – Folia Geobot. 50: 161–174. Valachovič M. (ed.) (2001): Rastlinné spoločenstvá Slovenska. 3 Vegetácia mokradí. – Veda, Bratislava. van Diggelen R., Middleton B., Bakker J., Grootjans A. & Wassen M. (2006): Fens and floodplains of the temperate zone: Present status, threats, conservation and restoration. – Appl. Veg. Sci. 9: 157–162. Vicherová E., Hájek M., Šmilauer P. & Hájek T. (2017): Sphagnum establishment in alkaline fens: Importance of weather and water chemistry. – Sci. Tot. Envir. 580: 1429–1438. Wassen M.J., Olde Venterink H.G.M. & De Swart E.O.A.M. (1995): Nutrient concentrations in mire vegetation as a measure of nutrient limitation in mire ecosystems. – J. Veg. Sci. 6: 5–16. Waughman G.J. (1980): Chemical Aspects of the Ecology of Some South German Peatlands. – J. Ecol. 68: 1025– 1046. Wheeler B.D. & Proctor M.C.F. (2000): Ecological gradients, subdivisions and terminology of north-west European mires. – J. Ecol. 88: 187–203. Willner W., Jiménez-Alfaro B., Agrillo E., Biurrun I., Campos J.A., Čarni A., Casella L., Csiky J., Dušterevska R., Didukh Ya.P., Ewald J., Jandt U., Jansen F., Kącki Z., Kavgacı A., Lenoir J., Marinšek A., Onyshchenko V., Rodwell J., Schaminée J., Šibík J., Škvorc Ž., Svenning J.-C., Tsiripidis J., Turtureanu P.D., Tzonev R., Vassilev K., Venanzoni R., Wohlgemuth T. & Chytrý M. (2017a): Classification of European beech forests: a Gordian Knot? – Appl. Veg. Sci. 20: 494–512. Willner W., Kuzemko A., Dengler J., Chytrý M., Bauer N., Becker T., Biţæ-Nicolae C., Botta-Dukát Z., Čarni A., Csiky J., Igid R., Kącki Z., Korotchenko I., Kropf M., Krstivojevid-Duk M., Krstonošid D., Rédei T., Ruprecht E., Schratt-Ehrendorfer L., Semenishchenkov Y., Stančid Z., Vashenyak Y., Vynokurov D. & Janišová M.

(2017b): A higher-level classification of the Pannonian and western Pontic steppe grasslands (Central and Eastern Europe). – Appl. Veg. Sci. 20: 143–158. Zohlen A. & Tyler G. (2000): Immobilization of tissue iron on calcareous soil: differences between calcicole and calcifuge plants. – Oikos 89: 95–106.

Paper 1 Peterka T., Plesková Z., Jiroušek M. & Hájek M. (2014): Testing floristic and environmental differentiation of rich fens in the Bohemian Massif. – Preslia 86: 337–366.

Preslia 86: 337–366, 2014 337

Testing floristic and environmental differentiation of rich fens on the Bohemian Massif

Testování floristické a ekologické diferenciace bohatých slatinišť v Českém masivu

Tomáš P e t e r k a1, Zuzana P l e s k o v á1, Martin J i r o u š e k1,2 & Michal H á j e k1,3

1Department of Botany and Zoology, Faculty of Science, Masaryk University, Kotlářská 2, CZ-611 37 Brno, Czech Republic, e-mail: [email protected], pleskovicova @gmail.com, [email protected], [email protected]; 2Department of Plant Biology, Faculty of Agronomy, Mendel University in Brno, Zemědělská 1, CZ-61300 Brno, Czech Republic; 3Institute of Botany, Academy of Sciences of the Czech Republic, Department of Vegetation Ecology, Lidická 25/27, CZ-602 00 Brno, Czech Republic

Peterka T., Plesková Z., Jiroušek M. & Hájek M. (2014): Testing floristic and environmental differentiation of rich fens on the Bohemian Massif. – Preslia 86: 337–366.

The south-eastern part of the Bohemian Massif (the Bohemian-Moravian Highlands, the Třeboň basin, Czech Republic) is an important hotspot of fen biodiversity. Especially rich fens with calcium-tolerant peat mosses (the Sphagno warnstorfii-Tomentypnion alliance) currently harbour highly endangered organisms. In this study we gathered phytosociological and environmental (water chemistry, water table depth) data from 57 unique and well-preserved fens. The ISOPAM algorithm reproduced the expert-based classification at the alliance level presented in the Vegeta- tion of the Czech Republic monograph. Particular types of vegetation were nearly completely differentiated in the PCA of environmental data and all their pairs differed significantly with respect to pH, which together with calcium was correlated with the major vegetation gradient. The secondary gradient coincided with the concentration of nitrate and potassium, but was not apparent in the bryophyte subset. When only data for vascular plants were analyzed, the major gradient reflected increasing number of species from poor to extremely-rich fens, including ubiquitous grassland species, and only partially coincided with pH and calcium. Contrary to expectations, neither the extremely rich or rich fens were associated with low concentration phosphorus in the water. In addition, particular vegetation types did not differ in the N:P ratio of bryophyte biomass. Species composition of extremely rich fens thus seemed to be determined predominantly by a high pH/calcium level and waterlogging, low iron concentration and absence of sphagna that would hamper regeneration of some competitively weak vascular plants. We demonstrated that the delimitation of the major vegetation types (alliances) along the poor-rich gradient makes great floristic and ecological sense also in the Hercynian Mountains and that pH and calcium rather than nutrient availability differentiate causally major vegetation types by determining structure of the moss layer.

K e y w o r d s: Bohemian-Moravian Highlands, bryophytes, classification, gradients, ISOPAM, mire, Třeboň basin, vegetation

Introduction Fens (minerotrophic mires of the Scheuchzerio palustris-Caricetea nigrae Tüxen 1937 class) are remarkable habitats with a specific species composition. In central Europe they are among the most endangered habitats, hosting a large number of ecological specialists and rare species of different taxonomic groups (Grootjans et al. 2005, Poulíčková et al. 2005, Schenková et al. 2012, Hettenbergerová et al. 2013). Their botanical and zoological species compositions vary predominantly along a complex gradient of pH, calcium and 338 Preslia 86: 337–366, 2014 total mineral richness, usually called the “poor-rich gradient” (du Rietz 1949, Sjörs 1952, Fransson 1972, Malmer 1986, Tahvanainen 2004, Hájek et al. 2006, Conradi & Fried- mann 2013). The old ecophysiological premise that mineral-rich soils are also richer in nitrogen, phosphorus and potassium was abandoned, because in fens the gradients of increasing N, P and K availabilities can be largely independent of the pH/calcium gradient or may even correlate with pH negatively (Waughmann 1980, Wheeler & Proctor 2000, Bragazza & Gerdol 2002, Bragazza et al. 2005, Rozbrojová & Hájek 2008, Kooijman & Hedenäs 2009). On the other hand, either an over supply or a deficiency of a particular element may underlie the observed poor-rich vegetation gradient. Productivity of the most calcium-rich fens is strongly limited by phosphorus that is immobilized by calcium into forms unavailable to plants (Boyer & Wheeler 1989, Bedford et al. 1999). Some authors (Paulissen et al. 2004, Kooijman & Hedenäs 2009) further stress the importance of particular forms of nitrogen (ammonium versus nitrate) whose ratio may change along a pH gradient. Assessment of changes in nutrient availability along a poor-rich gradient is however difficult because of great seasonal variation in macronutrient concentrations in water (Hájek & Hekera 2004, Jiroušek et al. 2013). The alternative way of assessing the nature of nutrient limitation, plant stoichiometry (Güsewell & Koerselman 2002, Olde Venterink et al. 2003, Rozbrojová & Hájek 2008, Pawlikowski et al. 2013), is on the other hand affected by differences in element concentrations among the species (Malmer et al. 1992, Wojtuń 1994, Bombonato et al. 2010). In addition, some other factors apart from pH/calcium level and macronutrient availability may contribute somehow to forming the main vegetation gradients, such as, among others, water table depth and its dynamics (Bragazza & Gerdol 1996, Jablońska et al. 2011, Schenková et al. 2014), iron toxicity (Rozbrojová & Hájek 2008, Aggenbach et al. 2013) or historical-biogeographical factors (Nekola 1999, Hájek et al. 2011b, Jiménez-Alfaro et al. 2012). The relationships between nutrient availabilities and species composition of fen vegetation are therefore not definitively resolved and studies from other than traditionally explored regions, especially those that display specific patterns of water chemistry, are needed. Understanding these relationships is a prerequisite of better conservation of endangered fen species in agriculture landscapes where high nutrient input seriously threatens persisting fen remnants (Zechmeister et al. 2002, Navrátilová et al. 2006, Koch & Jurasinski 2014). In the boreal zone of Europe there have been numerous studies testing the importance of particular environmental factors for the species composition of fen communities (Persson 1962, Mörnsjö 1969, Malmer 1986, Heikkilä 1987, Sjörs & Gunnarsson 2002) and the same holds true for North America (Vitt & Chee 1990, Anderson & Davis 1997, Nekola 2004). In central Europe, the relationship between environmental variables and species composition of minerotrophic mires have been studied mainly in spring fens in the Alps (Gerdol 1995, Bragazza & Gerdol 1999, Conradi & Friedmann 2013, Sekulová et al. 2013) and Western Carpathians (Hájek et al. 2002, Hájková & Hájek 2004, Sekulová et al. 2011, Koczur & Nicia 2013). The fens on the Bohemian Massif are little studied despite the fact that they are an important but deteriorating hotspot of central-European fen biodiversity. The regions in the Bohemian-Moravian Highlands and Třeboň basin are especially important for conservation of central-European fen biodiversity. This landscape is exceptional within the Czech Republic in the occurrence of minerotrophic mires (Divíšek et al. 2014) and especially in the occurrence of rich fens with calcium-tolerant peat mosses of the Sphagno warnstorfii-Tomentypnion alliance (Rybníček et al. 1984, Peterka et al.: Differentiation of rich fens 339

Hájek & Hájková 2011). The latter type of vegetation, despite having been seriously affected by human activities since the 1970s (e.g. Růžička 1989), still harbours highly endangered vascular plants such as Carex limosa, C. dioica, C. chordorrhiza and Tricho- phorum alpinum (Růžička 1999, Navrátilová & Navrátil 2005), bryophytes such as Meesia triquetra, Paludella squarrosa and Hamatocaulis vernicosus (Štechová et al., in prep.) and invertebrates such as the glacial relict snails Vertigo geyeri and V. liljeborgii that are extremely rare in temperate Europe (Schenková & Horsák 2013, Schenková et al. 2013). Currently there is only data on the vegetation-environmental relationships for the Třeboň basin (Navrátilová et al. 2006) and for only small regions of the Bohemian-Moravian Highlands (Rybníček 1974, Štechová et al. 2012, Peterka 2013). Therefore there is need for a study that covers the entire hotspot area and directly analyses the relationships between water chemistry and vegetation diversity. The specific question that should be addressed by such a study on the Bohemian Massif is the floristic and environmental delimitation of the Sphagno warnstorfii-Tomentypnion fens. This vegetation alliance is well known among Czech vegetation scientists and nature conservationists because it includes a large number of red-listed plants and animals. Sur- prisingly, the Sphagno warnstorfii-Tomentypnion alliance is not currently recognized in neighbouring Germany (Pott 1992, Berg et al. 2004), Poland (Matuszkiewicz 1982) and even Austria, which shares the eastern part of the Bohemian Massif with the Czech Repub- lic (Steiner 1993, Zechmeister & Steiner 1995). Analogous vegetation types are recognized in these countries only at a very fine (subassociation) level (Steiner 1992). The reason is in these countries different classification criteria are used to delimit alliances and associations. In Germany and Austria authors predominantly use a syntaxonomical system based on floristic differentiation by dominance of vascular plants specialized to fens but having a wide pH-niche. This system was introduced by Oberdorfer (1957, 1998) and Dierssen (1982) and accepted in many other vegetation surveys across Europe (e.g. Steiner 1992, Coldea et al. 1997, Lájer 1998, Jermacâne & Laivin‚š 2001). In this system, the major division is between topogenic, extremely waterlogged fens (Caricion lasiocarpae, Rhynchosporion albae) and spring fens plus fen grasslands (Caricion davallianae, Caricion fuscae). An alternative classification at the alliance level reflects the “poor-rich” gradient as the main compositional change within fens and in particular emphasising the role of bryophytes. This concept was introduced by Fennoscandian botanists (du Rietz 1949, Dahl 1956, Persson 1961, Eurola 1962, Moen et al. 2012) and was adopted among others in the former Czechoslovakia (Rybníček et al. 1984) and thereafter in Czech and Slovak republics (Dítě et al. 2007, Hájek & Hájková 2011). The Sphagno warnstorfii- Tomentypnion fens are a separate unit in this system. Some syntaxonomical systems are transitional but the Sphagno warnstorfii-Tomentypnion alliance or an analogous alliance is distinguished in some regions of Bulgaria (Hájek et al. 2008), France (Gillet 1982), Scot- land (Prentice & Prentice 1975), Russia (Koroleva 2001, 2006, Lapshina 2010), Green- land (Molenaar 1976) and partially in Italy (Gerdol & Tomaselli 1997) and Ukraine (Felbaba-Klushina 2010). Hájek et al. (2006) have demonstrated that in the Western Carpathians and Balkans the four vegetation types delimited along the poor-rich gradient (Caricion davallianae, Sphagno warnstorfii-Tomentypnion nitentis, Caricion fuscae [=Caricion canescenti-nigrae], Sphagno-Caricion canescentis) are well separated in terms of water pH and total mineral richness, i.e. the factors that shape the major gradient in vegetation. However, one may argue that this pattern may not be so simple in regions 340 Preslia 86: 337–366, 2014 with different water chemistry, or in regions where topogenic fens are common. The chal- lenge is therefore to test directly the floristic delimitation of the major alliances distinguished along a poor-rich gradient using numerical classification on the Bohemian Massif. The main aims of this study are summarized as follows: (i) to reveal major gradients in species composition of vascular plants and bryophytes in fens in the eastern part of the Bohemian Massif and their relationships to water chemistry, pH and water table depth, (ii) to test the validity of delimiting the main vegetation alliances as parts of a poor-rich gradient, including the Sphagno warnstorfii-Tomentypnion alliance, on the Bohemian Massif, (iii) to test the differences in environmental factors among these alliances.

Material and methods Study area The Bohemian Massif is a large crystalline massif located in the central part of the Czech Republic, eastern Germany, southern Poland and northern Austria. The study area (Fig. 1) is in the south-eastern part, namely the Bohemian-Moravian Highlands and Třeboň basin, where there is a great diversity and wide distribution of fens (Hájek & Hájková 2011). Two localities were sampled on the East Bohemian cretaceous table close to the boundary with the Bohemian-Moravian Highlands. The geological substrate of the Bohemian-Moravian Highlands consists mostly of crystalline rocks of proterozoic and paleozoic age, i.e. of different kinds of gneiss, migmatite, granite, granodiorite or phyllite with small bodies of amphibolites, marbles, serpentinites or erlans. Calcium-rich cretaceous sandstone and claystone occur locally in the “Dlouhé meze” area. The Bohemian-Moravian Highlands are in the cold-temperate climatic region with a mean annual temperature of 5.0–6.5 °C and mean annual precipitation of 600–900 mm (Čech et al. 2002). The altitude at the localities studied ranged between 450 and 730 m a.s.l. The geological bedrock in the Třeboň basin is made up of siliceous cretaceous and ter- tiary sandstones. The climate is temperate with a mean annual temperature of 7.8 °C and mean annual precipitation of 600–700 mm (Albrecht et al. 2003). All the study sites are located at altitudes of between 410–480 m a.s.l.

Vegetation data sampling We sampled all the well-preserved fens with rare species in the study area. We omitted only those fens with abundance of grassland species, which are usually drained or eutrophicated; they mostly belong to the Calthion alliance. Further, we omitted some depauperate poor fens that lacked rare species in order to balance the data set, because poor fens prevail over rich and extremely rich fens in the current Bohemian Massif landscape. Usually a single phytosociological relevé was gathered per fen (in central, visually the most preserved part), but in some cases, two plots of two distinct types of vegetation (according to Hájek et al. 2006) were sampled at one large well-preserved locality. Both vascular plants and bryophytes were recorded within each plot (16 m2). Their cover was estimated using the nine grade Braun-Blanquet’s scale (van der Maarel 1979). Altogether the vegetation at 57 plots (see Table 1) was recorded. Bryophytes were collected from the plots and their identification was confirmed or revised using light microscope. Coordinates of relevés were obtained Peterka et al.: Differentiation of rich fens 341

Fig. 1. – Map of plots studied on the Bohemian Massif. Their classification as particular types of fens follows the results presented in Table 1. using the WGS84 system. Nomenclature of vascular plants follows Danihelka et al. (2012). Nomenclature of bryophytes follows Kučera et al. (2012), but the species Plagiomnium affine, P. elatum, P. ellipticum and P. medium were merged into the Plagiomnium affine aggregate, because of their similar indication values within fens and identification uncertainities in the case of some specimens. By analogy, Chiloscyphus polyanthos and C. pallescens as well as Campylium stellatum and C. protensum were merged. For author references of syntaxa see Hájek & Hájková (2011).

Sampling of environmental data In summer 2011, the following environmental factors were recorded in each vegetation plot in the field using shallow bore holes dug in the peat: pH, corrected conductivity and water table depth (WTD). Samples of groundwater and biomass of (1) 2–3 (4) dominant species of moss were also collected (for details see Electronic Appendices 1, 2). Water pH and conductivity, both standardized at 20 °C, were measured in situ using portable instruments (GMH 3410 and GMH 3530 Greisinger). Conductivity due to H+ was subtracted (Sjörs 1952). Water table depth was expressed as the mean distance between the surface of the moss cushion (i.e. the apical part of acrocarpous and pleurocarpous mosses or capitula of Sphagna) and actual water level. Water samples were also collected from shallow bore holes, water was pumped out of them, which were then allowed to refill before sampling. Water samples were immediately filtered through microfibre glass filters, Fisher F261, with pores of 1.2 μm and placed in plastic bottles. Preservatives were added to two separate samples: for metallic elements (0.5 ml of 65% HNO3 per 100 ml of sample) and for anions 342 Preslia 86: 337–366, 2014

(0.3 ml of chloroform per 100 ml). The bottles were kept and transported to the laboratory in a portable fridge. For the analysis of N:P ratios in the biomass of moss, capitulas of sphagna and apical segments of other species of moss (length of about 2 cm) were collected using clean stainless-steel tweezers. All biomass samples were put in paper bags and left to dry out.

Water and biomass analyses

+ - 3- + 2+ 2+ Water samples were analyzed for concentrations of NH4 ,NO3 ,PO4 ,K,Ca ,Mg and Fe. Ammonium, nitrates and phosphates were analyzed using flow injection analysis (FIA). Other elements were determined using an atomic absorption spectrometer (AAS) novAA ® 350 (Analytik Jena AG). Flame method was used for all the above mentioned elements and ions. Lanthanum chloride was used as an ionization suppressant for calcium and magnesium analyses, and cesium chloride was used as the deionization agent when determining potassium. Concentrations of nitrogen and phosphorus in bryophyte biomass were determined using FIA after the dried moss shoots were digested in acid (perchloric acid for determining total phosphorus and Kjeldahl digestion with sulphuric acid for ammonium).

Data processing The phytosociological relevés were exported into JUICE 7.0 software (Tichý 2002). In order to test the applicability of the classification system based on four vegetation types (alliances) of fens, defined as part of the pH/calcium gradient (Hájek et al. 2006, Hájek & Hájková 2011), we used an unsupervised non-hierarchical numerical classification algorithm ISOPAM (Schmidtlein et al. 2010) at the level of four clusters. The ISOPAM algorithm is based on the classification of ordination scores from isometric feature mapping. Ordination and classification are repeated in a search for groups rich in diagnostic species and high overall fidelities of species to particular clusters. This approach is beneficial when the data sets have a bad “signal to noise ratio” (Schmidtlein et al. 2010) such as those for small fens and fen grasslands where a few fen specialists are accompanied by a high number of more ubiquitous wetland and grassland species coming from the surroundings. In the ISOPAM algorithm, we applied default threshold for diagnostic species filtering and the Bray-Curtis distance to ordinate the relevés. The ISOPAM algorithm considers only the presence-absence data. We identified three spuriously classified relevés (no. 2, 26 and 55 in Table 1) whose dominant species and/or overall species composition deviated from the characterization of this vegetation type (alliance) in the Vegetation of the Czech Republic monograph (Hájek & Hájková 2011). We checked their assignment to the resulting clusters using the normalized weirdness method (van Tongeren et al. 2008), which indicates they be placed in a more appropriate group. The diagnostic species of a particular cluster were determined using the phi-coefficient (Chytrý et al. 2002) with the size of all the groups standardized to the same size. For table presentation (Table 1) the species with fidelity to a particular type of vegetation with a phi > 0.3 were regarded as diagnostic. The significance of fidelity was verified using Fisher’s exact tests (P < 0.05). Three grassland generalist that appeared to be diagnostic for a small group of extremely-rich fens (Agrostis capillaris, Cirsium rivulare, Vicia cracca), although they occurred in only two relevés with low abundance, were reclassified as accompanying species. Peterka et al.: Differentiation of rich fens 343

Main gradients in the floristic composition of relevés were assessed using detrended correspondence analysis (DCA). The vegetation data were further subjected to canonical correspondence analysis (CCA) to find the best significant predictors of species composition using a forward selection procedure and the Monte Carlo permutation test (499 permuta- tions) with Holm correction of P values (referred to as Padjust). Three vegetation matrices, with transformed cover values (arcsin transformation) and down weighting of rare species were subjected to correspondence analyses (both DCA and CCA): (i) both vascular plants and bryophytes, (ii) only vascular plants and (iii) only bryophytes. With the exception of pH, no environmental variables had normal or uniform distributions (Shapiro-Wilk test). There- fore, their values were logarithmically transformed to approximate a normal distribution. The two major gradients were ecologically interpreted by a posteriori plotting the isolines of measured environmental factors using generalized additive models (GAMs) with Poisson distribution and stepwise selection of complexity using Akaike information criteria. Species richness of vascular plants and bryophytes were also modelled in order to illustrate changes in diversity of both taxonomic groups along the main floristic gradients. Principal component analysis (PCA) was used to describe the relationships between particular environmental variables, and check whether the group of measured environmental factors as a whole describes sufficiently the floristic differences between the different vegetation types. We applied PCA to the environmental data matrix, with centering by particular environmental variables, and plotted the delimited types of vegetation a posteriori onto the resulting plot. The CANOCO 5 package (Šmilauer & Lepš 2014) was used for the ordination analyses and GAM modelling. The N:P ratio in moss biomass was not included in the multidimensional analyses because of lack of statistical independence; moss element concentrations depend not only on the environment but also on species iden- tity (Hájek et al. 2014) and their effects may therefore be overestimated when environmental factors are confronted with the results of PCA or DCA. Significance of differences in measured environmental factors among the different vegetation types was tested using one-way ANOVAin the STATISTICA software (version 12, StatSoft Inc.) and the unequal N HSD post-hoc test. As conductivity, magnesium content and water table depth did not meet the assumptions required for parametric ANOVA (homoscedasticity in most cases) even after the logarithmic transformation, nonpara- matric Kruskal-Wallis statistics were used in these cases. We further compared the major vegetation types in terms of the N:P ratio in bryophyte samples to determine whether the observed differences in P concentration in water coincide with the N:P ratio, which may indicate P-limitation of aboveground production.

Results Classification of vegetation The ISOPAM algorithm at the level of four clusters gives a result that is similar to the expert-based classification presented in the Vegetation of the Czech Republic with the exception of only three relevés. In combination with the matching produced by the normalized wierdness method the end result was in complete agreement with the expert-based national vegetation classification. In particular four major vegetation types (Table 1) could be characterized as follows: 344 Preslia 86: 337–366, 2014

Table 1. – Phytosociological table of individual relevés with Braun-Blanquet cover codes (a = 2a; b = 2b). Diag- nostic species of particular vegetation types (in bold) are sorted according to fidelity, other species are sorted according to frequency. Codes in second column refer to red lists status of species (according to Grulich 2012, Kučera et al. 2012), except for category LC of bryophytes. Shortened names of localities and geographical coordinates are listed below the table. For the full header data see Electronic Appendix 1.

Relevé Nr. 1111111111222222222 2333333333344444 4444455555555 123456 7890123456789012345678 9012345678901234 5678901234567 Caricion davallianae Triglochin palustris C2 ++.b.+ ...... +...... Palustriella commutata a+...+...... Phragmites australis 1.a.ra ...a...+a..a...... 1...... +1...... Blysmus compressus C2 . .1a...... Eupatorium cannabinum +...1...... Carex davalliana C2 1...a1 ...... b....ab..a.a...... Carex paniculata C4a . ...a+ ...... +...... Fissidens adianthoides LC-att ++...+++...... +...1.+....+...... Scorpidium cossonii LR-nt 55...b ...... b31...++....+......

Sphagno warnstorfii-Tomentypnion Sphagnum warnstorfii LC-att ...... a. 53a1a. 1++b+4414aa443 . . a...... Anthoxanthum odoratum ...... ++11+1+++. 11+. +a+11++. ...1...... +.1.+...... Luzula multiflora ...... +++1++. ++. ++. . +++++. r + . . +...... +...+...... +. Sphagnum contortum LR-nt ...... 3..b+1..a1b.+1+.1+.+.4....4.+...... Festuca filiformis ...... +. . +++. +.....++..a+.+ ...... Sphagnum teres ...... 141335a3. . 33+b1+a11+11 1...+a.+33..34b. ....+...... Trichophorum alpinum C2 ...... +.....r1+1+...+....+.+...... Carex pulicaris C2 ...... 1....++.1..11..+1...... Briza media ...+.r ++1+.+1.+.11+1++..+11. ...+...... +...... Plagiomnium affine agg. ++. +1+ 11+++++++++13+++. ++1++ ...... +++...... Valeriana dioica C4a ++. 1. a 11+1+++++++1111abba1b+ .....+.r11+.1a1...... Holcus lanatus ...+.. +. 1. ++++r +1a+++1. r . . +. . . 11...... 11...... +...... Tephroseris crispa C4a ...... +.+++..++.....+r+...... r...... Galium uliginosum ...+++ +++++++++. ++++++++++++ . . ++. . . ++++. ++++ ...... Cirsium palustre ..1+.. +++++. ++++1+1++++11+. + r . 1+. . . +++. +++++ ...... Bryum pseudotriquetrum 11b1. + +++++++. ++. +1+++. +++++ ...... +...... Carex echinata ...... +. 1++1. 111+11. +1+a++. + . . 11...b++.a+a.+ .....+...+.a. Paludella squarrosa EN ...... r..+..+..+..+1...... Breidleria pratensis LC-att ...... 1+...+...++++1. 1+. . . . 1...... 1+...... Dactylorhiza majalis C3 .+...+ +++..+.+..1+.+.++.+.++...... +...... Potentilla erecta ++..+b +11+1111++11+1+1++111+ . . 1+....++.+1+++....+...++.+. Carex panicea b1. 11a 3aa1111a+b1+ba33baa333 ...+.1.+.a.+a3ba ....a...... +. Philonotis fontana ...... +...+...1...... ++...... Calliergonella cuspidata +..1a1 1. 11a1++3+1+31+b14aa++ . . +. 11++31b. b+b. ....+...... Carex demissa ...+.. +...+..1++..+.+++1...+ ..1.....+....+...... Aulacomnium palustre ...+.+ . b1+111. +. b+++a++1b11+ . . . ++++. ab+. . ++. . +. . +....+.+. Crepis paludosa +....+ 11...1...+.1.++a111+a+ ...... aa+++...... Festuca rubra agg. . . 1+r . ++++++1+. . +1+1+1+. ++1r +++....+++. ++1++ ...... +. Equisetum fluviatile ...... +1.++.1..++1+.++r.+.. +.+.+.++.r++...... +...... Carex flava C4a ...... ++.++...... Sphagnum subnitens LC-att ...... 1...... +.+.+...... Menyanthes trifoliata C3 .b...a . ++13.++..+aa3.+...b.1 ...+aa.4...... Ranunculus acris ....+. r +++. +.....++++...... r.+...... Chiloscyphus pallescens/polyanthos ...... +....++.+r.. ..+...... Myosotis nemorosa ...... +....+.....+r+...... r...... Succisa pratensis .+...... +.+.+..++.+1...+a+ ..++...+.+...... Climacium dendroides ...+.+ a1...+..+..1++.++.a+...... +.1+...... Peterka et al.: Differentiation of rich fens 345

Relevé Nr. 1111111111222222222 2333333333344444 4444455555555 123456 7890123456789012345678 9012345678901234 5678901234567 Caricion canescenti-nigrae Carex canescens ...... ++.++...... b11.1.+1+1++.1++...... +.++ Galium palustre agg. . . 1+. . . . ++++...++.++...+...+ ++++. +++r +1. ++++ ...... Veronica scutellata C4a ...... ++r.++...... Agrostis canina ...... +.++++++. . ++1+++++++++ 1111++111+++1b1+ . . . +++. ++. . a+ Comarum palustre C4a ...... b+aa11..ba.+...... + . baa3aa.+33...4. ...a...... Straminergon stramineum ...... +++++++. +. . +. ++1++. + ++++++11++. . ++++ . . +. +. . . +++++ Bistorta officinalis ...... r...... +...... ++.11.+...... Carex nigra . +++. . ++1+1ba. 1. a1a1. . 1+. +. + b1b+++11+31+51a1 . +++++. . +1++. Ranunculus flammula ...... +...... +...... ++..+1......

Sphagno-Caricion canescentis Polytrichum commune ...... +...... 1....+...... ++++. 1+. +baa+ Sphagnum fallax ...... 1..++a..++...... 5...... 3.....+.1 543ba51154515 Vaccinium oxycoccos C3 ...... 1.11...... 1a.1aa1...1.. Pinus sylvestris juv. ..+...... ++...... +1...... ++++++++..... Sphagnum papillosum ...... +...... 3b.4...1.... Rhynchospora alba C2 ...... r...... +.1.1+..... Drosera rotundifolia C3 ...+...... +.+++1+...+.+.+r+1.....1+...... ++111++++.... Calluna vulgaris ...... +...... ++..+...... Picea abies juv...... +...... +...... ++.. Utricularia ochroleuca C1 ...... +...... +..++..... Avenella flexuosa ...... a.+. Sphagnum capillifolium ...... 1..1......

Caricion davallianae and Sphagno warnstorfii-Tomentypnion Tomentypnum nitens LR-nt +114. 4 . a+1. +. +...1.1b41+abb...... Campylium stellatum LR-nt 11+a+b +1. +++1+131++1+.+.+1+1..+.+.+.+......

Sphagno warnstorfii-Tomentypnion and Caricion canescentis-nigrae Viola palustris ...... +. +++1+1+++++++. 1a++++ 1+1+. +1a+111++++ ...++...rr.1.

Other species Eriophorum angustifolium +. +1. . ++aa+111++1+a1+14+ab+b 1a1++1+113. 1+b3a ba+1+11b1. ++. Lysimachia vulgaris ..1++. ....+.++++++. +++++. . ++ . r +. 1+1+r ++1. +++ ...++..+...+. Carex rostrata ++a14. ...++.b.+.+1a+.+...+.+...+11+++++...... a+a..+....3 Epilobium palustre C4a . . a++r . . +. . ++. ++. ++. . r +. . r . r . . +++. . +r ++r ++++ ...+..r...... Molinia caerulea agg. 1+a..+ .a...+...1+...3+...a1. ...1+a...a+...... +a.a11+a.1.. Sphagnum palustre ...... +.b.+.1a..1...a+.+ .++411+. . +...1.3 ...4.+51.1.1. Rumex acetosa ..a+.. +.+..+++..+++..++.+...... +r+.++++...... r. Sphagnum flexuosum ...... a...+a1..1+a.1b.a.....5.b14 ..ba1.1....5. Cardamine pratensis . . ++. . . . +. +++. . . ++. +. . . +++. + +...... +++. +...... Equisetum palustre +1...+.+1...... 1..++..+++. +...... +.++.b+...... Juncus articulatus ...+++....++.+.+1.+1+++....+.....+...... +...... +...... Juncus effusus ...... +++....++r...... +...+...+++++ .....++..+.+. Lychnis flos-cuculi ..+.+. +.++.+....+..+..r.r..+ ...... r..+++...... Aneura pinguis .+.+.. .+..+..+++....+.+....+..+..++...... Angelica sylvestris ....+. ..+...+.+.++...rr..+.. ..+.....+.....1...... Alnus glutinosa juv...... + ....r1...+.....+.+..++.....1..+..+.+...... Sanguisorba officinalis ...... +1...... 1++.+...... a1++...... 1. Juncus bulbosus ...... +....+...... r1+..+.....+...... +.. ...++..1..... Filipendula ulmaria ....+r .+...... a.+.r...+.r ..a...... +...... Mentha arvensis ....+. +....++...+..+...+...... ++....1...... Ranunculus auricomus agg...... +...... +.+.r+...... ++..+++...... Peucedanum palustre ...... +..+.....+.. ...++++. 1...... +...+..... 346 Preslia 86: 337–366, 2014

Relevé Nr. 1111111111222222222 2333333333344444 4444455555555 123456 7890123456789012345678 9012345678901234 5678901234567 Hamatocaulis vernicosus VU .1.+.. .+.+...... +...+.+.+...... +...... Parnassia palustris C2 .+.1.. .+.+....+.+.++...... +...... Salix aurita juv. ...+.. ...++...+...... r...+ ..+..++...... Juncus conglomeratus ...... +..+....r.+.+...r...... +++...... Polytrichum strictum ...... r.+.1...... +.. ...+...... 1.++...+.... Betula pendula juv...... +...... +.++.....+1...... +....+.+. Caltha palustris ...... +....r++...... +...... +.+..+.r...... Equisetum arvense ...+r. 1....++...... +r.+...... Frangula alnus juv...... + ...... +...+...... +.+...... +...... ++...... Sarmentypnum exannulatum ...... 1+...... ++...11+...... +...... Eriophorum latifolium C2 ++.... 1b...... 1.....+.....+...... Carex diandra C2 ...a...... b..+.....1...... +.....1....b...... Cirriphyllum piliferum ....++...... ++.+...... +..+...... Nardus stricta ...... +...... ++...... +...... +.+...... +.... Sphagnum subsecundum ...... 11...... 1..3b...a...... a...... Carex lasiocarpa C3 ...... +.+...... 1.4.1...... +..1...... Equisetum sylvaticum ...... +....+.+...... r+...... +.+. Linum catharticum .+.... .+...... +..+.r...... + ...... Lycopus europaeus ...++. ....r...... +...... +...... r..... Scutellaria galericulata ....+...... +...... ++++...... Lysimachia thyrsiflora C3 ...... r...++...... ++.1...... Rhytidiadelphus squarrosus ...... +..+..1...... ++...... +. Calliergon giganteum VU .+.+...... +...... +.1...... Carex dioica C1 ...... 1....+...... +.....+...... +...... Anemone nemorosa ...... +++. 1...... +...... Epipactis palustris C2 +....a ...... 1....a...... Scirpus sylvaticus ...+a...... +...... +... Lythrum salicaria ...+...... r...... ++...... Salix pentandra juv. C4a ....+. .+.+...+...... Salix cinerea juv...... +.....+...... a...... +...... Quercus petraea juv...... +...... +...... +...+...... Juncus filiformis ...... +...... +.+....+...... Sphagnum inundatum ...... +...+...... +.. ...1...... Carex lepidocarpa C2 .a...... 1...... r...... Cirsium rivulare ....++...... +...... Ranunculus repens ....+. +...... r...... Vicia cracca ....r+ .+...... Dicranum bonjeanii LR-nt ...... +...... +...... +...... Leontodon hispidus ...... +...... a...1...... Carex chordorrhiza C1 ...... 1...... a...... +.. Juncus bufonius agg...... +++...... Brachythecium mildeanum ...... +1+...... Prunella vulgaris ...... 1+..+...... Calliergon cordifolium ...... +...... 4.+...... Utricularia intermedia C1 ...... 1.1...... 1...... Agrostis capillaris ..1.+...... Poa trivialis ..1...... +...... Polygala amarella C4b .....+ .+...... Selinum carvifolia ...... 1...... +...... Lathyrus pratensis ...... ++...... Geum rivale ...... +...... 1...... Carex flacca ...... +...... a...... Sphagnum angustifolium LC-att ...... 1...... 5...... Betula pubescens juv...... 1...... +...... Sphagnum auriculatum ...... 1...... 5..... Peterka et al.: Differentiation of rich fens 347

Relevé Nr. 1111111111222222222 2333333333344444 4444455555555 123456 7890123456789012345678 9012345678901234 5678901234567 Carex limosa C2 ...... r...... a...... Riccardia multifida LC-att ...... +.+...... Drosera anglica C1 ...... +...... +...... Pleurozium schreberi ...... +...... +...... Luzula sudetica C3 ...... +...... r...... Achillea millefolium agg...... +...... r...... Danthonia decumbens ...... +...+...... Carex hostiana C2 ...... +.....+...... Carex hartmanii C4a ...... 1...... +...... Chiloscyphus cuspidatus ...... +...... +. Sorbus aucuparia juv...... rr...... Vaccinium myrtillus ...... +...... +...... Hypericum maculatum ...... r...... +...... Sphagnum fimbriatum ...... 4...4...... Brachythecium rivulare ...... 1...... +...... Deschampsia cespitosa ...... +...... +. Sphagnum russowii ...... 1...... +... Carex elongata ...... +...... 1...... Carex elata C2 ...... 1...... +...... Holcus mollis ...... +...... +. Calamagrostis villosa ...... a ...... 3... Trientalis europaea C4a ...... + ...... a... Eriophorum vaginatum ...... + ...... 1..

Species recorded within one relevé. Vascular plants: Gymnadenia densiflora (C1) 1: +; Eleocharis quinqueflora (C1) 2: a; Utricularia minor (C2) 2: +; Typha angustifolia 4: r; Galium mollugo agg. 5: +; Poa pratensis 5: +; Aegopodium podagraria 5: +; Calamagrostis epigejos 5: +; Acer pseudoplatanus juv. 7: +; Carex appropinquata (C3) 8: a; Lotus corniculatus 8: +; Dactylorhiza fuchsii (C4a) 10: +; Laserpitium prutenicum (C3) 13: +; Juncus alpinoarticulatus (C3) 15: +; Drosera intermedia (C1) 15: +; Pinguicula vulgaris (C2) 15: +; Drosera ×obovata 15: +; Quercus robur juv. 16: +; Poa palustris 18: +; Cirsium heterophyllum 19: +; Equisetum ×litorale 19: +; Persicaria maculosa 20: +; Scorzonera humilis (C4a) 22: b; Carex pilulifera 22: +; Alchemilla sp. 22: +; Listera ovata (C4a) 25: +; Primula elatior 25: +; Ajuga reptans 25: +; Maianthemum bifolium 25: +; Salix euxina juv. 28: r; Carex vesicaria 31: +; Calamagrostis canescens 33: 1; Crepis mollis subsp. succisifolia (C3) 37: +; Pedicularis palustris (C1) 39: +; Eleocharis mamillata (C4a) 39: +; Lotus pedunculatus 41: 1; Pedicularis sylvatica (C2) 42: +; Fraxinus excelsior juv. 42: +; Dryopteris sp. 43: +; Sparganium natans (C2) 48: 1; Typha latifolia 48: +; Nymphaea candida (C1) 48: +; Vaccinium vitis-idaea 49: +; Eriophorum gracile (C1) 49: +; Galium saxatile 54: +; Senecio nemorensis agg. 54: +; Andromeda polifolia (C2) 55: 1; Melampyrum pratense 55: +. Bryophytes: Philonotis calcarea (LC-att) 2: 1; Cratoneuron filicinum 3: 4; Atrichum undulatum 7: +; Calliergonella lindbergii 7: +; Sphagnum centrale (LC-att) 9: 1; Sphagnum magellanicum 13: 3; Sphagnum obtusum (LR-nt) 14: 4; Calypogeia azurea 15: +; Pseudocampylium radicale (LC-att) 31: +; Polytrichum longisetum 31: +; Brachythecium rutabulum 31: r; Amblystegium serpens 33: +; Dichodontium palustre (LC-att) 34: +; Pohlia nutans 34: +; Pohlia drummondii 49: +; Plagiothecium denticulatum 54: +. Localities of relevés (BHM = Bohemian-Moravian Highlands, TR = Třeboň basin): 1. Eastern Bohemia, Opatov, 0.5 km S of the Nový rybník pond, 49°49'39.1", 16°29'18.9". 2. BMH, Hluboká, Řeka Nature Reserve, 0.5 km NNW of the village, 49°39'59.8", 15°51'10.7". 3. BMH, Bory-Dolní Bory, 0.3 km NW of Horník pond, 49°25'52.6", 16°01'24.6". 4. BMH, Černíč, 1.2. km NW of village, 49°08'15.6", 15°27'09.2". 5. Eastern Bohemia, Rudoltice v Čechách, 3.5 km NW of train station, 49°54'52.2", 16°32'10.1". 6. BMH, Sobíňov, Zlatá louka Nature Reserve, 2 km N of village, 49°42'49.6", 15°46'23.0". 7. BMH, Věcov-Odranec, S margin of village, 49°36'41.3", 16°08'23.1". 8. BMH, Hluboká, Řeka Nature Reserve, 0.5 km NNW of village, 49°39'58.1", 15°51'10.7". 9. BMH, Milíčov, N of village, 49°24'11.3", 15°23'43.1". 10. BMH, Dušejov 1 km W of village, 49°24'26.2", 15°25'09.1". 11. BMH, Šimanov, S of village, 49°27'00.6", 15°26'49.3". 12. BMH, Nový Rychnov-Čejkov, 1 km N of village, 49°23'06.8", 15°19'46.5". 13. BMH, Švábov, 0.5 km WNW of train station, 49°18'58.8", 15°20'55.1". 14. BMH, Jihlávka, 1 km S of village, 49°15'00.7", 15°17'48.6". 15. TR, Borovany-Hluboká 348 Preslia 86: 337–366, 2014 u Borovan, 1,5 km SE of the village, 48°53'30.6", 14°41'16.5". 16. TR, Libín-Spolí, in the valley of the Spolský potok stream, N of village, 48°59'09.6", 14°42'32.1". 17. TR, Kunžak-Suchdol, N of village, 49°07'54.7", 15°14'14.4". 18. BMH, Jihlávka, 1,2 km SE of village, 49°14'51.6", 15°16'44.4". 19. BMH, Žďár nad Sázavou- Plíčky, 49°33'57.2", 15°58'27.6". 20. BMH, Žďár nad Sázavou, N of town, 49°35'07.8", 15°56'32.4". 21. BMH, Trhová Kamenice, Buchtovka Nature Reserve, S of village, 49°46'26.6", 15°48'38.9". 22. BMH, Hlinsko, Ratajské rybníky Nature Reserve, SE of town, 49°46'06.1", 15°55'58.3". 23. BMH, Pustá Rybná, Damašek Nature Reserve, 1.5 NW of village, 49°43'08.8", 16°07'29.7". 24. BMH, Borová, 0.5 km W of train station, 49°44'34.3", 16°09'10.0". 25. BMH, Korouhev, 1.5 km SE of village, 49°38'45.7", 16°16'31.4". 26. BMH, Kameničky, Louky v Jeníkově Nature Reserve, 49°44'19.0", 15°57'51.0". 27. BMH, Sobíňov, Zlatá louka Nature Reserve, 2 km N of village, 49°42'48.4", 15°46'21.2". 28. BMH, Vortová, Zlámanec Nature Reserve, 49°42'18.9", 15°55'55.7". 29. BMH, Věcov-Odranec, 1 km S of village, 49°36'37.6", 16°08'12.1". 30. BHM, Hojkov, 1 km S of village, 49°22'56.8", 15°24'50.6". 31. BMH, Jihlávka, 1,2 km SE of village, 49°15'06.7", 15°17'55.4". 32. TR, Ratiboř, 1.5 km E of village, 49°09'06.0", 14°55'53.4". 33. TR, Bošilec, 49°09'04.3", 14°41'27.9". 34. TR, Odměny, near Svět pond, 48°59'31.3", 14°43'33.5". 35. TR, Chlum u Třeboně 48°58'44.3", 14°53'49.4". 36. BMH, Trhová Kamenice, Buchtovka Nature Reserve, 49°46'24.5", 15°48'44.3". 37. BMH, Vortová, Návesník Nature Reserve, 49°42'41.9", 15°55'36.5". 38. BMH, Kameničky, Bahna Nature Reserve, 49°45'11.3", 15°59'28.3". 39. BMH, Hlinsko, Ratajské rybníky Nature Reserve, SE of town, 49°46'09.8", 15°56'01.4". 40. BMH, Pustá Rybná, Damašek Nature Reserve, 1.5 NW of village, 49°43'08.6", 16°07'36.0". 41. BMH, Borová, 2 km NW of village, 49°45'05.0", 16°08'26.0". 42. BMH, Borová, 0.5 km W of train station, 49°44'32.8", 16°09'07.4". 43. BMH, Kameničky-Filipov, S margin of village, 49°44'35.5", 15°59'18.5". 44. BHM, Svratouch, 1 km NE of village, 49°44'12.7", 16°02'44.0". 45. BMH, Radostín, Radostínské rašeliniště Nature reserve, 49°39'25.7", 15°53'18.4". 46. TR, Borovany, SE of Žemlička pond, 48°53'27.7", 14°41'23.7". 47. TR, Lišov- Dolní Slověnice, 2 km NW of village, 49°04'11.4", 14°38'55.7". 48. TR, Ponědrážka, 1.5 km WNW of village, 49°08'09.1", 14°40'46.4". 49. TR, Ponědrážka, 1 km NWN of village, 49°08'33.7", 14°41'39.3". 50. TR, Hamr, SW of Kukla pond, 48°57'19.2", 14°53'21.0". 51. TR, Hamr, 2 km NW of village, 48°57'43.0", 14°53'19.2". 52. TR, Třeboň, 49°02'21.7", 14°50'12.1". 53. BMH, Polnička, Pod Kamenným vrchem Nature Reserve, 49°36'59.4", 15°53'53.3". 54. BMH, Borová, 2 km NW of village, 49°45'03.6", 16°08'13.4". 55. BMH, Radostín, Dářko Nature reserve, 2 km S of village, 49°38'14.4", 15°52'10.3". 56. BHM, Pustá Kamenice, S of village, 49°44'56.5", 16°05'31.4". 57. BMH, Vortová, Malý Černý pond, 49°42'10.6", 15°54'44.9".

1. Caricion davallianae (extremely rich fens) Absence of Sphagnum species and presence of low, calcium-demanding graminoids (Blysmus compressus, Triglochin palustris) differentiates this alliance from the others. The herb layer is further composed of sedges such as C. davalliana, C. panicea, C. rostrata and calcicole herbaceous plants (Eupatorium cannabinum, Valeriana dioica). The moss layer is usually dominated by Tomentypnum nitens or Scorpidium cossonii, accompained by Bryum pseudotriquetrum, Campylium stellatum or Palustriella commutata. Well-preserved stands of Caricion davallianae were recorded very rarely within the study area. All the sites studied occur at localities with stable water regimes and are regularly mown.

2. Sphagno warnstorfii-Tomentypnion (rich fens) This community is characterized by presence, and often also strong dominance, of calcium-tolerant species of Sphagnum (Sphagnum contortum, S. teres, S. warnstorfii and, occasionally, S. subnitens). The moss layer is further enriched by so called “brown mosses”, i.e. non-sphagnaceous weft-forming bryophytes (Campylium stellatum, Hamato- caulis vernicosus, Scorpidium cossonii) and bryophytes with boreal distributions considered to be glacial relicts in central Europe (Rybníček 1966), e.g. Paludella squarrosa, Tomentypnum nitens and Breidleria pratensis. The herb layer is mostly made up of low sedges (Carex demissa, C. nigra, C. panicea, C. pulicaris), accompained by other Peterka et al.: Differentiation of rich fens 349

Cyperaceae (Eriophorum angustifolium, E. latifolium, Trichophorum alpinum). Tomen- typnum nitens and Sphagnum warnstorfii often form small hummocks, on which species preferring drier (i.e. oxic) conditions can grow (Anthoxanthum odoratum, Festuca filiformis, Luzula multiflora). Both the bryophyte and herb layers are usually species-rich and host a large number of rare or endangered species (according to Grulich 2012, Kučera et al. 2012), e.g. Calliergon giganteum, Carex dioica, C. hostiana, C. pulicaris, Dactylorhiza majalis, Drosera rotundifolia, Hamatocaulis vernicosus, Paludella squarrosa, Parnassia palustris and Trichophorum alpinum. The vegetation is restricted to protected and annu- ally mown fens and fen meadows.

3. Caricion canescentis-nigrae (= Caricion fuscae; moderately rich fens) This community has a relatively low number of diagnostic species and is frequently dominated by Carex nigra, Eriophorum angustifolium and Comarum palustre. The moss layer comprises mostly Sphagnum teres, but other species of moss can also prevail (e.g. Calliergonella cuspidata, Sphagnum subsecundum). Both the herb and bryophyte layers are medium species-rich and almost lack calcicole species of plants. In some cases, the moderately rich fens in the study area lack sharp boundaries with Calthion palustris meadows (namely the Angelico sylvestris-Cirsietum palustris association) and poor fens. These transitional stands are indicated by the occurrence of broad-leaved herbaceous plants (Angelica sylvestris, Bistorta officinalis, Caltha palustris, Lychnis flos-cuculi, Ranunculus auricomus agg. or Sanguisorba officinalis) and/or an enhanced cover of Sphagnum flexuosum.

4. Sphagno-Caricion canescentis (poor fens) This, the last alliance represents species-poor minerotrophic fens without calcium-tolerant mosses and vascular plants. Frequent dominants of the moss layer are Sphagnum sect. Cuspidata (S. fallax, S. flexuosum), Sphagnum sect. Palustria (S. palustre, S. papillosum) and Polytrichum commune. Other non-sphagnaceous mosses are rarely present, with the exception of Straminergon stramineum. The herb layer mostly consists of Cyperaceae (Carex nigra, C. rostrata, Eriophorum angustifolium) and shrubs (Calluna vulgaris, Vaccinium oxycoccos). Some mires in the Třeboň basin are characterized by a fine-scale mosaic of (i) poor fens and (ii) oligotrophic pools with rare macrophytes (e.g. Sparganium natans, Utricularia ochroleuca, U. intermedia) or strongly waterlogged microhabitats with Rhynchospora alba and Sphagnum auriculatum, whereas similar habitats in the Bohemian-Moravian Highlands are rather uniform. Mire vegetation of the Sphagno recurvi-Caricion canescentis is widespread on the Bohemian Massif and occurs in wet meadows, at the margins of fishponds, in bog laggs or treeless patches in coniferous forests.

Ecological differences between the different types of fens The differences in the environmental variables in the four vegetation types are shown in Fig. 2. One-way ANOVA or the Kruskal-Wallis test confirmed the hypothesis that the different types of fens are well-characterized by water chemistry, especially pH, conductivity and calcium content of the groundwater. All groups differed significantly (F = 95.61, P < 0.00001) in pH, with the highest values recorded in the extremely rich fens, lower values in rich fens and moderately rich fens and the lowest values in poor fens. Similar results 350 Preslia 86: 337–366, 2014

8 700 a b c d a a b b

600

7 ]

-1 500

.cm

6 ěS 400

pH 300 5

200

conductivity [ 4 100

3 0 1234 1234

120 80 a b bc c a a a b

100 60

] ] 80

-1

[mg.l [mg.l 60 40

Ca 40 20 20

0 0 1234 1234

600 25 ab b ab a n.s.

500 20

400

]

] -1 15

-1

g.l

ě 300

[

[mg.l

4 + 10

PO 200

5 100

0 0 1234 1234

1000 7000 n.s. n.s. 6000 800 5000

]

-1 ] 600

-1

g.l 4000

g.l

[

4 3000 400

NH 2000 200 1000

0 0 1234 1234 Peterka et al.: Differentiation of rich fens 351

200 60 n.s. ab bb

50 150 40

]

-1

100 30

WDT [cm] Fe [mg.l 20 50 10

0 0 1234 1234

60 n.s. 50

N:P (ratio in bryophyte biomass)

0 1 (n = 18) 2 (n = 79) 3 (n = 38) 4 (n = 35)

most frequent species sampled within vegetation type: Campylium Sphagnum Sphagnum Sphagnum stellatum warnstorfii teres fallax Fig. 2. – Comparison of the environmental variables and the N:P ratio in bryophyte biomass measured in the four vegetation types: 1 – Caricion davallianae (extremely rich fens), 2 – Sphagno warnstorfii-Tomentypnion (rich fens), 3 – Caricion canescenti-nigrae (moderately rich fens), 4 – Sphagno-Caricion canescentis (poor fens). Medians are indicated by horizontal lines. Significant differences between groups (P > 0.05, the post-hoc test) are indicated by different letters, n.s. = no significant differences. were also recorded for conductivity (F = 58.30, P < 0.00001), calcium (F = 14.02, P < 0.00001) and magnesium (KW-H = 24.05, P = 0.00002), but these chemical variables did not differ between all pairs of vegetation types. Water in rich fens contained significantly more phosphorus (F = 10.74, P = 0.0132) than that in poor fens. In contrast, no significant differences were detected in the N:P ratio in the bryophyte biomass. By analogy, concen- + - + trations of NH4 ,NO3 ,K in water samples were similar in all vegetation types. Iron concentration increased from rich to poor fens (Fig. 2), but the differences between vegetation types were not statistically significant. Extremely rich fens are characterized by significantly lower water table than other types of fens (KW-H = 10.51, P = 0.0147).

Multivariate analyses PCA of environmental variables indicated two major gradients, one connected with pH, conductivity, calcium and magnesium concentrations and one with nutrient availability (ammonium, nitrate, potassium). Water table depth is greater in both, nutrient-rich fens and acidic fens. The different types of vegetation were particularly well-separated along the first axis, with the exception of those rich and moderately rich fens that are enriched in nutrients (Fig. 3). 352 Preslia 86: 337–366, 2014

NH4

NO3 K WTD

PO4 Fe

Mg Ca cond pH -1.0 1.0 -1.0 1.0 Fig. 3. – PCA ordination of samples based only on environmental variables. Eigenvalues of the first two axes are 0.376 and 0.180. Plots of different vegetation types are indicated by different symbols: Caricion davallianae (extremely rich fens), n Sphagno warnstorfii-Tomentypnion (rich fens), × Caricion canescenti-nigrae (moderately rich fens), [] Sphagno-Caricion canescentis (poor fens).

A simple DCA ordination diagram based on both vascular plants and bryophytes (Fig. 2) indicates that each group (alliance) is clearly separated along the main vegetation gradient (first DCA axis) stretching from extremely rich fens (with Carex davalliana, Tomentypnum nitens or Scorpidium cossonii) to poor fens (with Polytrichum commune or Sphagnum fallax). The second DCA axis is of minor importance, with less than half the eigenvalue (Fig. 4), and can be interpreted as fen-to-meadow gradient, largely coinciding with the water level gradient) stretching from waterlogged sites with strictly wetland species (e.g. Carex diandra, Sphagnum contortum, S. subsecundum) to plots with broad-leaved herbaceous plants of rather mesic conditions (e.g. Ranunculus auricomus agg., Sanguisorba officinalis). Water pH significantly decreased along the main vegetation gradient towards poor fens (Fig. 4). A similar result was also recorded for conductivity and both calcium and magnesium concentrations (Table 2, scatters not shown). Concentrations of potassium and nitrates correlated with the second axis. Concentration of total iron in water slightly increased towards the “poor” end of the first axis and towards the “wet” end of the second axis.

Fig. 4. – DCA ordination of all the plots sampled using pooled data on species compositions of both vascular plants and bryophytes. Position of relevés and species along two first ordination axes are shown. The eigenvalues of the axes are 0.475 (12.4% of total inertia) and 0.193 (5.0%). Only the species with a weight above 10% are shown (for full names see Electronic Appendix 3). Plots of different vegetation types are indicated by different symbols: Caricion davallianae (extremely rich fens), n Sphagno warnstorfii-Tomentypnion (rich fens), × Caricion canescenti-nigrae (moderately rich fens), [] Sphagno-Caricion canescentis (poor fens). Isolines of selected environmental variables and species richness along two main vegetation gradients were created using generalized additive models (GAMs). ¤ Peterka et al.: Differentiation of rich fens 353

SanOff SphFle

JunEff RanAur CrePal CarEch EquPal AntOdo RumAce PolCom CarCan RanAcr BrePra FesRub CliDen CirPal CarNig SphPal DacMaj CarPan LuzMul PotEre AgrCan SphFal ValDio LycFlo SphTer BriMed VioPal EriAng PlaAff GalUli SphWar FesFil HolLan EpiPal StrStr TomNit CarPul SucPra CarPra AulPal CarDav CarDemCalCus GalPal EquFlu JunBul ScoCos BryPse FilUlm MenTri LysVul CamSte ComPal PolStr TriAlp PhrAus MolCae SphCon AnePin DroRot VacOxy CarRos PeuPal CarDia

SphSub -2 4 -1 5

species richness 25 40 20 15

45 35 30

10 40 0.0 2.5 0.0 2.5 0404

pH log_NO3 5.2 4.6 4.4 5.4 4.2 6.5 5.6 4 5.8

6 6 6.2 5 6.4 5 6.6 4.8 5.5

4.5 6.8 4 3 3.5 2.5 0.0 2.5 0.0 2.5 0404

lok_K 2.2 log_Fe 3 2.8 3.2 3.4 3.6 1.6 1.8 2.8 2.6 3.8 2.4 2.2 2 1.6 2 1.8 1.4 2.8 1 0.8 3.2 1.4 0.4 0.6 1.2 3.4 3.6 0.0 2.5 0.0 2.5 0404 354 Preslia 86: 337–366, 2014

SanOff RanAur CarCan ComPal

CarNig JunEff CarEch EquFlu AgrCan RumAce GalPal VioPal EriAng PeuPal FesRub LuzMul CarPra JunBul EquPal CirPal CrePal AntOdo LycFlo HolLan ValDio GalUli DacMaj CarDem EpiPal LysVul VacOxy MenTri CarPan RanAcr FesFil PotEre DroRot BriMed AngSyl CarDia FilUlm SucPra CarPul TriAlp MolCae JunArt ParPal PhrAus CarRos CarDav -1 4 -1 5

species_richness pH 22 12 5.5 4 28 10 30 32 6 18

4.5 6.5 5 26 24 16 14 7 0.0 2.5 0.0 2.5 0.0 3.5 0.0 3.5

log_K log_Mg 1 0.8 2 -0.6 1.81.61.4 0.8 -0.8 1 1.2 1.6 1.4 1.8

0.8 1.2

-0.4

0.4 0.6 -0 -0.2 0.2 0.2 0.80.60.4 1.41.2 1 0.0 2.5 0.0 2.5 0.0 3.5 0.0 3.5 Fig. 5. – DCA ordination of all the plots sampled using only the data on species compositions of vascular plants. Positions of the species along the first two ordination axes are shown. The eigenvalues of the axes are 0.328 (10.0% of total inertia) and 0.208 (6.3%). Only species with a weight above 10% are shown (for full names see Electronic Appendix 3). Isolines of selected environmental variables and species richness along the two main vegetation gradients were created using generalized additive models (GAMs). Peterka et al.: Differentiation of rich fens 355

Table 2. – The relationships between the two principal DCA ordination axes and environmental variables modelled and tested using generalized additive models. Two significance levels are presented. The unadjusted P-values lower than 0.05 are presented, and those which are significant after the Holm correction (P < 0.00208) are indicated by *. The last column describes the axis with which the tested variable coincided. Species data Variable Deviance DF F P Fitting to axes Vascular plants + pH 9.2538 4 84.1 < 0.00001* 1st axis (Fig. 4) bryophytes conductivity (log) 11.644 6 24.3 < 0.00001* 1st axis Ca2+ (log) 22.516 6 13.5 < 0.00001* 1st axis Mg2+ (log) 24.375 7 10.2 < 0.00001* 1st axis, right-skewed 3- PO4 (log) 12.413 5 4.7 0.00255 1st axis K+ (log) 47.944 4 6.1 0.00124* 2nd axis (Fig. 4) + NH4 (log) 38.219 5 3.9 0.00770 2nd axis - NO3 (log) 148.35 5 5.3 0.00109* 2nd axis (Fig. 4) total Fe (log) 72.75 7 5.1 0.00038* both, non-linearly (Fig. 4) WTD (log) not significant Vascular plants pH 13.466 3 71.3 < 0.00001* diagonally (Fig. 5) conductivity (log) 12.361 6 21.8 < 0.00001* diagonally Ca2+ (log) 11.119 3 5.1 0.00846 diagonally Mg2+ (log) 28.641 5 11.3 < 0.00001* 1st axis, skewed (Fig. 5) 3– PO4 (log) 12.637 6 3.3 0.01143 both, non-linearly K+ (log) 43.42 5 5.4. 0.00110* 1st axis, bimodally (Fig. 5) + NH4 (log) 38.733 5 2.7 0.03828 both, non-linearly – NO3 (log) 161.95 6 2.9 0.02241 both, non-linearly total Fe (log) 94.746 4 3.7 0.01509 both, non-linearly WTD (log) 19.053 4 3.6 0.02019 both, non-linearly Bryophytes pH 9.9109 5 55.8 < 0.00001* 1st axis (Fig. 6) conductivity (log) 9.6566 5 39.2 < 0.00001* 1st axis Ca2+ (log) 27.708 4 15.3 < 0.00001* 1st axis Mg2+ (log) 28.474 5 11.5 < 0.00001* 1st axis 3– PO4 (log) 14.041 3 5.3 0.00724 both, non-linearly K+ (log) not significant + NH4 (log) not significant – NO3 (log) not significant total Fe (log) 89.757 4 5.3 0.00280 1st axis WTD (log) 2900.7 7 2.9 0.01753 1st axis

In the DCA of only vascular plant data, the main gradient was dominated by increasing species richness, governed by the representation of grassland species, and coincided with pH only partially. The pH gradient stretches diagonally as a resultant of both the first and the second axis (Fig. 5). Potassium concentration shows a bimodal relationship with the first axis, with maxima at opposite ends of the main gradient: in species-rich fen grasslands and poor fens. The DCA of bryophyte data yielded a much simpler result, with the dominant main axis sorting the species from calcicolous brown mosses (Tomentypnum nitens, Scorpidium cossonii) to poor-fen species (Polytrichum commune, Sphagnum fallax), which were tightly linearly correlated with pH (Fig. 6). The forward selection in the CCA revealed the key role of water pH in the entire data set

(explained variance: 39.5%, F = 5.8, P = 0.002, Padjust = 0.018). The residual variance was partially explained by nitrate concentration with marginal significance (expl. var.: 10.3%,

F = 1.6, P = 0.008, Padjust = 0.064). The variation within the vascular plant subset was 356 Preslia 86: 337–366, 2014

PolCom SphFal SphSub PolStr StrStr SphFle CalCus CliDen AulPal ScoCos BrePra SphTer TomNit PlaAff SphCon BryPse CamSteAnePin SphWar

SphPal -1.0 2.5 -1 6

species_richness pH 4.5 65 5 8 7 9 4 10 5 5.5 3 6 4.5 6.5 10 7 6 9

12 11 0.0 3.0 0.0 3.0 0606

Fig. 6. – DCA ordination of all of the plots sampled using only the data on species composition of bryophytes. Positions of the species along the first two ordination axes are shown. The eigenvalues of the axes are 0.707 (14.9% of total inertia) and 0.345 (7.3%). Only species with a weight above 10% are shown (for full names see Electronic Appendix 3). Isolines of selected environmental variables and species richness along the two main vegetation gradients were created using generalized additive models (GAMs).

mostly determined by pH (expl. var.: 27.2%, F = 3.8, P = 0.002, Padjust = 0.02), while nitrate concentration appeared to be the second most important factor (expl. var.: 12.0%, F = 1.7,

P = 0.002, Padjust = 0.02). In the case of bryophytes, pH explained 47.7% of variance (F = 7.1, P = 0.002, Padjust = 0.018) and no other variable was a significant predictor of species composition.

Discussion The poor-rich gradient within fens on the Bohemian Massif The floristic composition of mires in the south-eastern part of the Bohemian Massif is associated with differences in pH and concentration of dissolved base cations. This result is not surprising and matches the results of other studies throughout the world (e.g. Malmer 1986, Gerdol 1995, Řkland et al. 2001, Hájek et al. 2002), including on parts of the Bohemian Massif (Navrátilová et al. 2006, Laburdová & Hájek 2014). Contrary to the fens in the Carpathian part of the Czech Republic (Hájek et al. 2002), water pH appeared to be more tightly correlated with the main vegetation gradient than calcium concentration. This difference can be explained by the poor-rich gradient in the study area being incomplete due to the absence of calcareous fens and rare occurrence of extremely rich Peterka et al.: Differentiation of rich fens 357 fens. This incompleteness is first of all caused by the prevalence of carbonate-poor rocks and lack of calcareous tufas in both the Bohemian-Moravian Highlands and Třeboň basin (Kovanda 1971). The second reason is the deterioration of the fens due to drainage, fertilization and abandonment and consequent successional changes in these communities. For example, brown-moss fens with boreal sedges (classified as Drepanoclado revolventis- Caricetum lasiocarpae and Scorpidio-Caricetum limosae in Rybníček et al. 1984) have not been recorded recently in the study area. In addition, this result indicates that water pH is a good proxy of the complex pH/calcium gradient and is similar to the results from other crystalline regions in Europe, such as Fennoscandia (Tahvanainen 2004) and alpine zones of high European mountains (Hájková et al. 2006, Sekulová et al. 2013). In contrast to previously explored regions (Boyer & Wheeler 1989, Boeye et al. 1997, Rozbrojová & Hájek 2008, Kooijman & Hedenäs 2009), the most calcium-rich habitats (Caricion davallianae) were not generally determined by low phosphorus availability. The data from the entire Bohemian Massif did not confirm that the availability of phosphorus in the Sphagno warnstorfii-Tomentypnion fens is lower than in poor fens as indicated previously by data from the Carpathians (Hájek et al. 2002) and the Třeboň basin (Navrátilová et al. 2006). Concentration of phosphates in water increased towards high- pH fens in our study area and the N:P ratio in bryophyte biomass indicated a similar level of phosphorus limitation in all vegetation types. Such an important difference from what is recorded in other regions is probably caused by generally rather high concentrations of dissolved iron in the study area, which makes phosphorus unavailable to plants (Zak et al. 2004, Cusell et al. 2013). The lack of coincidence between dissolved phosphorus or nitrogen in the soil water and differentiation of vegetation types along the poor-rich gradient is also evident from the results of PCA of environmental factors, where vegetation types were differentiated along the first (pH/calcium) axis but not along the second, the nutrient- availability axis. Moreover, nutrient-enriched fens with a high score along the second axis were recorded for all vegetation types other than extremely rich fens. If dissolved nutrients are not associated with the poor-rich gradient on the Bohemian Massif, what factors are responsible for vegetation differentiation along the pH/calcium gradient? In drier habitats, iron unavailability in calcareous soils and high concentration of toxic aluminium in acid soils (pH < 4.5) are considered to be a major causal explanation of the calcicole-calcifuge behaviour of species (Zohlen & Tyler 2000, Tyler 2003). On the Bohemian Massif, low pH that enables the mobilization of aluminium occurs only in poor fens, but also other vegetation types were mutually well-differentiated with respect to pH. Iron concentration increased towards poor fens (see also Rozbrojová & Hájek 2008), but differences among different vegetation types were not statistically significant. Moreover, there are very high concentrations of dissolved iron (10–200 mg·l–1) throughout the area studied, suggesting iron toxicity (Snowden & Wheeler 1993, Aggenbach et al. 2013) affecting all vegetation types. Hence, iron alone cannot explain the species turnover along the pH/calcium gradient. If this is the case, what is the ecological explanation of the poor-rich gradient? We suggest that interactions between pH, calcium concentration in the water and water level affect the poor-rich gradient in a complicated way. All these factors determine species composition of vegetation, especially of its moss layer, and mosses are generally recognized to be crucial ecosystem engineers of mires (Jones et al. 1994, Vitt 2000). Clymo (1973) reports a negative up to a lethal effect of rich-fen water (having high pH and a high 358 Preslia 86: 337–366, 2014 calcium concentration) on most of the species of Sphagnum he studied. Granath et al. (2010) states that inundation of capitula by rich-fen water is lethal for the bog species Sphagnum fuscum, but does not affect the calcium-tolerant species S. teres. Apart from S. fuscum, several other species of Sphagnum avoid calcium by forming hummocks (Brehm 1971, Hájek et al. 2014). We conclude that sphagna are generally intolerant of an elevated water table in calcium-enriched fens. Thus, the combination of pH with calcium and water level determines whether a fen will be dominated by either sphagna or brown mosses, or both. Sphagna and brown mosses may affect ecosystem processes differently. Sphagna acidify the environment (Kooijman 2012) and drive the succession towards poor fens (Paulissen et al. 2013), take up most nutrients (Malmer et al. 1994, Fritz et al. 2014), hamper seed germination or seedling establishment (Neuhäusl 1975, Soudzilovskaia et al. 2011) and decrease decomposability of organic matter and hence nutrient mineralization (Hájek et al. 2011a). Great competitive ability of Sphagnum species can result in competitive exclusion of some vascular plants and a decrease in species richness (Hájková & Hájek 2003, Malmer et al. 2003, van der Welle et al. 2003). On the other hand, hummock- forming calcium-tolerant sphagna may provide a specific niche for shallow-rooting vascular plants that may avoid iron toxicity and reducing conditions by growing in aerated but permanently wet Sphagnum hummocks. Ecosystem role of brown mosses is less well known, but some studies indicate they have a specific role in nutrient cycling and uptake by plants by affecting redox conditions (Crowley & Bedford 2011). In conclusion, we suggest that pH differences in fens control the occurrence of particular species of moss, which may act as ecosystem engineers and regulate the vegetation structure to which phane- rogamic species respond. Some short-lived vascular plants tightly associated with calcareous fens may not be calcium-demanding, but just cannot reproduce generatively in dense Sphagnum carpets.

Other gradients The poor-rich gradient is commonly identified as the main vegetation gradient in fens, but not always. Floristic and faunistic composition of Polish lowland fens, is, for example, more affected by factors connected with hydrology and phosphorus availability (Pawli- kowski et al. 2013, Schenková et al. 2014). We expected an increasing role of hydrology and nutrient availability in our data set, which contains floristically unique topogenic fens and fens eutrophicated by polluted water (Navrátilová et al. 2006), and fens naturally enriched by potassium from weathering feldspars on granite bedrock. In comparison with the data reported for fens in the literature (Sjörs 1948, Malmer 1962, Persson 1962, Mörnsjö 1969, Elveland 1976, Zoltai & Vitt 1995, Wind-Mulder et al. 1996, Hájek et al. 2002, Hedenäs & Kooijman 2004, Tahvanainen 2004, Pawlikowski et al. 2013), groundwater in the study area contains generally more potassium and iron. Similar concentrations of potassium (up to 20 mg·l–1, with a mean value of about 5 mg·l–1) are reported only by Gąbka & Lamentowicz (2008) for poor fens in western Poland. Also phosphate concentration in groundwater in the study area is substantially higher than in Scandinavia (Mörnsjö 1969, Hedenäs & Kooijman 2004), slightly higher than that recorded in the Outer Western Carpathians (Hájek et al. 2002), much higher than in the Inner Western Carpathians (Hájek et al. 2014) but similar to the concentration in north-eastern Poland (Pawlikovski et al. 2013). Despite these differences, the gradient structure fits the general Peterka et al.: Differentiation of rich fens 359 pattern found across temperate Europe, i.e. primary gradient of pH and calcium and secondary gradient of fertility (Gerdol 1995, Wheeler & Proctor 2000, Hrivnák et al. 2008). In contrast in the Western Carpathians where absolute concentrations of nutrients in water are less important than stoichiometry (compare Hájek et al. 2002, Hájek & Hekera 2004 and Rozbrojová & Hájek 2008), the fertility gradient on the Bohemian Massif coincided with the absolute concentrations of particular nutrients, especially potassium and nitrate. It is partially correlated with water table depth, because water table decline causes nutrients in peat to mineralise (Grootjans et al. 1986). Water table depth further correlates with pH, because acidic fens may develop from alkaline fens after a water table decline that isolates the fen surface from the effect of groundwater (Granath et al. 2010, Paulissen et al. 2013). The complex gradient of fertility and water table depth (the fen-to-meadow gradient) was, however, much more strongly pronounced in vascular plant data. The result of a more complex control of vascular plant distribution in fens, including nutrient availability, is in accordance with results from the Western Carpathians (Hájková & Hájek 2004), Canada (Vitt & Chee 1990), the Netherlands (van Baaren et al. 1988) and the Alps (Bragazza & Gerdol 2002, Miserere et al. 2003, Sekulová et al. 2013).

Implications for fen classification Four vegetation types distinguished in this study matched the classification of central- European minerotrophic mires proposed by Hájek et al. (2006), which follows the tradition of Scandinavian mire ecologists (Nordhagen 1943, Malmer 1986, Sjörs & Gunnarsson 2002). In other words, the main fen vegetation types (alliances) on the Bohemian Massif correspond to parts of the poor-rich gradient and clearly differ from each other in species composition and site conditions. Water conductivity, calcium concentration and most importantly pH seem to be the variables best reflecting the floristic delimitation of particular vegetation types. Bryophytes were found to play an important role in vegetation diver- sification, because they mainly reflect a single dominant gradient of water pH and calcium. Moreover, they play a crucial role in mire ecosystem functioning and via direct interactions with vascular plants they affect the overall species composition of fen vegetation. The Scandinavian classification system delimiting major types of fens according to base saturation and associated structure of the bryophyte layer, thus appeared to be more suitable for our study area than the German-Austrian system based on hydrological gradients and dominance of particular vascular plants such as Rhynchospora alba, Carex lasiocarpa, C. limosa, C. nigra or Menyanthes trifoliata (Koch 1926, Oberdorfer 1957, Dierssen 1982, Steiner 1992), but nevertheless it is applied to fens in the Austrian part of the Bohemian Massif (Zechmeister & Steiner 1995). We aimed initially to address the specific question of floristic and environmental delimitation of the Sphagno warnstorfii-Tomentypnion fens that are currently disappearing but are extremely important in terms of biodiversity conservation. We conclude that our results confirm the meaningfulness of distinguishing the Sphagno warnstorfii-Tomen- typnion alliance, which was clearly differentiated based on both its floristic composition and water chemistry in our study. It further formed a quite compact cluster in the DCA ordination diagram. The presence of habitat specialists and rare and endangered species is high, which conforms with results from the Western Carpathians and Bulgaria (Hájek et al. 2007). High species richness together with a high representation of habitat specialists 360 Preslia 86: 337–366, 2014 suggests continuity over longer periods of time in the study area (compare Hájek et al. 2007). Thus the Sphagno warnstorfii-Tomentypnion fens can be characterized as mineral- rich fens where either a slight decrease in the water table, or suitable pH and calcium levels, enable the co-occurrence of calcium-tolerant sphagna (Sphagnum warnstorfii, S. contortum, S. teres, S. subnitens) with boreal species of brown mosses. They are rich in habitat specialists, with a group of shallow-rooting boreal fen plants. In the boreal zone these fens are more widespread but poorer in grassland species and calcareous-fen specialists than in central Europe. Similar alliances occur in European Russia and Siberia (Smagin 1999, 2007, Lapshina 2010). The variation in the Sphagno warnstorfii-Tomentypnion alliance on a European scale thus deserves further research.

See www.preslia.cz for Electronic Appendices 1–3

Acknowledgements We would like to thank Jan Beťák and Daniel Dítě for their help with assembling the relevés and collecting water and biomass samples. Petr Bureš, Filip Lysák, Táňa Štechová, Petra Hájková and Jana Navrátilová recommended several localities and shared our enthusiasm for the fascinating ecosystem of rich fens. Ondřej Hájek created the map. Tomáš Hájek leads our join project on calcium-tolerant peat mosses and provided many useful insights into factors affecting the occurrences of species of mosses in different environments. Tony Dixon kindly improved our English. This research was funded by the Czech Science Foundation (grant number: P505/10/0638), institutional support of Masaryk University and long-term research development project of Institute of Botany, Czech Acad- emy of Science (RVO 67985939).

Souhrn Jihovýchodní část Českého masivu (Českomoravská vrchovina, Třeboňsko) je významným centrem slatinné vegetace a její biodiverzity. Ohrožené druhy rostlin a živočichů zde hostí zejména slatiniště svazu Sphagno warnstorfii-Tomentypnion, jejichž prostředí představuje specifický úsek gradientu pH a vápnitosti, který je nejvý- znamnějším gradientem uvářejícím druhové složení rašelinišť. Floristické a ekologické vymezení hlavních vege- tačních typů (svazů) podél tohoto gradientu, od chudých (přechodových) slatinišť po vápníkem bohatá slatiniště, bylo dosud testováno zejména na datech ze Západních Karpat a Bulharska. Tyto studie nelze jednoznačně extra- polovat na území Českého masivu, kde jsou častá topogenní rašeliniště a kde podzemní voda obsahuje celkově více draslíku, železa a fosforu než v jiných oblastech Evropy. Aktuální vegetační přehledy sousedních zemí sdíle- jících části Českého masivu (Rakousko, Německo, Polsko) vymezení hlavních typů rašelinné vegetace podle komplexního gradientu pH/vápnitosti nepřijímají a svaz Sphagno warnstorfii-Tomentypnion tedy nerozlišují. V této studii jsme shromáždili data o vegetaci a proměnných prostředí (chemismu vody a hloubce vodní hladiny) z 57 unikátních zachovalých slatinišť. Klasifikace získaných fytocenologických snímků pomocí algoritmu ISOPAM téměř beze zbytku odpovídala vymezení svazů v monografii Vegetace ČR. Jednotlivé vegetační typy byly téměř odděleny v analýze hlavních komponent, která zohledňovala jen data o prostředí. Všechny vegetační typy se vzá- jemně signifikantně lišily v pH vody, jehož hodnoty, stejně jako koncentrace vápníku ve vodě, korelovaly s hlav- ním vegetačním gradientem vyjádřeným první osou detrendované korespondenční analýzy. Podél druhé osy, představující sekundární vegetační gradient, se měnila koncentrace dusičnanů a fosforu. Ordinační analýzy uká- zaly poněkud odlišné výsledky, když byla společenstva mechorostrů a cévnatých rostlin analyzována odděleně. Analýza společenstev mechorostů nevytvořila sekundární gradient spojený s přístupností živin a analýza spole- čenstev cévnatých rostlin vytvořila primární gradient, který odrážel vzrůstající počet druhů, včetně generalistů, od chudých k velmi bohatým slatiništím a jen částečně koreloval s pH. Oproti našemu očekávání nebyla bohatá slatiniště svazu Sphagno warnstorfii-Tomentypnion, ani vápníkem bohatá slatiniště svazu Caricion davallianae, vymezena nízkou dostupností fosforu, jako tomu bylo v jiných studiích ze střední Evropy. Druhové složení nej- vápnitějších slatin tedy pravděpodobně určuje vysoké pH a velká koncentrace vápníku, vysoká hladina podzemní vody a možná i nízká koncentrace přístupného železa. Velká alkalinita vede spolu s trvalým zamokřením k absen- ci rašeliníků a umožňuje tak výskyt některých kompetičně slabých druhů cévnatých rostlin, které nejsou vždy a priori vápnomilné, ale nemohou se generativně množit v souvislých porostech rašeliníků. Naše data ukazují, že Peterka et al.: Differentiation of rich fens 361 vymezení hlavních vegetačních typů (svazů) rašelinné vegetace podél gradientu pH a vápnitosti má značný floris- tický i ekologický smysl také v hercynských pohořích a že výskyt jednotlivých vegetačních typů je předurčen zejména úrovní pH a koncentrací vápníku v prostředí. Uvedené faktory přímo ovlivňují výskyt jednotlivých funkčních skupin mechorostů, které pak rozhodujícím způsobem ovlivňují jak výskyt jednotlivých druhů cévna- tých rostlin, tak i fungování rašelinného ekosystému jako celku.

References Aggenbach C. J. S., Backx H., Emsens W. J., Grootjans A. P., Lamers L. P. M., Smolders A. J. P., Stuyfzand P. J., Wołejko L. & van Diggelen R. (2013): Do high iron concentrations in rewetted rich fens hamper restoration? – Preslia 85: 405–420. Albrecht J. (ed.) (2003): Chráněná území ČR 8. Třeboňsko [Protected areas of the Czech Republic, 8. Třeboň region]. – Agentura ochrany přírody a krajiny ČR, Praha & EkoCentrum Brno. Anderson D. S. & Davis R. B. (1997): The vegetation and its environments in Maine peatlands. – Can. J. Bot. 75: 1785–1805. Bedford B. L., Walbridge M. R. & Aldus A. (1999): Patterns in nutrient availability and plant diversity of temperate North American wetlands. – Ecology 8: 2151–2169 Berg C., Dengler J., Abdank A. & Isermann M. (eds) (2004): Die Pflanzengesellschaften Mecklenburg- Vorpommerns und ihre Gefährdung. – Wiessdorn-Verlag, Jena. Boeye D., Verhagen B., van Haesebroeck V. & Verheyen R. F. (1997): Nutrient limitation in species-rich lowland fens. – J. Veg. Sci. 8: 415–424. Bombonato L., Siffi C. & Gerdol R. (2010): Variations in the foliar nutrient content of mire plants, effects of growth-form based grouping and habitat. – Pl. Ecol. 211: 235–251. Boyer M. L. H. & Wheeler B. D. (1989): Vegetation patterns in spring-fed calcareous fens: calcite precipitation and constrains on fertility. – J. Ecol. 77: 597–609. Bragazza L. & Gerdol R. (1996): Response surfaces of plant species along water-table depth and water pH gradients in a poor mire on the Southern Alps. – Ann. Bot. Fenn. 33: 11–20. Bragazza L. & Gerdol R. (1999): Ecological gradients in some Sphagnum mires in the south-eastern Alps (Italy). – Appl. Veg. Sci. 2: 55–60. Bragazza L. & Gerdol R. (2002): Are nutrient availability and acidity-alkalinity gradients related in Sphagnum- dominated peatlands? – J. Veg. Sci. 13: 473–482. Bragazza L., Rydin H. & Gerdol R. (2005): Multiple gradients in mire vegetation: a comparison of a Swedish and an Italian bog. – Pl. Ecol. 177: 223–236. Brehm K. (1971): Ein Sphagnum-Bult als Beispiel einer naturlichen Ionenaustauschersaule. – Beitr. Biol. Pflanzen 47: 287–312. Čech L., Šumpich J. & Zabloudil V. (eds) (2002): Chráněná území ČR, 7. Jihlavsko [Protected areas of the Czech Republic, 7. Jihlavsko region]. – Agentura ochrany přírody a krajiny ČR, Praha & EkoCentrum Brno. Chytrý M., Tichý L., Holt J. & Botta-Dukát Z. (2002): Determination of diagnostic species with statistical fidelity measures. – J. Veg. Sci. 13: 79–90. Clymo R. S. (1973): The growth of Sphagnum. Some effects of environment. – J. Ecol. 61: 849–869. Coldea G., Sanda V., Popescu A. & Ştefan N. (1997): Les associations végétales de Roumanie. 1. Les associations herbacées naturelles [Plant association of Roumania. 1. Forb communities]. – Presses Universitaires, Cluj- Napoca. Conradi T. & Friedmann A. (2013): Plant communities and environmental gradients in mires of the Ammergauer Alps (Bavaria, Germany). – Tuexenia 33: 133–163. Crowley K. F. & Bedford B. L. (2011): Mosses influence phosphorus cycling in rich fens by driving redox conditions in shallow soils. – Oecologia 167: 253–264. Cusell C., Lamers L. P. M., van Wirdum G. & Kooijman A. (2013): Impacts of water level fluctuation on mesotrophic rich fens: acidification vs. eutrophication. – J. Appl. Ecol. 50: 998–1009. Dahl E. (1956): Rondane. Mountain vegetation in south Norway and its relation to the environment. – Skr. Norske Vidensk.-Akad. Oslo, Mat.-Naturvidensk. Kl. 3: 1–374. Danihelka J., Chrtek J. & Kaplan Z. (2012): Checklist of vascular plants of the Czech Republic. – Preslia 84: 647–811. de Molenaar J. G. (1976): Vegetation of the Angmagssalik District, Southeast Greenland. II. Herb and snow-bed vegetation. – Meddel. Grřnland 198/2: 1–266. Dierssen K. (1982): Die wichtigsten Pflanzengesellschaften der Moore NW-Europas. – Conservatoire et Jardin botaniques Genéve, Genf. 362 Preslia 86: 337–366, 2014

Dítě D., Hájek M. & Hájková P. (2007): Formal definitions of Slovakian mire plant associations and their application in regional research. – Biologia 62: 400–408. Divíšek J., Chytrý M., Grulich V.& Poláková L. (2014): Landscape classification of the Czech Republic based on the distribution of natural habitats. – Preslia 86: 209–231. du Rietz G. E. (1949): Huvudenheter och huvudgränser i svensk myrvegetation [Main units and main limits in Swedish mire vegetation]. – Svensk Bot. Tidskr. 43: 279–304. Elveland J. (1976): Myrar pĺ Storön vid norrbottenskusten [Coastal mires on the Storön peninsula, Norrbotten, N Sweden]. – Wahlenbergia 3: 1–274. Eurola S. (1962): Über die regionale Einteilung de südfinnischen Moore. – Ann. Bot. Soc. Zool. Bot. Fenn. Vanamo 33: 1–243. Felbaba-Klushina L. (2010): Prodromus syntaksonib roslynnosti bolit i cholodnych džerel Ukrains’kych Karpat (Klasy Scheuchzerio-Caricetea fuscae Tx. 1937, Oxycocco-Sphagnetea Br.-Bl. et Tx. ex Westhoff et al. 1946, Montio-Cardaminetea Br.-Bl. et Tx. ex Klika et Hadač 1944) [Prodromus of bogs, fens and cold spring syntaxa of the Ukrainian Carpathians]. – Nauk. Visn. Užhor. Un-tu. Ser. Biol. 28: 73–82. Fransson S. (1972): Myrvegetation i sydvästra Värmland [Mire vegetation in south-westem Värmland, Sweden]. – Acta Phytogeogr. Suec. 57: 1–133. Fritz C., Lamers L. P. M., Riaz M., van den Berg L. J. L. & Elzenga T. J. T. M. (2014): Sphagnum mosses: masters of efficient N-uptake while avoiding intoxication. – PLoS ONE 9: e79991. Gąbka M. & Lamentowicz M. (2008): Vegetation-environment relationships in peatlands dominated by Sphag- num fallax in western Poland. – Folia Geobot. 43: 413–429. Gerdol R. (1995): Community and species-performance patterns along an alpine poor-rich mire gradient. – J. Veg. Sci. 6: 175–182. Gerdol R. & Tomaselli M. (1997): Vegetation of wetlands in the Dolomites. – Diss. Bot. 281: 1–197. Gillet F. (1982): L’alliance du Sphagno-Tomenthypnion dans le Jura. – Doc. Phytosoc., N. S., 6: 155–180. Granath G., Strengbom J. & Rydin H. (2010): Rapid ecosystem shifts in peatlands: linking plant physiology and succession. – Ecology 91: 3047–3056. Grootjans A., Alserda A., Bekker C. W., Janáková M., Madaras M., Stanová V., Ripka J., van Delft B. & Wolejko L. (2005): Calcareous spring mires in Slovakia; jewels in the crown of the mire kingdom. – Stapfia 85: 97–116. Grootjans A. P., Schipper P. C. & van der Windt H. J. (1986): Influence of drainage on N-mineralization and vegetation response in wet meadows. 2. Cirsio-Molinietum stands. – Acta Oecol. 7: 3–14. Grulich V. (2012): Red List of vascular plants of the Czech Republic: 3rd edition. – Preslia 84: 631–645. Güsewell S. & Koerselman W. (2002): Variation in nitrogen and phosphorus concentrations of wetland plants. – Persp. Pl. Ecol. Evol. Syst. 5: 37–61. Hájek M. & Hájková P. (2011): Vegetace slatinišť, přechodových rašelinišť a vrchovištních šlenků (Scheuchzerio palustris-Caricetea nigrae) [Vegetation of fens, transitional mires and bog hollows]. – In: Chytrý M. (ed.), Vegetace České republiky 3. Vodní a mokřadní vegetace [Vegetation of the Czech Republic 3. Aquatic and wetland vegetation], p. 614–704, Academia, Praha. Hájek M., Hájková P. & Apostolova I. (2008): New plant associations from Bulgarian mires. – Phytologia Balcanica 14: 377–399. Hájek M. & Hekera P. (2004): Can seasonal variation in fen water chemistry influence the reliability of vegetation-environment analyses? – Preslia 76: 1–14. Hájek M., Hekera P. & Hájková P. (2002): Spring fen vegetation and water chemistry in the Western Carpathian flysch zone. – Folia Geobot. 37: 205–224. Hájek M., Horsák M., Hájková P. & Dítě D. (2006): Habitat diversity of central European fens in relation to environmental gradients and an effort to standardise fen terminology in ecological studies. – Persp. Pl. Ecol. Evol. Syst. 8: 97–114 Hájek M., Horsák M., Tichý L., Hájková P., Dítě D. & Jamrichová E. (2011b): Testing a relict distributional pattern of fen plant and terrestrial snail species at the Holocene scale: a null model approach. – J. Biogeogr. 38: 742–755. Hájek M., Plesková Z., Syrovátka V., Peterka T., Laburdová J., Kintrová K., Jiroušek M. & Hájek T. (2014): Pat- terns in moss element concentrations in fens across species, habitats, and regions. – Persp. Pl. Ecol. Evol. Syst. 16: 203–218. Hájek M., Tichý L., Schamp B. S., Zelený D., Roleček J., Hájková P., Apostolova I. & Dítě D. (2007): Testing the species pool hypothesis for mire vegetation: exploring the influence of pH specialists and habitat history. – Oikos 116: 1311–1322. Peterka et al.: Differentiation of rich fens 363

Hájek T., Ballance S., Limpens J., Zijlstra M. & Verhoeven J. T. A. (2011a): Cell-wall polysaccharides play an important role in decay resistance of Sphagnum and actively depressed decomposition in vitro. – Biogeochemistry 103: 45–57. Hájková P. & Hájek M. (2003): Species richness and above-ground biomass of poor and calcareous spring fens in the flysch West Carpathians, and their relationships to water and soil chemistry. – Preslia 75: 271–287. Hájková P. & Hájek M. (2004): Bryophyte and vascular plant responses to base-richness and water level gradients in Western Carpathian Sphagnum-Rich Mires. – Folia Geobot. 39: 335–351. Hájková P., Hájek M. & Apostolova I. (2006): Diversity of wetland vegetation in the Bulgarian high mountains, main gradients and context-dependence of the pH role. – Pl. Ecol. 184: 111–130. Hedenäs L. & Kooijman A. M. (2004): Habitat differentiation within Palustriella. – Lindbergia 29: 40–50. Heikkilä H. (1987): The vegetation and ecology of mesotrophic and eutrophic fens western in Finland. – Ann. Bot. Fenn. 24: 155–175. Hettenbergerová E., Hájek M., Zelený D., Jiroušková J. & Mikulášková E. (2013): Changes in species richness and species composition of vascular plants and bryophytes along a moisture gradient. – Preslia 85: 369–388. Hrivnák R., Hájek M., Blanár D., Kochjarová J. & Hájková P. (2008): Mire vegetation of the Muránska Planina Mts: formalised classification, ecology, main environmental gradient and influence of geographical position. – Biologia 63: 368–377. Jabłońska E., Pawlikowski P., Jarzombkowski F., Chormański J., Okruszko T. & Kłosowski S. (2011): Impor- tance of water level dynamics for vegetation patterns in a natural percolation mire (Rospuda fen, NE Poland). – Hydrobiologia 674: 105–117. Jermacâne S. & Laivin‚š M. (2001): Latvijâ aprakstîto augu sabiedrîbu sintaksonu saraksts [List of syntaxa described in Latvia]. – Latvijas Veěetâcija 4: 115–132. Jiménez-Alfaro B., Fernández Pascual E., Díaz González T. E., Pérez Haase A. & Ninot J. M. (2012): Diversity of fen vegetation and related plant specialists in mountain refugia of the Iberian Peninsula. – Folia Geobot. 47: 403–419. Jiroušek M., Poulíčková A., Kintrová K., Opravilová V., Hájková P., Rybníček K., Kočí M., Bergová K., Hnilica R., Mikulášková E., Králová Š. & Hájek M. (2013): Long-term and contemporary environmental conditions as determinants of the species composition of bog organisms. – Freshw. Biol. 58: 2196–2207. Jones C. G., Lawton J. H. & Shachak M. (1994): Organisms as ecosystem engineers. – Oikos 69: 373–386. Koch M. & Jurasinski G. (2014): Four decades of vegetation development in a percolation mire complex following intensive drainage and abandonment. – Plant Ecol. Divers. (in press, doi: 10.1080/17550874.2013.862752). Koch W. (1926): Die Vegetationseinheiten der Linthebene unter Berücksichtigung der Verhältnisse in der Nordostschweiz. Systematisch-kritische Studie. – Jahresber. St. Gallischen Naturwiss. Ges. 61/2: 1–144. Koczur A. & Nicia P. (2013): Spring fen Scheuchzerio-Caricetea nigrae in the Polish Western Carpathians: vegetation diversity in relation to soil and feeding waters. – Acta Soc. Bot. Pol. 82: 117–124. Kooijman A. (2012): ‘Poor rich fen mosses’: atmospheric N-deposition and P-eutrophication in base-rich fens. – Lindbergia 35: 42–52. Kooijman A. & Hedenäs L. (2009): Changes in nutrient availability from calcareous to acid wetland habitats with closely related brown moss species: increase instead of decrease in N and P. – Pl. Soil 324: 267–278. Koroleva N. E. (2001): Sintaksonomičeskij obzor bolot tundrovogo pojasa Chibinskich gor (Murmanskaja oblast) [Syntaxonomic survey of tundra belt mires of Khibiny mountains (Murmansk region)]. – Rastiteľnosť Rossii 2: 49–57. Koroleva N. E. (2006): Bezlesnye rastitelnye soobščestva poberežja Vostočnogo Murmana (Kol’skij poluostrov, Rossija) [Treeless plant communities of the East Murman shore (Kola peninsula, Russia)]. – Rastiteľnosť Rossii 9: 20–42. Kovanda J. (1971): Kvartérní vápence Československa [Quarternary limestones of Czechoslovakia]. – Sborn. Geol. Věd (Antropozoikum) A7: 1–236. Kučera J., Váňa J. & Hradílek Z. (2012): Bryophyte flora of the Czech Republic: updated checklist and Red List and a brief analysis. – Preslia 84: 813–850. Laburdová J. & Hájek M. (2014): Vztah vegetace pramenišť západočeské zřídelní oblasti k chemismu prostředí [Relationships between spring vegetation and environment chemistry in West-Bohemian mineral-spring region]. – Zpr. Čes. Bot. Společ. 49: 49–71. Lájer K. (1998): Bevezetés a magyarországi Lápok vegetáció-ökológiájába [Introduction to Hungarian wetland vegetation ecology]. – Tilia 6: 84–238. Lapshina E. (2010): Rastiteľnosť bolot jugo-vostoka Zapadnoj Sibiri [Mire vegetation of south-eastern part of Western Siberia]. – Izd. NGU, Novosibirsk. 364 Preslia 86: 337–366, 2014

Malmer N. (1962): Studies on mire vegetation in the Archean area of southwestern Götland (south Sweden). I. Vegetation and habitat conditions on the Ĺkhult mire. – Opera Bot. 7: 1–309. Malmer N. (1986): Vegetation gradients in relation to environmental conditions in northwestern European mires. – Can. J. Bot. 64: 375–383. Malmer N., Albinsson C., Svensson B. M. & Wallén B. (2003): Interferences between Sphagnum and vascular plants: effects on plant community structure and peat formation. – Oikos 100: 469–482. Malmer N., Horton D. G. & Vitt D. H. (1992): Element concentrations in mosses and surface waters of western Canadian mires relative to precipitation chemistry and hydrology. – Ecography 15: 114–128. Malmer N., Svensson B. M. & Wallén B. (1994): Interactions between Sphagnum mosses and field layer vascular plants in the development of peat-forming systems. – Folia Geobot. Phytotax. 29: 483–496. Matuszkiewicz W. (1982): Przewodnik do oznaczania zbiorowisk roślinnych Polski [Identification key to plant communities of Poland]. – Państwove Wydawnictwo Naukowe, Warszawa. Miserere L., Montacchini F. & Buffa G. (2003): Ecology of some mire and bog plant communities in the Western Italian Alps. – J. Limnol. 62: 88–96. Moen A., Lyngstad A. & Řien D. (2012): Boreal rich fen vegetation formerly used for haymaking. – Nord. J. Bot. 30: 226–240. Mörnsjö T. (1969): Studies on vegetation and development of a peatland in Scania, South Sweden. – Oper. Bot. 24: 1–187. Navrátilová J. & Navrátil J. (2005): Stanovištní nároky některých ohrožených a vzácných rostlin rašelinišť Třeboňska [Environmental factors of some endangered and rare plants in Třeboň region mires]. – Zpr. Čes. Bot. Společ. 40: 279–299. Navrátilová J., Navrátil J. & Hájek M. (2006): Relationships between environmental factors and vegetation in nutrient-enriched fens at fishpond margins. – Folia Geobot. 41: 353–376. Nekola J. C. (1999): Paleorefugia and neorefugia: the influence of colonization history on community pattern and process. – Ecology 80: 2459–2473. Nekola J. C. (2004): Vascular plant compositional gradients within and between Iowa fens. – J. Veg. Sci. 15: 771–780. Neuhäusl R. (1975): Hochmoore am Teich Velké Dářko. – Vegetace ČSSR A9, Academia, Praha. Nordhagen R. (1943): Sikilsdalen og Norges fjellbeiter. En plantesosiologisk monografi [Sikilsdalen and Norwe- gian mountain pastures: a plant sociological monograph]. – Bergens Mus. Skr. 22: 1–607. Oberdorfer E. (1957): Süddeutsche Pflanzengesellschaften. – Pflanzensoziologie 10: 1–564. Oberdorfer E. (ed.) (1998): Süddeutsche Pflanzengesellschaften. Teil I: Fels- und Mauergesellschaften, alpine Fluren, Wasser-, Verlandungs- und Moorgesellschaften. Ed. 4. – Gustav Fischer Verlag, Jena/Stuttgart/ Lübeck/Ulm. Řkland R. H., Řkland T. & Rydgren K. (2001): A Scandinavian perspective on ecological gradients in north-west European mires: reply to Wheeler and Proctor. – J. Ecol. 89: 481–486. Olde VenterinkH., Wassen M. J., VerkroostA. W. M. & de Ruiter P. C. (2003): Species richness-productivity patterns differ between N-, P-, and K-limited wetlands. – Ecology 84: 2191–2199. Paulissen M. P. C. P., Schaminée J. H. J., During H. J., Wamelink G. W. W. & Verhoeven J. T. A. (2013): Expan- sion of acidophytic late-successional bryophytes in Dutch fens between 1940 and 2000. – J. Veg. Sci. 25: 525–533. Paulissen M. P. C. P., van Der Ven P. J., Dees A. J. & Bobbink R. (2004): Differential effects of nitrate and ammonium on three fen bryophyte species in relation to pollutant nitrogen input. – New Phytol. 164: 451–458. Pawlikowski P., Abramczyk K., Szczepaniuk A. & Kozub Ł. (2013): Nitrogen: phosphorus ratio as the main ecological determinant of the differences in the species composition of brown-moss rich fens in north-eastern Poland. – Preslia 85: 349–367. Persson Ĺ. (1961): Mire and spring vegetation in an area north of lake Torneträsk, Torne Lappmark, Sweden. I. Description of the vegetation. – Opera Bot. 6/1: 1–187. Persson Ĺ. (1962): Mire and spring vegetation in an area north of lake Torneträsk, Torne Lappmark, Sweden. II. Habitat conditions. – Opera Bot. 6/3: 1–100. Peterka T. (2013): Doplněk k rozšíření druhu Paludella squarrosa na Českomoravské vrchovině [A supplement to the distribution of Paludella squarrosa in the Bohemian-Moravian Highlands (Czech Republic)]. – Bryonora 52: 31–35. Pott R. (1992): Die Pflanzengesellschaften Deutschlands. – Eugen Ulmer, Stuttgart. Poulíčková A., Hájek M. & Rybníček K. (2005): Ecology and palaeoecology of spring fens of the West Carpathians. – Univerzita Palackého, Olomouc. Prentice H. C. & Prentice I. C. (1975): The hill vegetation of North Hoy, Orkney. – New Phytol. 75: 313–367. Peterka et al.: Differentiation of rich fens 365

Rozbrojová Z. & Hájek M. (2008): Changes in nutrient limitation of spring fen vegetation along environmental gradients in the West Carpathians. – J. Veg. Sci. 19: 613–620. Růžička I. (1989): Výsledky záchranného výzkumu ohrožené květeny mizejících rašelinišť a rašelinných luk na Jihlavsku [The resuts of rescue research of endangered flora of disappearing mires and mire meadows in the Jihlavsko region]. – Vlastiv. Sborn. Vysočiny, Odd. Věd Přír. 9: 135–176. Růžička I. (1999): Floristický materiál z území CHKO Žďárské vrchy [The floristic material from the region of the Protected Landscape Area Žďárské vrchy Hills]. – Vlastiv. Sborn. Vysočiny, Odd. Věd Přír. 14: 63–93. Rybníček K. (1966): Glacial relics in the bryoflora of the highlands Českomoravská vrchovina (Bohemian- Moravian Highlands); their habitat and cenotaxonomic value. – Folia Geobot. Phytotax. 1: 101–119. Rybníček K. (1974): Die Vegetation der Moore im südlichen Teil der Böhmisch-Mährischen Höhe. – Vegetace ČSSR A6, Academia, Praha. Rybníček K., Balátová-Tuláčková E. & Neuhäusl R. (1984): Přehled rostlinných společenstev rašelinišť a mokřadních luk Československa [Plant communities of mires and wet meadows in Czechoslovakia]. – Stud. ČSAV 1984/8: 1–124. Schenková V., Čáp Hlaváč J. & Horsák M. (2013): Vrkoč rašelinný – další z glaciálních reliktů. Z červené knihy našich měkkýšů [Vertigo liljeborgi: another glacial relict. From the Red book of our molluscs]. – Živa 61: 73–74. Schenková V. & Horsák M. (2013): Nové nálezy vrkoče Geyerova potvrzují jeho ohroženost – z červené knihy našich měkkýšů [New findings of Vertigo geyeri confirm its vulnerability: from the Red book of our molluscs]. – Živa 61: 238–239. Schenková V., Horsák M., Hájek M., Plesková Z., Dítě D. & Pawlikowski P. (2014): Mollusc and plant assemblages controlled by different ecological gradients at Eastern European fens. – Acta Oecol. 56: 66–73. Schenková V., Horsák M., Plesková Z. & Pawlikowski P. (2012): Habitat preferences and conservation of Vertigo geyeri (Gastropoda: Pulmonata) in Slovakia and Poland. – J. Mollusc. Stud. 78: 105–111. Schmidtlein S., Tichý L., Feilhauer H. & Faude U. (2010): A brute-force approach to vegetation classification. – J. Veg. Sci. 21: 1162–1171. Sekulová L., Hájek M., Hájková P., Mikulášková E. & Rozbrojová Z. (2011): Alpine wetlands in the West Carpathians: vegetation survey and vegetation-environment relationships. – Preslia 83: 1–24. Sekulova L., Hájek M. & Syrovátka V. (2013): Vegetation environment relationships in alpine mires of the West Carpathians and the Alps. – J. Veg. Sci. 24: 1118–1128. Sjörs H. (1948): Myrvegetation i Bergslagen [Mire vegetation in Bergslagen, Sweden]. – Acta Phytogeogr. Suec. 21: 1–340. Sjörs H. (1952): On the relation between vegetation and electrolytes in north Swedish mire waters. – Oikos 2: 241–258. Sjörs H. & Gunnarsson U. (2002): Calcium and pH in north and central Swedish mire waters. – J. Ecol. 90: 650–657. Smagin V. A. (1999): Rastiteľnosť jevtrofnych bolot severa jevropejskoj Rossii [Vegetation of eutrophic fens in the north of European Russia]. – Bot. Zhurn. 84: 75–85. Smagin V. A. (2007): Sojuz Bistorto-Caricion diandrae all. nov. na bolotach taježnoj zony jevropejskoj časti Rossii [Fen vegetation of alliance Bistorto-Caricion diandrae all. nov. of taiga zone of European Russia]. – Bot. Zhurn. 92: 1340–1365. Šmilauer P. & Lepš J. Š. (2014): Multivariate analysis of ecological data using CANOCO 5. – Cambridge Univ. Press, Cambridge. Snowden R. E. D. & Wheeler B. D. (1993): Iron toxicity to fen plant species. – J. Ecol. 81: 35–46. Soudzilovskaia N. A., Graae B. J., Douma J. C., Grau O., Milbau A., Shevtsova A., Wolters L. & Cornelissen J. H. C. (2011): How do bryophytes govern generative recruitment of vascular plants? – New. Phytol. 190: 1019–1031. Štechová T., Kučera J. & Šmilauer P. (2012): Factors affecting population size and vitality of Hamatocaulis vernicosus (Mitt.) Hedenäs (Calliergonaceae, Musci). – Wetl. Ecol. Manage. 20: 329–339. Steiner G. M. (1992): Österreichischer Moorschutzkatalog. Ed. 4. – Verlag Ulrich Moser, Graz/Wien. Steiner G. M. (1993): Scheuchzerio-Caricetea fuscae. – In: Grabherr G. & Mucina L. (eds), Die Pflanzengesellschaften Österreichs, Teil II. Natürliche waldfreie Vegetation, p. 131–165, Gustav Fischer Verlag, Jena etc. Tahvanainen T. (2004): Water chemistry of mires in relation to the poor-rich vegetation gradient and contrasting geochemical zones of north-eastern Fennoscandian shield. – Folia Geobot. 39: 353–369. Tichý L. (2002): JUICE, software for vegetation classification. – J. Veg. Sci. 13: 451–453. 366 Preslia 86: 337–366, 2014

Tyler G. (2003): Some ecophysiological and historical approaches to species richness and calcicole/calcifuge behaviour – contribution to a debate. – Folia Geobot. 38: 419–428. van Baaren M., During H. & Leltz G. (1988): Bryophyte communities in mesotrophic fens in the Netherlands. – Holarct. Ecol. 11: 32–40. van der Maarel E. (1979): Transformation of cover-abundance values in phytosociology and its effects on community similarity. – Vegetatio 39: 97–114 van der Welle M. E. W., Vermeulen P. J., Shaver G. R. & Berendse F. (2003): Factors determining plant species richness in Alaskan arctic tundra. – J. Veg. Sci. 14: 711–720. van Tongeren O., Gremmen N. & Hennekens S. (2008): Assignment of relevés to pre-defined classes by supervised clustering of plant communities using a new composite index. – J. Veg. Sci. 19: 525–536. Vitt D. H. (2000): Peatlands: ecosystems dominated by bryophytes. – In: Shaw A. J. & Goffinet B. (eds), Bryophyte biology, p. 312–343, Cambridge Univ. Press, Cambridge. Vitt D. H. & Chee W. L. (1990): The relationships of vegetation to surface water chemistry and peat chemistry in fens of Alberta, Canada. – Vegetatio 89: 87–106. Waughman G. J. (1980): Chemical aspects of the ecology of some south German peatlands. – J. Ecol. 68: 1025–1046. Wheeler B. D. & Proctor M. C. F. (2000): Ecological gradients, subdivisions and terminology of north-west Euro- pean mires. – J. Ecol. 88: 187–203. Wind-Mulder H. L., Rochefort L. & Vitt D. H. (1996): Water and peat chemistry comparisons of natural and post- harvested peatlands across Canada and their relevance to peatland restoration. – Ecol. Engin. 7: 161–181. Wojtuń B. (1994): Element contents of Sphagnum mosses of peat bogs of lower Silesia (Poland). – Bryologist 97: 284–295. Zak D., Gelbrecht J. & Steinberg C. E. W. (2004): Phosphorus retention at the redox interface of peatlands adjacent to surface waters in northeast Germany. – Biogeochemistry 70: 357–368. Zechmeister H., Tribsch A., Moser D. & Wrbka T. (2002): Distribution of endangered bryophytes in Austrian agricultural landscapes. – Biol. Conserv. 103: 173–182. Zechmeister H. G. & Steiner G. M. (1995): Quellfluren und Quellmoore des Waldviertels, Österreich. – Tuexenia 15: 161–197. Zohlen A. & Tyler G. (2000): Immobilization of tissue iron on calcareous soil, differences between calcicole and calcifuge plants. – Oikos 89: 95–106. Zoltai S. C. & Vitt D. H. (1995): Canadian wetlands: environmental gradients and classification. – Vegetatio 118: 131–137.

Received 5 June 2014 Revision received 2 October 2014 Accepted 7 October 2014 Appendix 1. – Header data of relevés including results of groundwater analyses.

+ - 3- 2+ + 2+ Relevé Nr. Cover total Cover E1 Cover E0 WTD pH conductivity NH4 NO3 PO4 Ca K Mg Fe [%] [%] [%] [cm] [µS.cm-1] [µg.l-1] [µg.l-1] [µg.l-1] [mg.l-1] [mg.l-1] [mg.l-1] [mg.l-1] 1 99 30 99 6 7.30 443.00 126.116 98.146 59.355 93.42 6.99 2.993 0.627 2 100 50 100 7 7.40 385.30 122.894 139.051 141.443 80.04 8.185 1.487 1.847 3 90 70 75 9 6.90 557.00 45.303 7.449 526.691 10.2 3.079 65.52 42.97 4 98 60 98 2 6.60 353.00 47.115 18.643 208.836 42.69 6.17 8.592 1.179 5 95 90 10 7 7.10 612.70 81.067 93.629 93.829 103.7 0.9425 13.21 109 6 100 50 100 7 7.60 334.30 155.458 399.536 95.436 59.53 10.365 2.044 0.6053 7 80 65 70 15 6.10 111.90 25.93 58.606 46.309 4.075 19.31 1.212 14.54 8 100 65 99 20 6.60 327.30 27.688 173.193 139.347 55.24 3.075 1.547 9.135 9 99 60 99 22 5.80 141.90 158.925 302.607 151.827 9.749 6.62 2.949 62 10 100 60 100 21 5.70 118.60 188.512 1124.001 91.728 17.09 3.409 2.718 34.48 11 90 70 85 14 5.70 109.80 108.54 293.548 168.756 3.44 3.682 2.002 6.708 12 98 60 96 14 6.00 108.10 234.504 276.576 80.298 5.158 19.15 2.091 71.04 13 99 60 98 17 6.20 210.30 53.174 449.013 148.873 25.22 2.765 4.697 3.82 14 99 40 99 13 5.70 87.20 58.154 96.07 102.925 4.56 9.965 1.687 32.27 15 98 35 95 12 6.40 144.00 157.461 398.536 152.509 11.9 3.472 3.519 4.435 16 100 60 97 6 6.30 345.70 65.477 6633 81.12 35.97 8.294 10.15 1.573 17 98 70 96 23 5.30 104.90 65.77 73.081 57.619 19.51 0.1578 2.259 10.7 18 100 70 97 11 6.20 117.00 40.577 24.893 126.552 12.35 2.795 1.934 35.32 19 95 80 85 12 6.50 248.00 40.77 790.26 125.789 30.52 1.6295 4.395 3.084 20 100 70 98 9 6.00 111.70 21.131 5.404 122.211 19.7 7.955 2.084 5.392 21 100 70 90 15 6.20 216.70 664.199 442.032 123.926 12.58 11.72 3.651 50.92 22 100 80 90 12 6.30 152.00 121.177 79.357 170.871 41.86 3.618 7.701 181.7 23 100 70 95 11 5.70 93.40 141.06 553.399 100.167 14.24 5.28 2.391 1.511 24 95 60 90 9 5.80 173.90 445.139 27.699 303.541 16.67 18.085 3.611 28.51 25 100 85 90 12 6.50 167.70 141.746 834.173 181.103 22.22 12.56 3.693 11.22 26 100 60 95 7 6.60 167.40 117.406 533.35 161.484 14.65 5.475 6.465 4.017

+ - 3- 2+ + 2+ Relevé Nr. Cover total Cover E1 Cover E0 WTD pH conductivity NH4 NO3 PO4 Ca K Mg Fe [%] [%] [%] [cm] [µS.cm-1] [µg.l-1] [µg.l-1] [µg.l-1] [mg.l-1] [mg.l-1] [mg.l-1] [mg.l-1] 27 100 85 95 11 6.90 273.00 88.952 554.076 63.772 68.54 3.34 1.292 1.9 28 95 45 95 7 5.80 88.40 249.733 634.734 158.067 21.07 4.433 3.77 4.087 29 99 50 99 8 5.30 106.10 71.629 213.233 34.13 1.496 6.165 1.21 13.48 30 100 45 100 15 5.20 62.40 165.956 101.696 143.274 4.11 3.146 1.832 13 31 99 65 99 17 5.10 92.60 24.759 27.891 90.855 6.232 3.543 1.098 61.02 32 100 40 99 16 5.80 115.40 140.177 568.65 117.6 10.28 0.3029 2.677 150.9 33 98 80 85 7 5.10 68.30 42.885 11.868 48.033 10.24 3.982 1.219 71.8 34 92 45 80 10 5.60 50.30 40.468 0 123.185 3.673 1.067 1.193 4.704 35 100 40 99 10 4.80 74.50 46.813 1.92 59.161 9.241 0.6166 1.283 44.26 36 100 99 40 10 5.20 87.5 553.126 113.247 97.627 20.36 8.135 2.641 95.9 37 95 60 70 7 5.60 109.70 105.407 94.263 83.026 18.78 3.098 2.181 25.23 38 100 95 60 8 5.80 91.20 119.462 400.391 76.753 21.45 3.566 2.769 16.73 39 95 60 90 5 5.60 109.50 148.259 8.32 263.256 21.21 7.745 3.889 9.252 40 100 30 100 7 5.30 57.5 124.262 2121 62.17 16.25 5.605 4.076 29.87 41 100 100 60 10 5.3 98.1 729.334 3965.0 143.957 16.05 11.48 2.344 19.09 42 100 80 100 15 5.4 107.2 975.134 505.445 166.141 16.18 18.105 2.989 34.38 43 100 100 40 7 5.6 110 321.039 717.527 108.562 15.99 7.88 2.213 18.67 44 100 30 100 20 4.6 55.7 390.631 219.247 64.649 12 4.563 1.47 14.3 45 100 30 100 11 3.70 16.10 522.465 98.747 46.708 2.472 13.455 0.505 7.331 46 100 25 100 30 4.00 80.70 199.351 433.557 89.018 8.561 5.36 2.613 24.09 47 98 25 96 10 4.80 69.70 399.723 6050.001 80.203 3.156 2.128 0.933 188.05 48 100 40 98 3 4.20 22.20 42.281 19.394 29.164 3.738 1.3605 0.5938 4.993 49 85 55 80 7 5.50 40.40 50.137 14.878 34.318 3.864 2.612 0.4537 33.41 50 98 25 98 50 4.70 15.30 259.22 13.269 89.074 3.013 3.015 0.2889 39.17 51 99 15 98 12 4.30 35.30 62.223 1.742 62.797 3.833 20.24 0.5904 49.54 52 98 30 98 0 3.90 55.50 69.172 63.614 51.274 5.784 0.1999 1.01 125.1 53 98 15 98 7 4.20 25.5 53.984 52.881 118.418 7.776 1.521 0.2627 39.81 54 100 50 100 32 4.20 76.2 576.095 374.333 132.473 11.77 21.06 1.518 57.44

+ - 3- 2+ + 2+ Relevé Nr. Cover total Cover E1 Cover E0 WTD pH conductivity NH4 NO3 PO4 Ca K Mg Fe [%] [%] [%] [cm] [µS.cm-1] [µg.l-1] [µg.l-1] [µg.l-1] [mg.l-1] [mg.l-1] [mg.l-1] [mg.l-1] 55 100 15 100 32 3.90 30.70 71.125 42.132 83.158 9.506 3.082 0.2089 61.17 56 100 20 100 15 3.9 60.2 71.468 226.305 55.116 6.5 2.593 2.299 3.386 57 100 35 100 10 4 24.6 45.071 247.288 138.626 9.27 1.77 0.4572 11.14

Appendix 2. – Contents of phosphorus, nitrogen and N:P ratios in biomass of dominant moss species.

Relevé Nr. Species N P N:P [mg.g-1] [mg.g-1] 1 Campylium stellatum 8.43 0.47 17.76 1 Palustriella commutata 6.97 0.31 22.19 1 Scorpidium cossonii 9.33 0.53 20.00 2 Campylium stellatum 15.64 0.44 35.33 2 Philonotis calcarea 8.92 0.43 20.80 2 Scorpidium cossonii 9.07 0.75 13.27 3 Bryum pseudotriquetrum 13.59 2.12 6.41 3 Cratoneuron filicinum 14.99 2.10 7.14 4 Bryum pseudotriquetrum 13.27 0.89 14.91 4 Campylium stellatum 12.72 0.50 25.64 4 Tomentypnum nitens 9.20 0.70 13.12 5 Calliergonella cuspidata 18.30 1.84 9.94 5 Plagiomnium affine agg. 13.67 2.32 5.89 6 Campylium stellatum 10.18 0.87 11.71 6 Scorpidium cossonii 12.73 1.27 10.01 6 Tomentypnum nitens 18.72 0.81 23.13 7 Climacium dendroides 11.78 0.84 13.95 7 Sphagnum contortum 10.56 0.56 18.99 7 Sphagnum warnstorfii 9.94 0.65 15.26 8 Aulacomnium palustre 8.86 0.60 14.70 8 Sphagnum teres 12.48 0.77 16.14 8 Sphagnum warnstorfii 9.86 0.53 18.67 8 Tomentypnum nitens 6.12 0.64 9.52 9 Aulacomnium palustre 9.26 1.40 6.62 9 Sphagnum warnstorfii 10.53 0.68 15.39 10 Sphagnum subsecundum 7.22 0.46 15.61 10 Sphagnum teres 9.08 0.86 10.56 11 Aulacomnium palustre 8.81 0.61 14.46 11 Calliergonella cuspidata 11.96 1.31 9.12 11 Sphagnum teres 8.73 1.14 7.66 11 Sphagnum warnstorfii 6.73 0.45 14.88 12 Aulacomnium palustre 7.47 0.70 10.60 12 Calliergonella cuspidata 10.47 0.64 16.42 12 Sphagnum contortum 6.76 0.66 10.24 12 Sphagnum teres 8.87 0.92 9.67 12 Sphagnum warnstorfii 6.29 0.66 9.58 13 Aulacomnium palustre 5.77 0.37 15.52 13 Polytrichum strictum 12.30 1.08 11.42 13 Sphagnum teres 5.81 0.56 10.32 13 Sphagnum warnstorfii 6.92 0.41 16.74 14 Sphagnum obtusum 6.31 0.47 13.29 14 Sphagnum subsecundum 6.31 0.52 12.18 14 Sphagnum teres 5.42 0.35 15.36 14 Sphagnum warnstorfii 4.55 0.38 12.01 15 Calliergonella cuspidata 10.94 0.47 23.07 15 Philonotis fontana 7.03 0.30 23.17 15 Scorpidium cossonii 10.52 0.39 26.71 15 Sphagnum contortum 7.06 0.43 16.53

Relevé Nr. Species N P N:P [mg.g-1] [mg.g-1] 15 Sphagnum subnitens 8.24 0.44 18.57 15 Sphagnum warnstorfii 7.87 0.57 13.75 16 Campylium stellatum 10.08 0.23 43.36 16 Scorpidium cossonii 14.20 0.33 42.45 16 Sphagnum contortum 10.96 0.52 21.18 16 Sphagnum warnstorfii 10.76 0.35 30.65 17 Aulacomnium palustre 9.15 1.21 7.55 17 Sphagnum contortum 12.78 0.55 23.21 17 Sphagnum fallax 5.74 0.61 9.44 17 Sphagnum palustre 8.51 0.80 10.69 17 Sphagnum teres 9.74 0.73 13.27 17 Sphagnum warnstorfii 5.47 0.86 6.39 18 Sphagnum palustre 11.05 1.32 8.37 18 Sphagnum teres 15.78 1.25 12.63 18 Sphagnum warnstorfii 8.93 0.65 13.64 18 Tomentypnum nitens 8.33 1.54 5.40 19 Calliergonella cuspidata 10.75 1.18 9.12 19 Sphagnum warnstorfii 7.32 0.34 21.66 20 Sphagnum teres 7.48 0.65 11.48 20 Sphagnum warnstorfii 9.96 0.41 24.08 21 Aulacomnium palustre 10.38 0.91 11.42 21 Sphagnum warnstorfii 11.81 0.64 18.42 21 Tomentypnum nitens 11.33 0.65 17.52 22 Calliergonella cuspidata 12.57 1.18 10.69 22 Sphagnum warnstorfii 10.42 0.63 16.55 22 Tomentypnum nitens 11.24 1.03 10.88 23 Sphagnum flexuosum 8.74 0.82 10.68 23 Sphagnum teres 13.19 0.58 22.86 23 Sphagnum warnstorfii 8.92 0.46 19.28 24 Calliergonella cuspidata 10.90 0.71 15.30 24 Sphagnum warnstorfii 11.04 0.75 14.77 25 Aulacomnium palustre 16.36 1.13 14.42 25 Calliergonella cuspidata 14.46 1.61 8.97 25 Climacium dendroides 15.15 0.80 18.86 25 Sphagnum palustre 9.80 0.60 16.37 25 Sphagnum teres 17.59 1.06 16.62 25 Sphagnum warnstorfii 9.42 0.79 11.87 25 Tomentypnum nitens 15.60 1.06 14.75 26 Campylium stellatum 13.00 0.60 21.72 26 Sphagnum warnstorfii 11.69 0.48 24.51 26 Tomentypnum nitens 10.12 0.58 17.41 27 Aulacomnium palustre 10.61 1.00 10.66 27 Sphagnum teres 9.27 0.68 13.72 27 Sphagnum warnstorfii 8.45 0.65 13.03 27 Tomentypnum nitens 9.61 1.07 8.94 28 Sphagnum contortum 6.04 0.30 19.82 28 Sphagnum warnstorfii 8.59 0.56 15.45 29 Sphagnum flexuosum 9.24 0.50 18.34 29 Sphagnum teres 16.99 1.09 15.53 30 Polytrichum commune 16.77 2.39 7.01 30 Sphagnum angustifolium 8.56 0.79 10.88

Relevé Nr. Species N P N:P [mg.g-1] [mg.g-1] 31 Sphagnum fimbriatum 15.79 0.78 20.19 31 Sphagnum flexuosum 7.96 0.58 13.60 31 Sphagnum subsecundum 7.86 0.46 16.94 31 Sphagnum warnstorfii 7.35 0.74 9.88 32 Sphagnum flexuosum 5.12 0.63 8.07 32 Sphagnum palustre 11.90 0.81 14.69 32 Sphagnum russowii 10.55 0.58 18.24 33 Sphagnum contortum 7.98 0.95 8.41 33 Sphagnum palustre 10.63 0.92 11.50 34 Calliergonella cuspidata 12.27 0.69 17.87 34 Sphagnum flexuosum 5.74 0.38 15.30 34 Sphagnum palustre 7.96 0.43 18.35 34 Sphagnum subsecundum 9.20 0.83 11.03 34 Sphagnum teres 15.62 1.22 12.77 35 Sphagnum fimbriatum 9.21 0.49 18.77 35 Sphagnum subsecundum 9.06 0.77 11.71 36 Sphagnum fallax 14.22 0.79 17.97 37 Aulacomnium palustre 8.11 0.50 16.09 37 Calliergonella cuspidata 12.67 0.63 20.03 37 Sphagnum teres 11.61 0.70 16.52 38 Aulacomnium palustre 9.55 0.92 10.36 38 Sphagnum teres 7.23 0.51 14.27 39 Calliergon cordifolium 23.19 2.47 9.38 39 Calliergonella cuspidata 11.49 1.52 7.57 39 Sphagnum subsecundum 12.34 1.11 11.13 40 Sphagnum flexuosum 8.41 0.65 12.93 41 Calliergonella cuspidata 14.46 1.35 10.70 41 Sphagnum teres 19.34 1.04 18.57 42 Sphagnum palustre 9.06 0.39 22.97 42 Sphagnum teres 11.78 0.63 18.64 43 Calliergonella cuspidata 15.62 1.00 15.57 43 Sphagnum teres 14.97 1.08 13.88 44 Sphagnum flexuosum 12.78 0.41 31.17 44 Sphagnum palustre 15.51 0.54 28.49 45 Sphagnum fallax 14.05 0.39 36.39 46 Aulacomnium palustre 10.04 0.47 21.38 46 Polytrichum strictum 6.43 0.63 10.21 46 Sphagnum capillifolium 4.28 0.24 17.57 46 Sphagnum fallax 4.48 0.28 15.99 46 Sphagnum papillosum 5.87 0.36 16.13 47 Sphagnum fallax 4.76 0.24 19.51 47 Sphagnum papillosum 9.00 0.67 13.48 48 Sphagnum fallax 10.45 0.57 18.47 48 Sphagnum inundatum 10.22 0.64 15.92 48 Sphagnum palustre 9.13 0.51 17.86 49 Sphagnum capillifolium 6.19 0.29 21.06 49 Sphagnum fallax 3.76 0.68 5.50 49 Sphagnum papillosum 12.13 0.32 38.32 49 Sphagnum teres 13.14 0.99 13.21 49 Sphagnum subsecundum 10.71 0.44 24.17 50 Polytrichum commune 16.01 1.14 14.04

Relevé Nr. Species N P N:P [mg.g-1] [mg.g-1] 50 Sphagnum fallax 6.17 0.44 13.91 51 Sphagnum fallax 12.15 0.45 27.23 51 Sphagnum palustre 16.15 0.42 38.09 52 Sphagnum denticulatum 9.67 0.56 17.29 52 Sphagnum fallax 4.98 0.32 15.80 52 Sphagnum palustre 9.68 0.64 15.16 53 Sphagnum fallax 13.11 0.38 34.86 53 Sphagnum papillosum 12.05 0.18 67.77 54 Polytrichum commune 17.38 0.88 19.80 54 Sphagnum fallax 14.04 0.39 35.95 54 Sphagnum palustre 6.02 0.58 10.42 54 Sphagnum russowii 10.37 0.57 18.30 55 Polytrichum commune 20.73 1.52 13.68 55 Sphagnum fallax 10.93 0.88 12.38 56 Polytrichum commune 14.65 1.31 11.16 56 Sphagnum flexuosum 7.64 0.64 11.94 56 Sphagnum palustre 9.90 1.92 5.16 57 Sphagnum fallax 11.84 0.62 19.13

Appendix 3. – Abbreviations and full names of plant species.

Vascular plants: AgrCan = Agrostis canina, AngSyl = Angelica sylvestris, AntOdo = Anthoxanthum odoratum, BriMed = Briza media, CarCan = Carex canescens, CarDav = Carex davalliana, CarDem = Carex demissa, CarDia = Carex diandra, CarEch = Carex echinata, CarNig = Carex nigra, CarPan = Carex panicea, CarPra = Cardamine pratensis, CarPul = Carex pulicaris, CarRos = Carex rostrata, CirPal = Cirsium palustre, ComPal = Comarum palustre, CrePal = Crepis paludosa, DacMaj = Dactylorhiza majalis, DroRot = Drosera rotundifolia, EpiPal = Epilobium palustre, EquFlu = Equisetum fluviatile, EquPal = Equisetum palustre, EriAng = Eriophorum angustifolium, FesFil = Festuca filiformis, FesRub = Festuca rubra agg., FilUlm = Filipendula ulmaria, GalPal = Galium palustre agg., GalUli = Galium uliginosum, HolLan = Holcus lanatus, JunArt = Juncus articulatus, JunBul = Juncus bulbosus, JunEff = Juncus effusus, LuzMul = Luzula multiflora, LycFlo = Lychnis flos-cuculi, LysVul = Lysimachia vulgaris, MenTri = Menyanthes trifoliata, MolCae = Molinia caerulea agg., ParPal = Parnassia palustris, PeuPal = Peucedanum palustre, PhrAus = Phragmites australis, PotEre = Potentilla erecta, RanAcr = Ranunculus acris, RanAur = Ranunculus auricomus agg., RumAce = Rumex acetosa, SanOff = Sanguisorba officinalis, SucPra = Succisa pratensis, TriAlp = Trichophorum alpinum, VacOxy = Vaccinium oxycoccos, ValDio = Valeriana dioica, VioPal = Viola palustris.

Bryophytes: AnePin = Aneura pinguis, AulPal = Aulacomnium palustre, BrePra = Breidleria pratensis, BryPse = Bryum pseudotriquetrum, CalCus = Calliergonella cuspidata, CamSte = Campylium stellatum, CliDen = Climacium dendroides, PlaAff = Plagiomnium affine agg., PolCom = Polytrichum commune, PolStr = Polytrichum strictum, ScoCos = Scorpidium cossonii, SphCon = Sphagnum contortum, SphFal = Sphagnum fallax, SphFle = Sphagnum flexuosum, SphPal = Sphagnum palustre, SphSub = Sphagnum subsecundum, SphTer = Sphagnum teres, SphWar = Sphagnum warnstorfii, StrStr = Straminergon stramineum, TomNit = Tomentypnum nitens.

Paper 2 Peterka T., Jiroušek M., Hájek M. & Jiménez-Alfaro B. (2015): European Mire Vegetation Database: a gap-oriented database for European fens and bogs. – Phytocoenologia 45: 291–297.

European Mire Vegetation Database: a gap-oriented database for European fens and bogs

Tomáš Peterka 1, Martin Jiroušek 1,2, Michal Hájek 1 & Borja Jiménez-Alfaro 1

1Department of Botany and Zoology, Faculty of Science, Masaryk University, Kotlářská 2, 61137 Brno, Czech Republic 2Department of Plant Biology, Faculty of Agronomy, Mendel University in Brno, Zemědělská 1, 61300 Brno, Czech Republic

Abstract: The attempt to produce a harmonized classification of European mires and to conduct a syntaxonomical analysis on the basis of individual relevés has led to the creation of the European Mire Vegetation Database (GIVD ID: EU-00-022, http://www.givd.info/ID/EU-00-022). The database is managed by the Mire Ecology Working Group (Department of Botany and Zoology, Masaryk University, Brno). In May 2015 this database was comprised of 10,147 relevés of the classes Scheuchzerio palustris-Caricetea nigrae and Oxycocco-Sphagnetea published in various monographs, manuscripts or journals, but not stored in any other national or regional electronic vegetation databases. Only relevés where bryophytes were identified as well as vascular plants were computerized. Most of the newly digitized data are from Northern and Southeastern European regions. Geographical coordinates are available for individual vegetation plots, their accuracy depends on the precision of the location description given. The European Mire Vegetation Database has been integrated in the European Vegetation Archive as a repository of mire vegetation not included in other national or regional databases.

Keywords: EVA; phytosociology; relevé; Turboveg; vegetation plot; vegetation sampling.

Abbreviations: EVA = European Vegetation Archive; GIVD = Global Index of Vegetation-Plot Databases.

Introduction Modern technologies and advances in software development, especially the Turboveg database management system (Hennekens & Schaminée 2001), have led to the creation of national and regional vegetation databases (Schaminée et al. 2009). Besides other uses, these databases enabled the realization of large-scale vegetation syntheses and syntaxonomical analyses spanning national boundaries (Botta-Dukát et al. 2005; Illyés et al. 2007; Michl et al. 2010; Sekulová et al. 2011; Eliáš et al. 2013). Many databases have recently been registered in the Global Index of Vegetation-Plot Databases (GIVD; Dengler et al. 2011). Moreover, the need for scientific cooperation and easier data exchange among European countries has sparked the creation of the European Vegetation Archive (EVA; Jiménez-Alfaro et al. 2013; Chytrý et al. 2015). In 2013, the Mire Ecology Working Group (Department of Botany and Zoology, Masaryk University, Brno), under the leadership of the third author (MH), started the project entitled "Classification of mire communities on the European scale". The main aim was to establish a consistent vegetation system of European fens (Scheuchzerio palustris-Caricetea nigrae Tx. 1937) and bogs (Oxycocco-Sphagnetea Br.-Bl. & Tx. ex Westhoff et al. 1946) to alliance level, since national systems are largely incompatible because different classification approaches are used in different countries (Rybníček 1981; Dierssen 1982; Malmer 1985; Hájek et al. 2002; Smagin 2012; Peterka et al. 2014). However, a pan-European classification

77 of both vegetation types is only possible by analyzing a comprehensive data set with individual phytosociological relevés from the whole of Europe, and this information is not fully available at the continental scale (Jiménez-Alfaro et al. 2014). Before digitizing new relevés, we checked the available data in (i) the databases stored within EVA, (ii) other vegetation databases registered in GIVD (e.g. Belarus Peatland Restoration Project, Swiss Mire Monitoring) and (iii) relevés sampled by members of the working group and of about twenty of their foreign collaborators. However, some European regions still had insufficient relevé cover or none at all. There was conspicuous contrast between the thousands of digitized vegetation plots from countries of Central and Western Europe to the very few or no computerized plots from countries of Northern and Southeastern Europe (Fig. 1), which seems to be the general pattern (see Schaminée et al. 2009; Jiménez-Alfaro et al. 2014). The easiest solution was to computerize vegetation relevés from individual “empty” regions which were scattered in monographs, manuscripts or local journals. This gave rise to the European Mire Vegetation Database (ID EU-00-022 in GIVD) containing previously undigitized relevés and covering regional gaps in digital data for mire communities. Since March 2015 the database has been stored in EVA. In this article we introduce the European Mire Vegetation Database and describe its content, data quality and further development of this new source of data for vegetation analysis.

Fig. 1. Distribution of the georeferenced vegetation plots assigned to the classes of Scheuchzerio palustris- Caricetea nigrae and Oxycocco-Sphagnetea in Europe. Available relevés stored in EVA, other GIVD or private databases and relevés stored in European Mire Vegetation Database are shown. The map refers to the state of the datasets in May 2015. The list of data sources is available from first author upon request. The map was created using the DIVA-GIS software (http://www.diva-gis.org/).

Database structure, content and constraints Because the database custodians and the people who helped prepare the data are most familiar with the Czech National Phytosociological Database (Chytrý & Rafajová 2003), the organizational structure of the European Mire Vegetation Database follows the model of this rigorously managed national database. As of 13th May 2015, the European Mire Vegetation Database contained 10,147 relevés obtained from articles, manuscripts and monographs (Supplement S1). The database was created in two steps. Firstly, all available fen and bog data were digitized from those regions with general lack or total absence of available digitized vegetation plots (Fig. 1), namely Iceland, Svalbard, Greenland, Fennoscandia and former Yugoslavia (with the exception of Croatia and Slovenia). Secondly, we computerized

78 relevés from selected publications to fill in the most apparent gaps in the national or regional databases. These gaps comprised of small or medium-sized geographic areas as well as gaps relating to specific vegetation types or plot size. These data sources covered different European regions, e.g. Estonia, Italy, Poland, Romania or Russian Federation (see Table 1).

Table 1. Numbers and sizes of vegetation relevés from individual regions stored in the European Mire Vegetation Database. Values in brackets indicate the size of one or a few relevés that fall beyond the range of plot sizes usual in the particular region. Region Number of plots Plot sizes (m2) Bosnia and Herzegovina 68 100(–200) Croatia 6 6–100 Denmark 9 1–4 Estonia 159 1 Faroe Islands and neighbouring parts of Great Britain 21 1 Finland 1635 1–100, unknown Greenland 153 1–4 Iceland 406 (0.5)1–25(100) Ireland 108 1–60 Italy (except Sicily) 314 1–100 Latvia 14 9–60 Montenegro and Kosovo 26 1–100 Norway (except Svalbard) 2784 1–20(60) Poland 203 5–100 Romania 179 2–100(200), unknown Russian Federation 1485 1–100, unknown Serbia 42 25–100 Sicily 12 15–100 Svalbard 179 1–10(25) Sweden 2316 (0.25)1–4(100) Ukraine 28 100

Individual relevés were chosen on the basis of (1) the original authors’ assignment to the classes Scheuchzerio palustris-Caricetea nigrae or Oxycocco-Sphagnetea and (2) expert evaluation by the database custodians. The relevés were originally sampled between 1910 and 2013, about half of them (48.5 %) were recorded between the 1920s and 1970s (Fig. 2). It is important to note that several quoted literary sources also contained non-mire relevés, which were not computerized. Only relevés that contained records of both bryophytes and vascular plants were included. Species data were digitized using the checklist of European vascular plants, bryophytes, lichens and macro-algae held within Turboveg 2. Present header data (covers of individual layers, total cover, plot size, altitude, locality, pH, original syntaxa) were computerized when they were available. One general problem found in all vegetation databases that comprise of diverse original data sources, is the varying quality of primary data and the heterogeneity of sampling design (Michalcová et al. 2011; Apostolova et al. 2012). In the data compiled for the European Mire Vegetation Database, vegetation plots vary mainly in their size (Table 1) and in the cover-abundance scale applied. In addition, some phytosociological approaches use specific scales that can be hard to convert into any percentage-based cover scale. This is especially the case of Norrlin´s scale used by several Finnish botanists in the first half of the 20th century (e.g. Paasio 1933; Brandt 1948). This scale expresses species abundance as shoot density (Lawesson et al. 2000). Thus, the most straightforward use of plot data available in the European Mire Vegetation Database seems to be to convert species cover values from Norrlin´s scale to simple presence/absence data.

1910–1919 1920–1929 1930–1939 1940–1949 1950–1959 1960–1969 1970–1979 1980–1989 1990–1999 2000–2009 2010–2013 Unknown/uncertain

0 500 1000 1500 2000 2500 3000 Number of relevés

Fig. 2. Counts of vegetation plots sampled within particular decades stored in European Mire Vegetation Database.

Another problem relating to the quality of relevé sources is the absence of important header data (e.g. covers of individual layers, total cover, altitude, date) and also the precision of geographic coordinates or the lack thereof. Only a minority of relevés were located using exact coordinates. However, the geographic sites of many phytosociological records can be determined quite accurately from maps or detailed descriptions given in the original source. Other vegetation plots were located rather broadly, i.e. with respect to the nearest village, town, lake or a large geographic area such as a mountain range or an administrative district. To reflect the heterogeneity of this information, the lack of precision of coordinates in the database is indicated in the Bias_min field by numbers expressing the inexactness of longitude and latitude in geographic minutes. Unfortunately, the coordinates of 163 relevés could not be traced due to very vague or missing location descriptions. Another crucial source of bias in vegetation databases was particularly the differences in the taxonomic treatment in different countries or time periods (Jansen & Dengler 2010). Although a certain homogenization of taxon names can be addressed through importing the data into Turboveg 2, merging the data with other databases (e.g. EVA) will require a new integration of species names in Turboveg 3 by using the SynBioSys Taxon Database (Chytrý et al. 2015). In any case, this tool provides a priori species links that must be carefully checked prior to any analysis. In particular, it will be necessary to merge selected taxa into species groups or aggregates to avoid potential taxonomic bias even at the cost of information loss (Jansen & Dengler 2010). As an example, table 2 shows the most frequent species in the European Mire Vegetation Database and possible species aggregates to be considered before data analysis.

Future perspectives After the digitization of a few missing relevés, the European Mire Vegetation Database will serve as one of the main data sources for the classification of European mires, supplementing relevés provided by regional collaborators and those available from EVA or other databases. Within the purview of this major project, fen and bog vegetation will be standardized and formalized to alliance level. Special attention will also be paid to internal variability and the syntaxonomy of extremely rich fens (alliance Caricion davallianae Klika

1934; Jiménez- Alfaro et al. 2014) and rich fens (alliance Sphagno warnstorfii-Tomentypnion nitentis Dahl 1956 and related communities). The database will be a suitable data source for other non-commercial scientific projects. The whole database or subsets of same are available through EVA (Chytrý et al. 2015) and may be provided to applicants upon approval by one of the database custodians (restricted access regime). For detailed instructions how to obtain data, see EVA website.

Table 2. The 30 most frequent species within the database. Species groups or aggregates are proposed to reduce taxonomic bias by including closely related taxa. N. = number of occurences. agg./gr. = proposed aggregates or species groups.

Taxon N. agg./gr. Taxon N. agg./gr. Andromeda polifolia 1,901 Straminergon stramineum 742 Eriophorum vaginatum 1,471 Carex limosa 733 Eriophorum angustifolium 1,359 Aulacomnium palustre 699 Vaccinium oxycoccos 1,324 Vaccinium oxycoccos s.lat Carex lasiocarpa 683 (incl. V. microcarpum) Carex rostrata 1,128 Sphagnum fuscum 638 Rubus chamaemorus 1,001 Potentilla palustris 637 Menyanthes trifoliata 983 Polytrichum strictum 622 Drosera rotundifolia 977 Molinia caerulea 615 Molinia caerulea agg. (incl. M. arundinacea) Betula nana 954 Scorpidium revolvens 608 Scorpidium revolvens s.lat. (incl. S. cossonii) Vaccinium uliginosum 861 Equisetum fluviatile 591 Sphagnum angustifolium 851 Sphagnum recurvum agg. Empetrum nigrum 587 Empetrum nigrum s.lat. (incl. S. fallax, S. (incl. E. hermaphroditum) flexuosum) Calluna vulgaris 790 Pinus sylvestris 586 (all layers merged) Campylium stellatum 783 Campylium stellatum s. Scorpidium scorpioides 580 lat. (incl. C. protesum) Carex nigra 769 Pleurozium schreberi 580 Sphagnum magellanicum 756 Polygonum viviparum 514

Acknowledgements We are very grateful to our collaborators who provided hard-to-obtain regional papers, monographs or manuscripts containing full relevés, namely P. Lazarevid, O. Kuznetsov, R. Heikkilä, C. Marceno, C. Bita-Nicolae, V. Randjelovid and T. Tahvanainen. Several important data sources were provided by K. Rybníček († 2014), who established the tradition of research of mires and their syntaxonomy and ecology in the Czech Republic. We also wish to thank all our colleagues who helped us create the database, especially those who computerized relevés and looked up their coordinates: D. Dítě, Z. Fajmonová, E. Hettenbergerová, J. Jiroušková, J. Němec, P. Novák, Z. Plesková, H. Sekerková, L. Sekulová, M. Táborská and K. Vincenecová. Special thanks are due to M. Chytrý, who suggested that we prepare this article and helped us to launch the entire project. We are very grateful to F. Jansen and A. Moen for comments on a previous version of the text. The creation of the database and the preparation of the article were funded by the Grant Agency of the Czech Republic (project No. 14- 36079G) and by Masaryk University, mainly as part of internal project MUNI/A/1456/2014.

References Apostolova, I., Sopotlieva, D., Pedashenko, H., Velev, N. & Vasilev, K. 2012. Bulgarian Vegetation Database: historic background, current status and future prospects. Biodiversity & Ecology 4: 141–148. Botta-Dukát, Z., Chytrý, M., Hájková, P. & Havlová, M. 2005. Vegetation of lowland wet meadows along a climatic continentality gradient in central Europe. Preslia 77: 89–111. Brandt, A. 1948. Über die Entwicklung der Moore im Künstengebiet von Süd-Pohjanmaa am Bottnischen Meerbusen. Annales Botanici Societatis Zoologicæ Botanicæ Fennicæ 'Vanamo' 23: 1–134.

Chytrý, M., Hennekens, S.M., Jiménez-Alfaro, B., Knollová, I., Dengler, J., Jansen, F., Landucci, F., Schaminée, J.H.J., Adid, S., (…) & Yamalov, S. 2015. European Vegetation Archive (EVA): an integrated database of European vegetation plots. Applied Vegetation Science. doi: 10.1111/avsc.12191. Chytrý, M. & Rafajová, M. 2003. Czech National Phytosociological Database: basic statistics of the available vegetation-plot data. Preslia 75: 1–15. Dengler, J., Jansen, F., Glöckler, F., Peet, R.K., De Cáceres, M., Chytrý, M., Ewald, J., Oldeland, J., Lopez- Gonzalez, G., (...) & Spencer, N. 2011. The Global Index of Vegetation-Plot Databases (GIVD): a new resource for vegetation science. Journal of Vegetation Science 22: 582–597. Dierssen, K. 1982. Die wichtigsten Pflanzengesellschaften der Moore NW-Europas. Conservatoire et Jardin botaniques, Genève, CH. Eliáš, P., Sopotlieva, D., Dítě, D., Hájková, P., Apostolova, I., Senko, D., Melečková, Z. & Hájek, M. 2013. Vegetation diversity of salt-rich grasslands in Southeast Europe. Applied Vegetation Science 16: 521– 537. Hájek, M., Hekera, P. & Hájková, P. 2002. Spring fen vegetation and water chemistry in the Western Carpathian flysch zone. Folia Geobotanica 37: 205–224. Hennekens, S.M. & Schaminée, J.H.J. 2001. TURBOVEG, a comprehensive database management system for vegetation data. Journal of Vegetation Science 12: 589–591. Illyés, E., Chytrý, M., Botta-Dukát, Z., Jandt, U., Škodová, I., Janišová, M., Willner, W. & Hájek, O. 2007. Semi-dry grasslands along a climatic gradient across central Europe: vegetation classification with validation. Journal of Vegetation Science 18: 835–846. Jansen, F. & Dengler, J. 2010. Plant names in vegetation databases – a neglected source of bias. Journal of Vegetation Science 21: 1179–1186. Jiménez-Alfaro, B., Apostolova, I., Čarni, A., Chytrý, M., Csiky, J., Dengler, J., Dimopoulos, P., Font, X., Golub, V., (...) & Yamalov, S. 2013. Unifying and analysing vegetation-plot databases in Europe: the European Vegetation Archive (EVA) and the Braun-Blanquet project. In: Walker, D.A., Breen, A.L., Raynolds, M.K., Walker, M.D. (eds.) Arctic Vegetation Archive (AVA) Workshop, Krakow, Poland, April 14–16, 2013, pp. 50–51. CAFF, IS. Jiménez-Alfaro, B., Hájek, M., Ejrnaes, R., Rodwell, J., Pawlikowski, P., Weeda, E.J., Laitinen, J., Moen, A., Bergamini, A., (...) & Díaz, T.E. 2014. Biogeographic patterns of base-rich fen vegetation across Europe. Applied Vegetation Science 17: 367–380. Lawesson, J.E. (ed.) 2000. A Concept for Vegetation Studies and Monitoring in the Nordic Countries. Nordic Council of Ministers, Copenhagen, DK. Malmer, N. 1985. Remarks to the classification of mires and mire vegetation – Scandinavian arguments. Aquilo Series Botanica 21: 9–17. Michalcová, D., Lvončík, S., Chytrý, M. & Hájek, O. 2011. Bias in vegetation databases? A comparison of stratified-random and preferential sampling. Journal of Vegetation Science 22: 281–291. Michl, T., Dengler, J. & Huck, S. 2010. Montane-subalpine tallherb vegetation (Mulgedio-Aconitetea) in central Europe: large-scale synthesis and comparison with northern Europe. Phytocoenologia 40: 117–154. Paasio, I. 1933. Über die Vegetation der Hochmoore Finnlands. Acta Forestalia Fennica 39(3): 1–190. Peterka, T., Plesková, Z., Jiroušek, M. & Hájek, M. 2014. Testing floristic and environmental differentiation of rich fens on the Bohemian Massif. Preslia 86: 337–366. Rybníček, K. 1981. Problematika klasifikace rašelinných společenstev *Problems of the Classification of Mire Communities]. Zprávy České botanické společnosti, Materiály 2: 65–70. [In Czech.] Schaminée, J.H.J., Hennekens, S.M., Chytrý, M. & Rodwell, J.S. 2009. Vegetation-plot data and databases in Europe: an overview. Preslia 81: 173–185. Sekulová, L., Hájek, M., Hájková, P., Mikulášková, E. & Rozbrojová, Z. 2011. Alpine wetlands in the West Carpathians: vegetation survey and vegetation–environment relationships. Preslia 83: 1–24. Smagin, V.A. 2012. Syntaxonomy of ridge dwarf shrub-grass-peatmoss communities in Aapa-mires and fens of European Russia. Botanicheskii zhurnal 97: 939–960.

Supplement S1. List of computerized articles, monographs and manuscripts.

Almquist, E. 1929. Upplands vegetation och flora [Uplands vegetation and flora]. Acta Phytogeographica Suecica 1: 1–732. [In Swedish.] Antipin, V.K. & Boichuk, M.A. 2004. Sfagnovyye soobshchestva s Molinia caerulea (Poaceae) na Onezhsko-Pechorskikh aapa bolotakh [Sphagnum communities with Molinia caerulea (Poaceae) in Onega-Pechora aapa mires]. Botanicheskii zhurnal 89: 244–251. [In Russian.] Arwidsson, T. 1943. Studien über die Gefässpflanzen in den Hochgebirgen der Pite Lappmark. Acta Phytogeographica Suecica 17: 10–274. Birks, H.J.B. & Walters, S.M. (1973): The flora and vegetation of Barno jezero, Durmitor, Montenegro. Glasnik Republičkog zavoda za zaštitu prirode – Prirodnjačkog muzeja Titograd 5: 5–23. Blinova, I.V. & Petrovskij, M.N. 2014. K harakteristike minerotrofnyh travjanyh bolot v central’noj časti Murmanskoj oblasti i o neobhodimosti ih ohrany [Base-Rich Fens in the Central Part of Murmansk Region and the Case for Their Protection]. Vestnik Kol’skogo naučnogo centry 18: 38–55. [In Russian.] Blinova, I.V. & Uotila, P. 2013. Schoenus ferrugineus (Cyperaceae) in Murmansk Region (Russia). Memoranda Soc. Fauna Flora Fennica 89: 65–74. Bogdanovskaya-Giyenef, I.D. 1928. Rastitel'nyy pokrov verkhovykh bolot russkoy Pribaltiki [Bog vegetation in Russian Baltic]. Trudy Petergofskogo estestvenno-nauchnogo instituta 5: 265–377. [In Russian.] Böcher, W. 1954. Oceanic and continental vegetational complexes in Southwest Greenland. Meddelelser om Grønland 148(1): 1–336. Booberg, G. 1930. Gisselçsmyren en växtsociologisk och utvecklingshistorisk monografi över en Jämtländsk kalkmyr. Almqvist & Wiksells Boktryckeri, Uppsala. Brandt, A. 1948. Über die Entwicklung der Moore im Künstengebiet von Süd-Pohjanmaa am Bottnischen Meerbusen. Annales Botanici Societatis Zoologicæ Botanicæ Fennicæ 'Vanamo' 23: 1–134. Burescu, P. & Togor, G. 2010. Phytocoenological studies on oligotroph peat bog of Bihorului mountains. Studia Universitatis “Vasile Goldiş”, Seria Ştiinţele Vieţii 20: 71–81. Coldea, G. & E. Plæmadæ. 1989. Vegetaţia mlaştinilor oligotrofe din Carpaţii Româneşti (Cl. Oxycocco-Sphagnetea Br.-Bl. et Tx. 43) [Oligotrophic wetland vegetation in the Romanian Carpaths (cl. Oxycocco-Sphagnetea Br.-Bl. et Tx. 43)]. Contribuţii Botanice 1989: 37-43. [In Romanian.] Coldea, G., & Plæmadæ, E. 1970. Contribuţii la studiul clasei Scheuchzerio–Caricetea fuscae Nordh. 1936 din România [Contributions to the study of the Scheuchzerio-Caricetea fuscae Nordh. 1936 class in Romania]. Hidrobiologia 11: 105– 116. [In Romanian.] Coldea, G., Plæmadæ, E. & Bartok, E. 1977. Contribuţii la studiul clasei Scheuchzerio-Caricetea fuscae Nordh. 1936 din România (IV) *Contributions to the study of the Scheuchzerio-Caricetea fuscae Nordh. 1936 class in Romania (IV)]. Contribuţii botanice 1977: 69–78. [In Romanian.] Čolid, D., 1965: Nova nalazišta rosulje (Drosera rotundifolia L.) na Staroj planini - istočna Srbija *New localities of Drosera rotundifolia L. in the Stara Planina - Eastern Serbia]. Zaštita prirode 29–30: 5–23. [In Serbian.] Críodáin, C.O. & Doyle, G.J. 1997. Schoenetum nigricantis, the Schoenus fen and flush vegetation of Ireland. Biology and Environment 97: 203–218. Dahl, E. 1956. Rondane. Mountain vegetation in south Norway and its relation to the environment. Skrifter utgitt av Det Norske Videnskaps-Akademi i Oslo, Mat.-Naturv. Klasse 3: 1–374. de Molenaar, J.G. 1976. Vegetation of the Angmagssalik District, Southeast Greenland. II. Herb and snow-bed vegetation. Meddelelser om Grønland 198/2: 1–266. Dierssen, K. & Dierssen, B. 2005. Studies on the vegetation of fens, springs and snow fields in West Greenland. Phytocoenologia 35: 849–85. Dierssen, K. 1982. Die wichtigsten Pflanzengesellschaften der Moore NW-Europas. Conservatoire et Jardin botaniques, Genève. Du Rietz, G.E. & Nannfeldt, J.A. 1925. Ryggmossen und Stigsbo Rödmosse, die letzten lebenden Hochmoore der Gegend von Upsala. Svenska Växtsociologiska Sällskapets Handlingar 3: 1–23. Elveland, J. 1976. Myrar pç Storön vid norrbottenskusten *Coastal mires on the Storön peninsula, Norrbotten, N Sweden+. Wahlenbergia 3: 1–274. [In Swedish.] Eurola, S. 1962. Über die regionale Einteilung der Südfinnischen Moore. Annales Botanici Societatis Zoologicæ Botanicæ Fennicæ 'Vanamo' 33(2): 1–243. Eurola, S. 1971. The middle arctic mire vegetation in Spitsbergen. Acta Agralia Fennici 123: 87–107. Fedotov, Yu.P. 1999. Bolota zapovednika «Bryanskiy les» i Nerusso-Desnyanskogo Poles'ya (flora i rastitel'nost') [Wetlands of the «Bryanskiy les» nature reserve and Nerusso-Desnyanskogo Poles'ya (flora and vegetation)]. Bryansk. [In Russian.] Fijałkowski, D. 1959. Szata roślinna jezior Łęczyosko-Włodawskich i przylegających do nich torfowisk *Plant Associations of Lakes Situated between Łęczna and Włodawa and of Peat-bogs Adjacent to These Lakes]. Annales Universitatis Mariae Curie-Skłodowska, Sectio C. 14: 131–206. [In Polish.] Fijałkowski, D. 1965. Zbiorowiska wodno-torfowiskowe rezerwatu Świerszczów koło Włodawy *Aquatic-Peat Communities of the Świerszczów Reserve near Włodawa+. Annales Universitatis Mariae Curie-Skłodowska, Sectio C. 20: 179–194. [In Polish.]

Foster, D.R. & Fritz, S.C. 1987. Mire Development, Pool Formation and Landscape Processes on Patterned Fens in Dalarna, Central Sweden. Journal of Ecology 75: 409–437. Fredskild, B. 1961. Floristic and ecological studies near Jakobshavn, West Greenland. Meddelelser om Grønland 163(4): 1– 82. Galanina, O.V. 2004. Rastitel'nost' sfagnovykh bolot i yeye kartografirovaniye na yugo-zapade tayezhnoy oblasti [Sphagnum-mire vegetation and its mapping in the south-west part of the taiga region]. MSc thesis, Botanicheskiy institut im. B.L. Komarova, Sankt-Peterburg. [In Russian.] Gargano, D. & Passalacqua, N.G. 2007. Bogs and Mires in Mediterranean Areas: the Vegetation of the Marshlands of the Lacìna Plain (Calabria, S. Italy). Phyton 47: 161–189. Gerdol, R. & Tomaselli, M. 1997. Vegetation of wetlands in the Dolomites. Dissertationes Botanicæ 281: 1–197. Gerdol, R. & Tomaselli, M. 1987. Mire vegetation in the Apuanian Alps (Italy). Folia Geobotanica Phytotaxonomica 22: 25– 33. Goncharova, N.N. 2007. Flora i Rastitel'nost' bolot yugo-zapada respubliki Komi [Flora and vegetation of mires in the southwestern part part of the Republic of Komi]. MSc thesis, Rossiyskaya akademiya nauk, Ural'skoye otdeleniye, Komi nauchnyy tsentr, Institut biologii, Petrozavodsk. [In Russian.] Hadač, E. 1969. Mire Communities of Reykjanes Peninsula, SW. Iceland (Plant communities of Reykjanes Peninsula, Part I.). Folia geobotanica et phytotaxonomica 4: 1–21. Hadač, E. 1985. Plant Communities of the Kaldidalur Area, WSW Iceland. Part 1. Syntaxonomy. Folia geobotanica et phytotaxonomica 20: 113–175. Hadač, E. 1989. Notes on Plant Communities of Spitsbergen. Folia geobotanica et phytotaxonomica 24: 131–169. Hallberg, H.P. 1971. Vegetation auf den Schalenablagerungen in Bohuslän, Schweden. Acta Phytogeographica Suecica 56: 1–136. Heikkilä, R. & Lindholm, T. 1988. Distribution and ecology of Sphagnum molle in Finland. Annales botanici Fennici 25: 11–19. Heikkilä, R., Lindholm, T., Kuznetsov, O., Aapala, K., Antipin, V., Djatshkova, T. & Shevelin, P. 2001. Complexes, vegetation, flora and dynamics of Kauhaneva mire system, western Finland. Finnish environment institude, Helsinki. Herbichowa, M. 1979. Roślinnośd atlantyckich torfowisk pobrzeża Kaszubskiego *The vegetation of the atlantic bogs on the Cashubian sea-coast]. Acta biologica 5: 1–52. [In Polish.] Hrsak, V. 1996. Vegetation succession at acidic fen near Dubravica in the Hrvatsko zagorje region. Natura Croatica 5: 1–10. Ivchenko, T. 2012. Redkiye bolotnyye soobshchestva s Schoenus ferrugineus na yuzhnom Urale (Chelyabinskaya oblast') [Rare mire communities with Schoenus ferrugineus in the southern Urals (Chelyabinsk region)]. Botanicheskii zhurnal 97: 79–86. [In Russian.] Ivchenko, T. 2013. Rastitel'nost' bolot Il'menskogo gosudarstvennogo zapovednika (Yuzhnyy Ural) [Mire vegetation of the Il’menski State Nature Reserve, the Southern Urals+. Rastitel'nost' Rossii 22: 38–62. [In Russian.] Jasioska, A.K., Iakushenko, D.M., Sobierajska, K., Tretiak, P.R. & Iszkuło, G. 2009. Pinus uliginosa G.E. Neumann ex Wimm., a new taxon for the Ukrainian flora. Ukrayinsʹkyy botanichnyy zhurnal 66(5): 640–646. Kalela, A. 1939. Über Wiesen und wiesenartige Pflanzengesellschaften auf der Fischerhalbinsel in Petsamo Lappland. Helsinki. Kalliola, R. 1939. Pflanzensoziologische Untersuchungen in der alpinen Stufe Finnish-Lapplands. Annales Botanici Societatis Zoologicæ Botanicæ Fennicæ 'Vanamo' 13/2: 1–321. Katz, N.I. 1929. Zur Kenntnis des Niedermoore im Norden den Moskauer Gouvernements. Repertorium specierum novarum regni vegetabilis. Beihefte. 56: 1–79. Klokk, T. 1982. Mire and forest vegetation from Klæbu, central Norway. Gunneria 40: 1–71. Koczur, A. 2014. Charakterystyka roślinności młak miasta Krakowa (Polska Południowa) *Spring fen vegetation in Kraków city (Southern Poland)]. Fragmenta Floristica et Geobotanica Polonica 21: 91–103. [In Polish.] Koroleva, N.E. 2001. Sintaksonomicheskiy obzor bolot tundrovogo poyasa Khibinskikh gor (Murmanskaya oblast) *Syntaxonomic survey of tundra belt mires of Khibiny mountains (Murmansk region)+. Rastitel’nost’ Rossii 2: 49–57. [In Russian.] Koroleva, N.E. 2006. Bezlesnyye rastitel'nyye soobshchestva poberezh'ya vostochnogo Murmana (Kol'skiy poluostrov, Rossiya) *Treeless plant communities of the East Murman shore (Kola peninsula, Russia)+. Rastitel’nost’ Rossii 9: 20–42. [In Russian.] Korotkov, K.O., Morozov, N.S., Morozova, O.V. & Alexeev, Yu.E. 1986. Cladium mariscus (Cyperaceae) na Valdaye (Novgorodskaya oblast') [Cladium mariscus (Cyperaceae) in Valday upland (Novgorod region)]. Botanicheskii zhurnal 71/10: 1341–1347. [In Russian.] Lakuśid, R. 1968. Planinska vegetacija jugoistočnih Dinarida *The mountain vegetation of southeastern Dinarids]. Glasnik Republičkog zavoda za zaštitu prirode i prirodnjačke zbirke 1: 9-75. [In Serbian.] Lakuśid, R., Grgid, P., Kutleša, L., Muratspahid, D., Redžid, S. & Omerovid, S. 1991. Struktura i dinamika fitocenoza u ekosistemima tresetišta na planinama Bosne [Structure and dynamics of phytocoenoses in peatland ecosystems in the mountains of Bosnia]. Bilten društva elokoga Bosne i Hercegovine, Ser. A 7: 35–84. [In Bosnian.] Layvin'sh, M. & Svars, D. 1993. Rastitel'nyye soobshchestva s Schoenus ferrugineus L. na territorii Latvii: vidovoy sostav ekologiya i klassifikatsiya [Plant communities with Schoenus ferrugineus L. in Latvia: species composition ecology and classification]. In: Boch, M.C. (ed.) Voprosy klassifikatsii bolotnoy rastitel'nosti, pp. 104–112. Nauka, Sankt-Peterburg. [In Russian.]

Lounamaa, J. 1961. Untersuchungen über die eutrophen Moore des Tulemajärvi-Gebietes im südwestlichen Ostkarelien, KASSR. Annales Botanici Societatis Zoologicæ Botanicæ Fennicæ 'Vanamo' 32: 1–63. Möller, I. 1999. Studien zur Vegetation Nordwestspitzbergens. Doctotal Doctoral thesis, Universität Hamburg, Hamburg. Neschatayev, V.Yu. 1986. Izmeneniye rastitel'nosti travyano-sfagnovykh sosnyakov pod bliyaniyem osusheniya [The changes of the vegetation in grass bog moss pine forest under the influence of drainage]. )]. Botanicheskii zhurnal 71/4: 429– 440. [In Russian.] Nordhagen, R. 1928. Die Vegetation und Flora des Sylenegebietes. Skrifter utgitt av Det Norske Videnskaps Akademi i Oslo, Mat.-Naturv. Klasse. 1927(1): 1–612. Nordhagen, R. 1943. Sikilsdalen og Norges fjellbeiter. En plantesosiologisk monografi [Sikilsdalen and Norwegian mountain pastures: a plant sociological monograph]. Bergens Museums Skrifter 22: 1–607. Osvald, H. 1923. Die Vegetation des Hochmoores Komosse. Svenska Växtsociologiska Sällskapets, Handlingar 1: 1–436. Osvald, H. 1925. Zur Vegetation der ozeanischen Hochmoore in Norwegen. Svenska Växtsociologiska Sällskapets, Handlingar 7: 1–106. Paasio, I. 1933. Über die Vegetation der Hochmoore Finnlands. Acta Forestalia Fennica 39(3): 1–190. Paasio, I. 1939. Zur Vegetation der eigentlichen Hochmoore Estlands. Annales Botanici Societatis Zoologicæ Botanicæ Fennicæ 'Vanamo' 11(2): 1–114. Pałczyoski, A. 1975. Bagna Jadwieskie (pradolina Biebrzy) *The Jadwieskie swamps (Biebrza valley)+. Roczniki nauk rolniczych, Ser. D – Monografie 145: 1–232. Passarge, G. & Passarge H. 1972. Beobachtungen über Wald- und Gebüschgesellschaften im Raum Leningrad. Feddes Repertorium 82: 629–657. Persson, Å. 1961: Mire and spring vegetation in an area north of lake Tornetrask, Torne Lappmark, Sweden. I. Description of the vegetation. Opera Botanica 6(1): 1–187. Petronici, C., Mazzola, P. & Raimondo, F.M. 1978. Nota introduttiva allo studio degli ambienti idromorfi delle Madonie [Introduction to the Madonie marshy areas study]. Naturalista Siciliano, s. IV 2: 11–24. [In Italian.] Pettersson, B. 1958. Dynamik och konstans i Gotlands flora och vegetation [Dynamics and constancy in Gotland flora and vegetation]. Acta Phytogeographica Suecica 40: 1–288. [In Swedish.] Philippi, G. 1973. Moosflora und Moosvegetation des Freeman-Sund-Gebietes (Südost-Spitzbergen). Franz Steiner Verlag GMBH, Wiesbaden. Pop, I., Cristea, V., Hodişan, I. & Raţiu, O. 1986. Vegetaţia tinoavelor de la Blæjoaia şi Dorna *Mire vegetation of Blæjoaia and Dorna+. Contribuţii botanice 1986: 123–129. Pop, I., Hodişan, I. & Cristea, V. 1987. La végétation de certaines turbières de la Vallée Izbuc (Départment de Cluj). Contribuţii botanice 1987: 111–120. Raimondo, F.M., Scialabba, A. & Dia, M.G. 1980. Note briogeografiche. III. Distribuzione in Italia di Aulacomnium palustre (Hedw.) Schwaegr. ed ecologia della specie nelle stazioni siciliane [Bryogeographical notes. III. Distribution of Aulacomnium palustre (Hedw.) Schwaegr. in Italy and its ecology in the Sicilian localities]. Naturalista Siciliano, s. IV 4: 79–99. [In Italian.] Randjelovid, V.N. & Zlatkovid, B.K. 2010. Flora i vegetacija Vlasinske visoravni [Flora and vegetation of the Vlasina plateau]. Univerzitet u Niou, Niš. [In Serbian.] Randjelovid, V.N., Zlatkovid, B.K & Amidžid, L. 1998. Flora and Vegetation of High-mounatain Peat-bogs of Mt. Šar-planina. The University thought 5: 91–98. Raţiu, O. 1965. Contribuţii la cunoaşterea vegetaţiei din bazinul Stîna de Vale *Contributions to the knowledge of the vegetation from the Stina de Vale basin+. Contribuţii botanice 1965: 151-175. [In Romanian.] Redžid, S., Trakid, S. & Barudanovid, S. 2013. Patterns of vegetation diversity of grasslands and pastures – Crvanj Mt. (Herzegovina, Western Balkan). Scientific Research and Essays 8(39): 1944–1965. Ruuhijärvi, R. 1960. Über die regionale Einteilung der Nordfinnischen Moore. Annales Botanici Societatis Zoologicæ Botanicæ Fennicæ 'Vanamo' 31(1): 1–360. Sambuk, S.G. 1987. Oligotrofnyye sfagnovyye sosnovyye lesa na severo-zapade yevropeyskoy chasti SSSR [Oligotrophic Sphagnum pine forest of the north-wets of the European part of the USSR]. Botanicheskii zhurnal 72/11: 1523–1532. [In Russian.] Sieg, B., Drees, B. & Daniëls, F.J.A. 2006. Vegetation and altitudinal zonation in continental West Greenland. Meddelelser om Grønland, Bioscience 57: 1–93. Skogen, A. 1973. Phytogeographical and ecological studies on Carex paniculata L. in Norway. Årbok for universitetet i Bergen, Mat.-Naturv. serie 3: 1–12. Skogen, A. 1974. Autoecological studies on Hammarbya paludosa at Hitra, Central Norway. Norwegian journal of botany 21: 53–68. Sonesson, M. & Kvillner, E. 1980. Plant communities of the Stordalen mire – a comparison between numerical and non- numerical classification methods. Ecological Bulletin 30: 113–125. Söyrinki, N., Salmela, R. & Suvanto, J. 1977. Oulangan kansallispuiston metsä- ja suokasvillisuus [The forest and mire vegetation of the Oulanka National Park, Northern Finland]. Acta Forestalia Fennica 154: 1–150. [In Finnish.] Thumark, S. 1931. Der See Fiolen und seine Vegetation. Acta Phytogeographica Suecica 2: 1–198. Vorobyov, Ye.O., Balashov, L.S. & Solomakha, V.A. 1997. Syntaksonomiya roslynnosti Polisʹkoho pryrodnoho zapovidnyka [The syntaxonomy of the vegetation of the Polesie natural reserve]. Ukr. Phytosociol. Col. Ser. B. 1: 1–128. [In Ukrainian.]

Vorren, K.D., Eurola, S. & Tveraabak, U. 1999. The lowland terrestrial mire vegetation about 69°N lat. in northern Norway. Tromura (Tromsø Museums Rapportserie) Naturvitenskap 84: 1–105. Warén, H. 1924. Untersuchungen über die botanische Entwicklung der Moore mit Berücksichtigung der chemischen Zusammensetzung der Torfes. Wissenschaftliche Veröffentlichungen des Finnischen Moorkulturvereins 5: 1–95. Warén, H. 1926. Untersuchungen über Sphagnumreiche Pflanzengesellschaften der Moore Finnlands unter Berücksichtigung der soziologischen Bedeutung der einzelnen Arten. Acta societatis pro fauna et flora Fennica 55: 1– 133.

Paper 3 (manuscript) Peterka T., Syrovátka V., Dítě D., Hájková P., Hrubanová M., Jiroušek M., Plesková Z., Singh P., Šímová A., Šmerdová E. & Hájek M. (unpubl.): Is variable plot size a serious constraint in broad-scale vegetation studies? A case study on fens.

Is variable plot size a serious constraint in broad-scale vegetation studies? A case study on fens

Tomáš Peterka1, Vít Syrovátka1, Daniel Dítě2, Petra Hájková1,3, Monika Hrubanová1, Martin Jiroušek1,4, Zuzana Plesková1, Patrícia Singh1, Anna Šímová1, Eva Šmerdová1, Michal Hájek1

1 Department of Botany and Zoology, Faculty of Science, Masaryk University, Brno, Czech Republic 2 Institute of Botany, Plant Science and Biodiversity Center, Slovak Academy of Sciences, Bratislava, Slovakia 3 Laboratory of Paleoecology, Institute of Botany, The Czech Academy of Sciences, Brno, Czech Republic 4 Department of Plant Biology, Faculty of AgriSciences, Mendel University in Brno, Brno, Czech Republic

Abstract Question: Filtering of vegetation plot records according to sampling size is an essential methodical step in vegetation studies. In mires, variation in traditionally used plot sizes seems to limit continental-scale syntheses. Which plot sizes provide mutually consistent results regarding both the number of habitat specialists (i.e. diagnostic/indicator species) and the capturing main compositional gradients? Location: Scandinavia, central Europe. Methods: The dataset of fen vegetation plot records was compiled using large databases and categorised into four distinct habitats. For each habitat, species-area curves of specialists and generalists were fitted using GAM. In the subset of 72 own plot size series (0.07, 0.25, 1, 4, 16 m2) we applied, separately for each plot size, Non-Metric Multidimensional Scaling (NMDS) and compared the resulting patterns with Procrustes analysis. We further used K-means clustering for a posteriori assessment whether plot size affects classifications. Results: Consistently across different fen habitats, the species-area curves of specialists increased steeply up to the plot size of 1 m2, while increased negligibly or approached an asymptote in the plot size range of 1–25 m2. Contrary, the species-area curves of generalists displayed mostly linear to linear-exponential trends. NMDS ordinations of medium (1 and 4 m2) and large plots (16 m2) were the most congruent, while the patterns captured in the ordination of the smallest plots (0.07 m2) differed most from the others. Clusters produced by K-means classification reflected different vegetation types or regions rather than different plot size. Conclusions: In fens, plot sizes of at least 1 m2 describe sufficiently the broad-scale pattern in specialists’ diversity as well as the main environmental gradients. The range of plot sizes of 1–25 m2 may be safely merged in broad-scale analyses of fen vegetation without introducing substantial bias, at least when compared with other possible uncertainty sources.

Keywords: fens; phytosociology; plant specialists; plot size; scale; species-area relationship; vegetation classification; vegetation plot; vegetation survey; wetlands

Introduction The increasing recent need for effective protection of natural habitats (Janssen et al., 2016) requests for harmonised vegetation classifications on the pan-European scale that could serve as a solid basis for habitat typologies. In the last decades, this necessity together with coincident development of vegetation databases (Chytrý et al., 2016) has led to several broad-scale syntheses of vegetation data and resultant attempts to create supra-national classifications of particular vegetation types (e.g. Eliáš et al., 2013; Douda et al., 2016; Peterka et al., 2017; Rodríguez-Rojo et al., 2017; Willner et al., 2017a, b; Marcenò et al., 2018). All these studies were based on the analysis of a large set of primary vegetation data (vegetation plots). From a methodological point of view, the creation of broad-scale vegetation surveys and syntheses is connected with a series of steps, elementary questions and decisions, such as data selection, filtering and resampling (Knollová, Chytrý, Tichý, & Hájek, 2005; Lengyel, Chytrý, & Tichý, 2011), choice of classification criteria (De Cáceres et al., 2015), decision between supervised and unsupervised classification approach (Tichý, Chytrý, & Botta-Dukát, 2014), choosing of proper classification algorithms or data transformation (Lengyel & Podani, 2015) etc. De Cáceres et al. (2015) stimulated an ongoing debate, trying to establish a common framework for plot-based vegetation classifications. Although the broad-scale vegetation surveys do not generally differ in basic principles, the absolute methodical consensus among vegetation scientists still does not exist. One of the fundamental, but unresolved question is whether to use the plots of different sizes jointly in vegetation surveys or not. Classical debate related to the matter of plot sizes concentrated predominantly on searching for “minimal”, “characteristic” or “representative” area, i.e. the smallest plot size representing the structure and composition of a plant community (Moravec, 1973; Dietvorst, van der Maarel, & van der Putten, 1982; Barkman, 1989), as well as distinguishing between plant communities and synuziae (Gillet & Gallandat, 1996). In the last two decades, this issue was resurrected in the context of broad- scale vegetation surveys and associated projects (e.g. Chytrý & Otýpková, 2003; Otýpková & Chytrý, 2006; Dengler, Löbel, & Dolnik, 2009). The question whether to use plots of different sizes is connected with one of the general macroecological patterns described by the species-area curves, i.e. the larger plots inevitably harbouring more species than the smaller ones (e.g. Arrhenius, 1921; Lomolino, 2000; Storch, 2016). In the field of vegetation surveys, this fact results in the premise that the joint use of different-sized-plots may affect the results of classification (Fekete & Szöcs, 1974; Podani, 1984; Dengler et al., 2009). The mainland species-area relationship (i.e. increasing number of species with increasing habitat or plot area) usually has a character of a logarithmic curve with an upper asymptote determined by the species pool size, and the slope determined by habitat and trophic level (Drakare, Lennon, & Hillebrand, 2006). In addition, habitat specialists may show different species-area relationship as compared to the habitat generalists (Matthews, Cottee-Jones, & Whittaker, 2014), simply because habitat specialists and generalists (the

89 latter group mostly derived from the surrounding matrix) are ecologically associated with different habitats and hence are assembled from the different species pools (Horsáková, Hájek, Hájková, Dítě, & Horsák, 2018). In phytosociology, habitat specialists act as the so- called diagnostic species whose representation is a crucial parameter in vegetation classification (Westhoff & van der Maarel, 1973; Chytrý, Tichý, Holt, & Botta-Dukát, 2002). To our best knowledge, surprisingly no study discussing the effect of the plot size on vegetation classification distinguished between the species-area relationships of habitat specialists (i.e. diagnostic species) and matrix-derived species (i.e. accidental species). Although general species-area curves usually increase over the entire range of plot sizes used in phytosociology, the slope of the species-area curve for specialists may be gentler because of limited species pool. This effect may appear in the low-productive habitats naturally occupying small areas such as springs and fens – compare their species pool sizes with those of grasslands or forests (Sádlo, Chytrý, & Pyšek, 2007). In addition, in the low- productive and stressed habitats such as fens, facilitative inter-specific interactions override the competitive ones (the Stress Gradient Hypothesis: Bertness & Callaway, 1994; Michalet, Le Bagousse–Pinguet, Maalouf, & Lortie, 2014), leading to increased species density on small plots. We hence hypothesise that in such habitats even quite small plots will contain most of the habitat specialists (i.e. diagnostic species in phytosociology) and the species-area curve will be gentler, while the species-area curve for matrix-derived species (generalists), a more numerous group of species in these habitats (Horsáková et al., 2018), will be steeper. The general agreement is that the clear dependence of community delimitation on the plot size is obvious when different spatial and structural organization levels are captured at different measuring scales considered, e.g. forest vegetation sampled on “large” plots may harbour patches of non-forest communities, which might be recorded on plots of much smaller spatial scale (Chytrý & Otýpková, 2003). However, much less attention has been paid to the question whether there is an effect of plot size on classification within one magnitude of plot sizes. Chytrý & Otýpková (2003) suggested that plots falling into a certain range of the most frequent sizes might be analysed together in a single dataset after excluding outliers, though recommended standards for plot sizes for individual vegetation classes. Later, Dengler et al. (2009) recommended a higher level of criticism. The main guidance of their seminal paper can be reproduced, with a certain simplification, as follows: it is necessary (i) to analyse plots within relatively narrow size ranges (even-sized plots) when using old data, (ii) to apply uniform (standardised) plot size for all vegetation types that will by classified jointly in future surveys and (iii) to establish standard plot sizes for classification. The potential effect of plot sizes on vegetation classification was also examined by Otýpková & Chytrý (2006) who detected that the variation in plot size influences ordination patterns in homogeneous datasets but is of lower importance in heterogeneous ones. Forbes & Sumina (1999) found that different plot sizes (1, 25 m2) had almost no effect on the ordination pattern. Lengyel & Podani (2015), however, found a significant effect of mean plot size on classifications, though without an interpretable biological pattern.

Due to the different sampling tradition in different regions, the size of vegetation plots stored in databases varies among European countries as exemplified especially for fens (Peterka, Jiroušek, Hájek, & Jiménez-Alfaro, 2015). When strictly implemented, the standardising of plot sizes or using plots of equal sizes exclusively as suggested by Dengler et al. (2009) might, however, cause loss or even absence of available data from several European regions. For example, plots of 1 m2 were most often used in northern Europe (Pakarinen, 1995), plots of 4 m2 prevail in Great Britain (Rodwell, 1991) and plots of 16–25 m2 represent the standard in central Europe (Chytrý & Otýpková, 2003). Another point is that some plant communities, which are important for nature conservation, might occasionally occupy naturally smaller areas than a common plot standard is, though have distinct borders, e.g. spring fens, dystrophic (bog) hollows in mountains of temperate Europe or some types of aquatic vegetation. To record the species composition of such a plant community and to maintain the assumptions of ecological uniformity at the same time, it is then necessary to apply smaller plot size. Furthermore, the problem of different plot sizes acquires broader significance when considering its overlap to other ecological topics. Besides studies on vegetation classification, the choice of appropriate plot size is essential also in designing experiments in the field of restoration ecology, community management and functional ecology. No less important matter is an uncertain level of comparability of results between the ecological studies conducted using different sizes of study plots (e.g. Wiens, 1989; Güsewell, Buttler, & Klötzli, 1998). In field ecological studies, the smaller plots are usually sampled to obtain more replications. The question arises whether even small plots adequately represent given plant community and capture its diversity and proportion of habitat specialists. The aim of this paper is (i) to assess which extent of plot sizes is applicable in broad- scale vegetation classifications and ecological studies in fens, (ii) to compare patterns in the increasing species richness with increasing plot size based on large set of independently sampled data, separately for both plant specialists and generalists, (iii) to assess whether different plot sizes have a crucial effect on the result of unsupervised vegetation classifications and (iv) to suggest the mutually comparable range of plot sizes for designing ecological experiments in fens.

Methods Data collection and filtering For this study, we compiled vegetation plots (phytosociological relevés) from two independent regions with a large diversity of mire habitats and long tradition of mire ecology research: (i) Scandinavia (Norway, Sweden) and (ii) central Europe (Czech Republic and Slovakia plus the neighbouring part of southern Poland). The plots were stored in the European Mire Vegetation Database (EU-00-022; Peterka et al., 2015), the Czech National Vegetation Database (EU-CZ-001; Chytrý & Rafajová, 2003) and the Slovak Vegetation Database (EU-SK-001; Šibík, 2012), which have been recently integrated within the European

Vegetation Archive (EVA; Chytrý et al., 2016). These datasets were accompanied by other plots sampled by authors of this study and stored in private databases. In the first step, all the plots originally classified by the authors as “mire vegetation” were included. Vegetation of bogs (ombrotrophic mires, Oxycocco-Sphagnetea class) was excluded following the formal definition of that class (Jiroušek et al., in. prep.), plots transient to wet meadows and marshes were excluded following Peterka et al. (2017). The resulting dataset thus comprised solely plots of fens (minerotrophic mires) that belong to the phytosociological class of the Scheuchzerio palustris-Caricetea fuscae class (Mucina et al., 2016; Joosten, Tanneberger, & Moen, 2017). Plots of size from the interval 0.01−25 m2 were selected. Larger plots were omitted in accordance with Chytrý (2001) who found that plots of sizes exceeding ~ 25–30 m2 stored in vegetation databases contain fewer species than smaller ones. There is no ecological explanation, though sampling artefact of phytosociologists (less careful research at a larger scale or tendency to enlarge a plot when sampling species poor vegetation). In the following text, plots falling into the interval of 0–0.9 m2 are denoted as “small”, plots of 1–9.9 m2 as “medium” and 10–25 m2 as “large”.

Nomenclature The nomenclature was harmoniszed following Euro+Med (2006–2018) for vascular plants, Hill et al. (2006) for mosses and Frey, Frahm, Fischer, & Lobin (2006) for liverworts. Closely related taxa or taxa of problematic and ambiguous taxonomical status were merged (Appendix S1). Taxa determined only at the genus level were omitted as well as lichens, algae, and fungi. Nomenclature of vegetation units follows Mucina et al. (2016); in other cases, the author citation is quoted with the first reference.

Data analysis To explore which plot sizes harbour a comparable number of specialists, i.e. assumed diagnostic species fundamental for vegetation classification (for reasoning see below), the species-area curves were constructed. Since the vegetation of the Scheuchzerio palustris- Caricetea fuscae class differs considerably in species richness and species pools among individual fen types and regions (Malmer, 1986; Hájková, Hájek, Apostolova, Zelený, & Dítě, 2008; Horsáková et al., 2018), the relationships between species richness and plot size were studied on distinct ecological types, separately for Scandinavia and central Europe. We defined four following ecological types: (i) calcareous and extremely rich fens with calcicole vascular plants and without Sphagnum species, (ii) rich fens with calcium-tolerant Sphagnum species, (iii) poor fens and (iv) dystrophic hollows. For further description of these ecological types see Hájek, Horsák, Hájková, & Dítě (2006), Joosten et al. (2017), Peterka et al. (2017) and Appendix S2. The definitions of the types follow logical formulas proposed in Peterka et al. (2017) with a certain simplification and aiming at functional species groups (according to Landucci, Tichý, Šumberová, & Chytrý, 2015). Application of functional groups gives species-

92 poor plots and species-rich plots equal chance to fulfil the assumptions of the definitions. At the same time, the use of functional groups avoids the effect of circularity, which would happen if plots are selected according to the presence of a defined quantity of species and after that analysed in terms of species richness. Due to a considerably small number of available plots of calcareous and extremely rich fens from Scandinavia, only seven data subsets were finally analysed (Table 1). Within each subset, the following intervals of plot sizes were established: 0.01–0.6, 0.7–1.5, 1.51–4, 4.1–12 and 12.1–25 m2. Within each of these intervals, the geographical stratification was performed to avoid the effect of oversampling. Maximum of 5 plots were selected from each grid cell of 1.25 minutes of longitude × 0.75 minutes of latitude (approximately 1.5 × 1.4 km). Plants species were afterwards divided into two groups: specialists and generalists (matrix-derived species, see also Horsáková et al., 2018). Habitat specialists were selected according to the list of diagnostic species of Scheuchzerio palustris-Caricetea fuscae (Mucina et al., 2016) and supplemented by few other species according to authors’ field knowledge (Appendix S3). The group of specialists includes pure fen specialists (e.g. Pseudocalliergon trifarium, Sphagnum majus, Trichophorum alpinum) as well as species that are also spanning into other vegetation types but having a clear optimum in fen vegetation or occurring casually in pristine fens (e.g. Carex nigra, C. rostrata, Utricularia minor agg.). All other, i.e. mostly matrix-derived, species are regarded as habitat generalists. The number of specialists and generalists along the increasing plot size was analysed using generalised additive models (GAM) with the Poisson distribution. Besides separate curves for specialists and generalists, one joint curve combining both groups was created. The chi-square statistics (χ2) were calculated to test whether two separate curves represent a better model as compared to one joint curve. Further, the Akaike’s information criterion (AIC) was computed to estimate which model is better. This criterion attempts to measure model parsimony using the number of model degrees of freedom (relatively lower AIC value signifies relatively higher parsimony). Further, we compared five ordinations based on plots of distinct sizes to assess the influence of different plot sizes on ordination patterns. For this purpose, we sampled series of plots, including small (0.07, 0.25 m2), medium (1, 4 m2) and large plot sizes (16 m2) at 72 localities in central Europe (the Western Carpathians and the Bohemian Massif) using non- nested design (Podani, 1984). The smallest plots (0.07 m2) were circular, the other plots were quadrate. Plots of each size (with square-root transformed species covers) were subjected separately to two-dimensional non-metric multidimensional scaling (NMDS) using Bray-Curtis distances. In the next step, these ordinations were subjected to the Procrustes analysis. In this procedure, one ordination is rotated to maximum similarity with another one, and correlation-like statistic (Procrustes R2) and the sum of squared differences (SS) of all objects are calculated. This procedure was applied to all pairs of ordinations and SS was each time recorded as a measure of the ordinations’ difference. To visualise the dissimilarity among ordinations of plots of different sizes, the SS values were arranged into the

93 dissimilarity matrix, which was finally projected into PCoA. Since the first two PCoA axes captured nearly 95% of the variance in ordinations’ dissimilarity, the ordination space was reduced into two dimensions. To make a posteriori assessment whether the plot size affects the results of classification, we performed the cluster analysis using the entire fen dataset. We work with the hypothesis that the different plot size substantially affects the result of classification, if clusters (groups of plots) of homogeneous plots sizes or with very narrow size extent arise. To avoid the effect of oversampling, the dataset had been again geographically stratified using a maximum of ten plots from each grid cell of 1.25 minutes of longitude × 0.75 minutes of latitude. The non-hierarchical K-means cluster analysis with 50 resulting clusters was applied. The species cover values were square-root transformed, and the classification algorithm was repeated 25 times. Diagnostic species of resulting clusters were calculated using the phi coefficient (Chytrý et al., 2002) and their significance was tested using Fisher’s exact test (P <0.01). Species with fidelity to a particular cluster of phi >0.3 were considered as diagnostic. For the Procrustes analysis and the analysis of species richness with increasing plot size, packages ggplot2 (Wickham, 2009), mgcv (Wood, 2006) and vegan (Oksanen et al., 2017) in the R 3.4 software were used. Other procedures including cluster analysis were done with the help of the JUICE 7.0 programme (Tichý, 2002).

Results Species-area curves displayed shifts in plant species richness along increasing plot size across the two study regions and four vegetation types (Fig. 1). At the semi-log scale, the counts of plant specialists increased mainly at the interval between the smallest plots up to plot size of ~ 1–2 m2. At the interval from ca 1 m2 to 25 m2, the curves increased rather negligible or even stagnated after exhibiting theoretical asymptote. Virtually, no difference in species richness was further apparent among plots of 4 m2 and 16–25 m2 across all subsets. This pattern concerned curves based on plots from two studied geographic regions and covering all vegetation types, although the shape of the curves differed among individual subsets. The greatest increase in species richness was detected for both datasets of rich fens, where average species richness increased of about three species between plots of 1 m2 and 16–25 m2. Contrary to specialists, the curves of generalists displayed mostly linear to linear- exponential trend along the analysed range of plot sizes. Except for communities of dystrophic hollows in Scandinavia, the chi-square statistics proved that the species-area curves created separately for specialists and generalists differed significantly from one joint curve combining both groups (Table 2). According to the AIC, separate curves for specialists and generalists represented a better model than one joint curve. Procrustes correlation proved relatively high similarities among all pairs of ordinations (r = 0.81–0.96, P <0.001). The highest congruences (Table 3, Fig. 2) were detected between the ordinations of medium (1 and 4 m2) and large plots (16 m2), which

94 were closely grouped in the PCoA diagram. On the other hand, the greatest dissimilarity was found for the smallest plots (0.07 m2), being shifted towards the opposite end of ordination space contrary to the ordinations of all other plot sizes. The ordination of plots of 0.25 m2 appeared to be distinct from the ordinations of medium and large plots. K-means clustering at the level of 50 resulting clusters produced groups (clusters) of vegetation plots well-differentiated by diagnostic species and having clear ecological and syntaxonomical interpretation; some clusters reflected also included biogeographic regions (Appendices S4, S5). Plots of small (<1 m2), medium (1–9.9 m2) and large (10–25 m2) sizes were distributed across all clusters (Fig. 3), although there were certain differences in sizes’ frequencies among clusters. No cluster harbouring exclusively one category of plot sizes arose.

Discussion

The effect of plot size on the species richness, ordinations, and unsupervised classification Vegetation scientists traditionally used species-area curves when searching for the smallest plot size comprising representative species combination of a plant community, so-called “minimal” area applicable for phytosociological research (Du Rietz, 1921; Moravec, 1973; Barkman, 1989; Toman, 1990). The concept of “minimal” area was later abandoned since there is no objective way how to delimit it (Dengler, 2003) and its definition involves circularity (Chytrý & Otýpková, 2003). According to our opinion, the species-area curves anyway provide information which plot sizes (areas) share the same or comparable species numbers. Our results suggest that the number of fen specialists increases mainly from the smallest plots up to approximately 1 m2 or slightly larger plots. At the interval from 1 to 25 m2, the rate of increase was lower or even negligible. Within this range, the number of fen specialists (i.e. diagnostic species) increases only in the order of units, and possible bias caused by different plot size is hence comparable with the bias introduced by different sampling effort of individual researchers (Lepš & Hadincová, 1992; Klimeš, Dančák, Hájek, Jongepierová, & Kučera, 2001; Vittoz & Guisan, 2007) or different phenology (Vymazalová, Tichý, & Axmanová, 2014). In this study, we follow the premise that the delimitation of main vegetation units within the Braun-Blanquet classification approach is generally based on plant specialists (i.e. diagnostic species; De Cáceres et al., 2015) having indicator significance and a high fidelity to a given system. On the other hand, generalists (matrix-derived species) play rather a secondary role and are considered merely in combination with specialists, especially in formalized classifications (Kočí, Chytrý, & Tichý, 2003). In the case of fen vegetation, generalists frequently mirror some level of transition to another vegetation types, successional changes or degradation (Hájek et al., 2006; Bergamini et al., 2009). If the assumption of the key role of habitat specialists for vegetation classification is accepted,

95 then plots of sizes within the range of 1–25 m2 might be possibly included in one analysis, since the specialists’ representation is rather similar within the entire interval of plot sizes. Contrary to the specialists, the curves of generalists increase continuously linearly or have even the linear-exponential trend along with increasing plot size. This pattern gives evidence that the increase in species richness along with increasing plot size is driven by non-identical mechanisms for both species groups. When the larger plot is sampled, we have a higher chance to capture individual generalist species within the plot. Generalists likely colonize fen “islands” from the surroundings (Whittaker, 1998) and their more frequent occurrence at larger plots hence might be driven by neutral processes or spatial-mass effect (Hettenbergerová & Hájek, 2011; Janišová, Michalcová, Bacaro, & Ghisla, 2014) rather than by site ecological conditions determining existence of a fen community. In general, this theory is supported by Öster, Cousins, & Eriksson (2007) who found that habitat diversity in the landscape increases the total species richness of grasslands, but is unimportant for specialists’ species richness. A large proportion of matrix-derived species at larger plots might also be one of the theoretical reasons for the hard-to-interpretable effect of plot size on classification detected by Lengyel & Podani (2015). The mutual position of individual ordinations in the PCoA indicates that the main species composition patterns are almost equally reflected by plots of 1, 4 and 16 m2, advocating similarity of species composition within particular plot series. Hence, we can conclude that plant communities in these plot sizes do not differ considerably from each other. The ordinations of small plots (0.07, 0.25) yielded different results, suggesting the small plots (<1 m2) deviate from medium and large ones, and their analysis insufficiently mirrors the main vegetation gradients. This result corroborates the finding of Forbes & Sumina (1999) that it is possible to combine plots of varying sizes (1 and 25 m2) sampled in tundra vegetation in multivariate analyses for classification purposes. On the other hand, this result is in partial disagreement with Otýpková & Chytrý (2006), who demonstrated that the ordination of plots of 1 m2 significantly differed from ordinations of larger plots. The discrepancy with our study might be related to different studied scale (1–49 m2) and different vegetation types having significantly higher species richness, i.e. meadows and dry grasslands, some of which come from the area where the world records in species richness per plot were obtained (Chytrý et al., 2015). This fact may suggest that the different plot size has low importance for ordination and classification of relatively species-poor vegetation (e.g. fens, tundra), whereas it is much more important for species-rich communities (e.g. meadows, dry grasslands). The resulting clusters of K-means classification reproduced well the individual vegetation units at the level of alliances and associations as they were recognised in relevant vegetation surveys (e.g. Nordhagen, 1943; Dahl, 1956; Rybníček, Balátová-Tuláčková, & Neuhäusl, 1984; Valachovič, 2001; Chytrý, 2011; Moen, Lyngstad, & Øien, 2012). Since all clusters are characterized by almost equal extent of plot sizes, we can admit that plot size plays minor importance for plot clustering, at least in the case of large and relatively

96 heterogeneous datasets we analysed. The more pronounced effect of plot size would have been proved if clusters of a narrow range of plot sizes had appeared. The ratio between particular plot sizes indeed differed across individual clusters, but this disproportion may be caused by uneven data availability, e.g. some vegetation types being sampled predominantly in the regions where either small or large plot sizes are traditionally used. For example, the arctic-alpine extremely rich fen community of the alliance Caricion atrofusco-saxatilis (see Appendix S5) has been traditionally sampled on somewhat smaller plot sizes (~ 1 m2), and larger plots are rare (Peterka et al., 2017). On the other hand, mesotrophic mires and mire meadows of the Caricetum nigrae Braun 1915 association (Caricion fuscae alliance), one of the most common fen community in central Europe (Chytrý, 2011), has been traditionally sampled using plots of 16–25 m2 and smaller ones are scarce.

Different plot size in broad-scale vegetation syntheses: the battle between methodological purisms and practical use The unquestionable and clear advantage of using plots of equal size or even-sized plots, as suggested by Dengler et al. (2009), is the greatest level of statistical and methodological purity. However, when only vegetation plots of very narrow size range would be selected for a broad-scale vegetation survey, many data from important regions would be omitted because of different sampling traditions (i.e., using of different plot sizes) in different regions. Furthermore, the inclusion of plots within the strictly given size range might favour plots from the regions where the target vegetation is not well developed. Fen vegetation used in this study might serve as a good theoretical example to illustrate this pattern: if the pan-European analysis and classification of fen communities would be performed exclusively using the plots of 16 m2 (i.e. common plot standard for grasslands in western and central Europe) or 10–25 m2 (presumable range of even sized plots), the resulting dataset would have been deprived of majority of plots from Fennoscandia, the Alps or Great Britain, where smaller plots are traditionally used (cf. Rodwell, 1991; Steiner, 1992; Pakarinen, 1995; Peterka et al., 2015). Thus, the resulting dataset would have been predominantly formed of depauperate plots from western and central Europe where majority of fens were disturbed, and consequently a wide array of sensitive specialists disappeared here in the second half of the 20th century (Kooijman, 2012; Hájek et al., 2015; Navrátilová, Hájek, Navrátil, Hájková, & Frazier, 2017; Rion, Gallandat, Gobat, & Vittoz, 2018). Therefore, an important number of plots out of these regions has been sampled at fen remnants lacking important diagnostic species that occur only in pristine fens. Except the problem of different plot sizes, vegetation scientists have to deal with a wide array of uncertainty sources when working on synthetic vegetation studies using thousands of plots stored in electronic databases. Besides different sampling efficiency of individual authors and seasonal variance of stands as mentioned above, the important problems concern for example degree of ecological uniformity within sampled plots,

97 different types of cover estimation and transformation (Jensen, 1970; Tichý, Chytrý, Hájek, Talbot, & Botta-Dukát, 2010), preferential versus random sampling (Botta-Dukát, Kovács- Láng, Rédei, Kertész, & Garadnai, 2007; Michalcová, Lvončík, Chytrý, & Hájek, 2011; Swacha, Botta-Dukát, Kącki, Pruchniewicz, & Żołnierz, 2017), oversampling of selected localities (Knollová et al., 2005), countries (Chytrý et al., 2016) or even “vegetation types” (Lengyel et al., 2011) as well as technical problems in data preparation such as inconsistency in taxonomy and plant nomenclature (Jansen & Dengler, 2010) and spelling mistakes in the databases (Wagner, 2016). All these factors may confound the classifications, similarly to using different plot sizes. Considering all other possible sources of bias, one could recommend using only the plots of the same size and phenology (i.e. the narrow range of date within a year), sampled by the same researcher, or the same group of researchers to guarantee similar field identification skills and similar taxonomic conceptions applied. However, using such strict rules, the number of analysed plots would be extremely small. Except for the purely technical issues such as non-unified nomenclature and spelling mistakes, the bias can be reduced by analysing the more representative, i.e. larger and more variable, dataset (Roleček, Chytrý, Hájek, Lvončík, & Tichý, 2007) even at the cost of different size of included plots. Here we demonstrated that the argument that the bias introduced by different plot sizes exceeds the other biases is probably not the case of fens, or ecologically analogous habitats when plots of the size of at least 1 m2 are considered. The crucial point is that large dataset consisting of thousands of vegetation plots are likely sufficiently robust to compensate for the potential effect of different plot sizes, at least in fens.

Different plot sizes in vegetation studies - future outlook and recommendations Vegetation plots stored in electronic databases are quite heterogeneous concerning plot sizes (Chytrý & Otýpková, 2003; Dengler et al., 2009; Peterka, et al. 2015) and this fact must be taken into account before any analysis. There is obviously no objective way how to determine the “correct” plot sizes for research of plant communities (Økland, 1990; Chytrý & Otýpková, 2003; Berg, Schwager, Pöltl, & Dengler, 2016) and the range of analysed plot sizes should hence depend on studied questions (Kenkel, Juhász-Nagy, & Podani, 1989; Jalonen, Vanha-Majamaa, & Tonteri, 1998). In this study, we present an indirect evidence that plots of different sizes might be jointly included within surveys aiming at a delimitation of fen vegetation units on a broad geographical scale when plots of extreme sizes are excluded. This approach represents a compromise between keeping a high level of methodological purism and inclusion of as many available data as possible. In the case of fen vegetation, plots of 1–25 m2 might be considered without introducing significant error. Plots smaller than 1 m2 seem to be inconvenient due to the considerably lower representation of specialists and less tight relation to the main environmental gradients. This scale might be, however, used for synusial approach to vegetation classification (Gillet & Julve, 2018) and it is regarded as optimal for sampling bryophyte communities in the parallel classification of cryptogam vegetation (Berg et al., 2016). The question of whether to use large plots (i.e.

98 larger than 25–30 m2) for classification of non-shrubby and non-forest plant communities deserves further research. Jiménez-Alfaro et al. (2013) and Peterka et al. (2017), while analysing subsets of different plot sizes by unsupervised classifications, have already found that the main environmental gradients were consistent across different plot sizes, including even those larger than 30 m2. Such large plots may be likely included in vegetation analyses in the case they were sampled in large and relatively homogeneous habitats, such as pristine mires in the boreal zone of Eurasia. It is important to emphasise that our suggestion of the possible joint use of different- sized plots might be implemented to broad-scale vegetation surveys focused on the delimitation of phytosociological units or detection of main gradients in target vegetation based on large and robust datasets (involving thousands of plots). Contrary, local vegetation studies should be apparently based on plots of equal size or even-sized plots as advocated by Dengler et al. (2009). Plots of different sizes are further unacceptable for research of community parameters varying with scale such as diversity patterns (Økland, Eilertsen, & Økland, 1990; Malanson, Fagre, & Zimmerman, 2018), evenness (Wilson, Steel, King, & Gitay, 1999) or assembly rules (Jonsson & Moen, 1998). The same holds for the design of in situ experiments scale (e.g. management or restoration treatments), whose results may be affected by the different processes at different spatial scale (Güsewell et al., 1998). What is then the optimal scale? For fen communities, the plot size of at least 1 m2 seems to provide sufficient information of specialists’ diversity and to mirror the main environmental gradients as well as community structure analogously to larger plots. At the same time, plots of medium sizes (1–10 m2) allow to establish more replications than those of 16–25 m2, the frequent standard in central-European countries and thus increase the sensitivity of the investigation (see also Wildi & Krüsi, 1992). We are aware that the results of our study cannot be simply generalised due to the restriction of our research to specific vegetation type (the Scheuchzerio palustris-Caricetea fuscae class). Our findings might be cautiously extrapolated to other plant communities having the analogous richness of species pool or sharing some environmental conditions such as high level of moisture, macro-nutrient limitation and island-like nature (e.g. bogs, oligotrophic marshes, aquatic vegetation, low-productive acidophilous grasslands). The potential impact of different plot sizes on the classification of other habitats deserves further research.

Acknowledgements We thank Milan Chytrý for providing data from the Czech National Phytosociological Database, Milan Valachovič for providing data from the Slovak Vegetation Database and Ilona Knollová for help with data preparation. Tereza Náhlíková kindly provided data from her master thesis. Dozens of colleagues joined us during our field expeditions in last years and helped with a sampling of vegetation plots; we are most indebted to Jana B. Jiroušková and Ondřej Knápek. The research was funded by the Czech Science Foundation (project no. 19-28491X) and Masaryk University (MUNI/A/0979/2017). PH was further supported by the long-term developmental project of the Czech Academy of Sciences (RVO 67985939).

References Arrhenius, O. (1921). Species and area. Journal of Ecology, 9, 95–99. Barkman, J. J. (1989). A critical evaluation of minimum area concepts. Vegetatio, 85, 89–104. Berg, C., Schwager, P., Pöltl, M., & Dengler, J. (2016). Plot sizes used for phytosociological sampling of bryophyte and lichen micro-communities. Herzogia, 29, 654–667. Bergamini, A., Peintinger, M., Fakheran, S., Moradi, H., Schmid, B., & Joshi, J. (2009). Loss of habitat specialists despite conservation management in fen remnants 1995–2006. Perspectives in Plant Ecology, Evolution, and Systematics, 11, 65–79. Bertness, M. D., & Callaway, R. (1994). Positive interactions in communities. Trends in Ecology & Evolution, 9, 191–193. Botta-Dukát, Z., Kovács-Láng, E., Rédei, T., Kertész, M., & Garadnai, J. (2007). Statistical and biological consequences of preferential sampling in phytosociology: theoretical considerations and a case study. Folia Geobotanica, 42, 141–152. Chytrý, M. (Ed.) (2011). Vegetation of the Czech Republic 3. Aquatic and wetland vegetation. Praha, CZ: Academia. Chytrý, M., Dražil, T., Hájek, M., Kalníková, V., Preislerová, Z., Šibík, J., ... & Vymazalová, M. (2015). The most species-rich plant communities in the Czech Republic and Slovakia (with new world records). Preslia, 87, 217–278. Chytrý, M., Hennekens, S. M., Jiménez-Alfaro, B., Knollová, I., Dengler, J., Jansen, ... & Yamalov, S. (2016). European Vegetation Archive (EVA): an integrated database of European vegetation plots. Applied Vegetation Science, 19, 173–180. Chytrý, M., & Otýpková, Z. (2003). Plot sizes used for phytosociological sampling of European vegetation. Journal of Vegetation Science, 14, 563–570. Chytrý, M., & Rafajová, M. (2003). Czech National Phytosociological Database: basic statistics of the available vegetation-plot data. Preslia, 75, 1–15. Chytrý, M., Tichý, L., Holt, J., & Botta-Dukát, Z. (2002). Determination of diagnostic species with statistical fidelity measures. Journal of Vegetation Science, 13, 79–90. Dahl, E. (1956). Rondane. Mountain vegetation in south Norway and its relation to the environment. Skrifter utgitt av Det Norske Videnskaps-Akademi i Oslo, Mat.-Naturvidensk. Klasse, 3, 1–374. De Cáceres, M., Chytrý, M., Agrillo, E., Attorre, F., Botta-Dukát, Z., Capelo, J., ... & Wiser, S. K. (2015). A comparative framework for broad-scale plot-based vegetation classification. Applied Vegetation Science, 18, 543–560. Dengler, J. (2003). Entwicklung und Bewertung neuer Ansätze in der Pflanzensoziologie unter besonderer Berücksichtigung der Vegetationsklassifikation. Archiv naturwissenschaftlicher Dissertationen, 14, 1–297. Dengler, J., Löbel, S., & Dolnik, C. (2009). Species constancy depends on plot size – a problem for vegetation classification and how it can be solved. Journal of Vegetation Science, 20, 754–766. Dietvorst, P., van der Maarel, E., & van der Putten, H. (1982). A new approach to the minimum area of a plant community. Vegetatio, 50, 77–91. Douda, J., Boublík, K., Slezák, M., Biurrun, I., Nociar, J., Havrdová, A., Doudová, J., Adid, S., Brisse, H., ... & Zimmermann, N. E. (2016). Vegetation classification and biogeography of European floodplain forests and alder carrs. Applied Vegetation Science, 19, 147–163. Drakare, S., Lennon, J. J., & Hillebrand, H. (2006). The imprint of the geographical, evolutionary and ecological context on species–area relationships. Ecology letters, 9, 215–227. Du Rietz, G. E. (1921). Zur methodologischen Grundlage der modernen Pflanzensoziologie. Holzhausen, AT: Wien. Eliáš, P. Jr., Sopotlieva, D., Dítě, D., Hájková, P., Apostolova, I., Senko, D., ... & Hájek M. (2013). Vegetation diversity of salt-rich grasslands in Southeast Europe. Applied Vegetation Science, 16, 521–537. Euro+Med (2006–2018). Euro+Med PlantBase – The information resource for Euro-Mediterranean plant diversity. Available at: http://ww2.bgbm.org/ EuroPlusMed/ (accessed August 2018). Fekete, G., & Szöcs, Z. (1974). Studies on interspecific association processes in space. Acta Botanica Academiae Scientiarum Hungarica, 20, 227–241. Forbes, B. C., & Sumina, O. I. (1999). Comparative ordination of low arctic vegetation recovering from disturbance: reconciling two contrasting approaches for field data collection. Arctic, Antarctic, and Alpine Research, 31, 389–399. Frey, W., Frahm J.-P., Fischer, E., & Lobin, W. (2006). The liverworts, mosses and ferns of Europe. Harley Books, UK: Colchester.

100

Gillet, F., & Gallandat, J. D. (1996). Integrated synusial phytosociology: some notes on a new multiscalar approach to vegetation analysis. Journal of Vegetation Science, 7, 13–18. Gillet, F., & Julve, P. (2018). The integrated synusial approach to vegetation classification and analysis. Phytocoenologia, 48, 141–152. Güsewell, S., Buttler, A., & Klötzli, F. (1998). Short‐term and long‐term effects of mowing on the vegetation of two calcareous fens. Journal of Vegetation Science, 9, 861–872. Hájek, M., Horsák, M., Hájková, P., & Dítě, D. (2006). Habitat diversity of central European fens in relation to environmental gradients and an effort to standardise fen terminology in ecological studies. Perspectives in Plant Ecology, Evolution and Systematics, 8, 97–114. Hájek, M., Jiroušek, M., Navrátilová, J., Horodyská, E., Peterka, T., Plesková, Z., ... & Hájek, T. (2015). Changes in the moss layer in Czech fens indicate early succession triggered by nutrient enrichment. Preslia, 87, 279– 301. Hájková, P., Hájek, M., Apostolova, I., Zelený, D., & Dítě, D. (2008). Shifts in the ecological behaviour of plant species between two distant regions: evidence from the base richness gradient in mires. Journal of Biogeography, 35, 282–294. Hettenbergerová, E., & Hájek, M. (2011). Is species richness of small spring fens influenced by the spatial mass effect? Community Ecology, 12, 202–209. Hill, M. O., Bell, N., Bruggeman-Nannenga, M. A., Brugués, M., Cano, M. J., Enroth, J., ... & Guerra, J. (2006). An annotated checklist of the mosses of Europe and Macaronesia. Journal of Bryology, 28, 198–267. Horsáková, V., Hájek, M., Hájková, P., Dítě, D., & Horsák, M. (2018). Principal factors controlling the species richness of European fens differ between habitat specialists and matrix-derived species. Diversity and Distributions, 24, 742–754. Jalonen, J., Vanha-Majamaa, I., & Tonteri, T. (1998). Optimal sample and plot size for inventory of field and ground layer vegetation in a mature Myrtillus-type boreal spruce forest. Annales Botanici Fennici, 35, 191– 196. Janišová, M., Michalcová, D., Bacaro, G., & Ghisla, A. (2014). Landscape effects on diversity of semi-natural grasslands. Agriculture, Ecosystems & Environment, 182, 47–58. Jansen, F., & Dengler, J. (2010). Plant names in vegetation databases – a neglected source of bias. Journal of Vegetation Science, 21, 1179–1186. Janssen, J. A., Rodwell, J. S., García Criado, M., Gubbay, S., Haynes, T., Nieto, ... & Valachovič, M. (2016). European Red List of Habitats. Part 2. Terrestrial and freshwater habitats. Publications Office of the European Union, L: Luxembourg. Jiménez-Alfaro, B., Hájek, M., Ejrnæs, R., Rodwell, J., Pawlikowski, P., Weeda, E. J., ... & Díaz, T. E. (2014). Biogeographic patterns of base-rich fen vegetation across Europe. Applied Vegetation Science, 17, 367– 380. Jonsson, B. G., & Moen, J. (1998). Patterns in species associations in plant communities: the importance of scale. Journal of Vegetation Science, 9, 327–332. Joosten, H., Tanneberger, F., & Moen, A. (Eds.) (2017). Mires and peatlands of Europe. Status, distribution and conservation. Schweizerbart Science Publishers, CH: Stuttgart Kenkel, N. C., Juhász-Nagy, P., & Podani, J. (1989). On sampling procedures in population and community ecology. Vegetatio, 83, 195–207. Klimeš, L., Dančák, M., Hájek, M., Jongepierová, I., & Kučera,T. (2001). Scale-dependent biases in species counts in grassland. Journal of Vegetation Science, 12, 699–704. Knollová, I., Chytrý, M., Tichý, L., & Hájek, O. (2005). Stratified resampling of phytosociological databases: some strategies for obtaining more representative data sets for classification studies. Journal of Vegetation Science, 16, 479–486. Kočí, M., Chytrý, M., & Tichý, L. (2003). Formalised reproduction of an expert-based phytosociological classification: A case study of subalpine tall-forb vegetation. Journal of Vegetation Science, 14, 601–610. Kooijman, A. M. (2012). ‘Poor rich fen mosses’: atmospheric N-deposition and P-eutrophication in base-rich fens. Lindbergia, 35, 42–52. Landucci, F., Tichý, L., Šumberová, K., & Chytrý, M. (2015). Formalized classification of species-poor vegetation: a proposal of a consistent protocol for aquatic vegetation. Journal of Vegetation Science, 26, 791–803. Lengyel, A., Chytrý, M., & Tichý, L. (2011). Heterogeneity constrained random resampling of phytosociological databases. Journal of Vegetation Science, 22, 175–183. Lengyel, A., & Podani, J. (2015). Assessing the relative importance of methodological decisions in classiﬁcations of vegetation data. Journal of Vegetation Science, 26, 804–815.

101

Lepš, J., & Hadincová, V. (1992). How reliable are our vegetation analyses. Journal of Vegetation Science, 3, 119–124. Lomolino, M. V. (2000). Ecology’s most general, yet protean pattern: the species–area relationship. Journal of Biogeography, 27, 17–26. Malanson, G. P., Fagre D. B., & Zimmerman D. L. (2018): Scale dependence of diversity in alpine tundra, Rocky Mountains, USA. Plant ecology, 219, 999–1008. Malmer, N. (1986). Vegetational gradients in relation to environmental conditions in northwestern European mires. Canadian Journal of Botany, 64, 375–383. Marcenò, C., Guarino, R., Loidi, J., Herrera, M., Isermann, M., Knollová, I., ... & Chytrý, M. (2018). Classification of European and Mediterranean coastal dune vegetation. Applied Vegetation Science, 21, 533–559. Matthews, T. J., Cottee‐Jones, H. E., & Whittaker, R. J. (2014). Habitat fragmentation and the species–area relationship: a focus on total species richness obscures the impact of habitat loss on habitat specialists. Diversity and Distributions, 20, 1136–1146. Michalcová, D., Lvončík, S., Chytrý, M., & Hájek, O. (2011). Bias in vegetation databases? A comparison of stratified-random and preferential sampling. Journal of Vegetation Science, 22, 281–291. Michalet, R., Le Bagousse–Pinguet, Y., Maalouf, J. P., & Lortie, C. J. (2014). Two alternatives to the stress‐gradient hypothesis at the edge of life: the collapse of facilitation and the switch from facilitation to competition. Journal of Vegetation Science, 25, 609–613. Moen, A., Lyngstad, A., & Øien, D. (2012). Boreal rich fen vegetation formerly used for haymaking. Nordic Journal of Botany, 30, 226–240. Moravec, J. (1973). The determination of the minimum area of phytocoenoses. Folia Geobotanica et Phytotaxonomica, 8, 23–47. Mucina, L., Bültmann, H., Dierßen, K., Theurillat, J.-P., Raus, T., Čarni, A., ... & Tichý, L. (2016). Vegetation of Europe: hierarchical floristic classification system of vascular plant, bryophyte, lichen, and algal communities. Applied Vegetation Science, 19 (Suppl. 1), 3–264. Navrátilová, J., Hájek, M., Navrátil, J., Hájková, P., & Frazier, R. J. (2017). Convergence and impoverishment of fen communities in a eutrophicated agricultural landscape of the Czech Republic. Applied Vegetation Science, 20, 225–235. Nordhagen, R. (1943). Sikilsdalen og Norges fjellbeiter. En plantesosiologisk monografi. Bergens Museums Skrifter, 22, 1–607. Oksanen, J., Blanchet, F. G., Friendly, M., Kindt, R., Legendre, P., McGlinn, D., ... Wagner, H. (2017). vegan: Community Ecology Package. R package version 2.4-2. Retrieved from https://CRAN.R-project.org/packa ge=vegan Økland, R. H. (1990). Vegetation ecology: theory, methods and application with reference to Fennoscandia. Sommerfeltia, Supplement, 1, 1–233. Økland, R. H., Eilertsen, O., & Økland, T. (1990). On the relationship between sample plot size and beta diversity in boreal coniferous forest. Vegetatio, 87, 187–192. Öster, M., Cousins, S., & Eriksson, O. (2007). Size and heterogeneity rather than landscape context determine plant species richness in semi-natural grasslands. Journal of Vegetation Science, 18, 859–868. Otýpková, Z., & Chytrý, M. (2006). Effects of plot size on the ordination of vegetation samples. Journal of Vegetation Science, 17, 465–472. Pakarinen, P. (1995). Classification of boreal mires in Finland and Scandinavia: a review. Vegetatio, 118, 29–38. Peterka T., Hájek M., Jiroušek M., Jiménez-Alfaro B., Aunina L., Bergamini A., ... & Chytrý M. (2017). Formalized classification of European fen vegetation at the alliance level. Applied Vegetation Science, 20, 124–142. Peterka, T. Jiroušek, M., Hájek, M., & Jiménez-Alfaro, B. (2015). European Mire Vegetation Database: a gap- oriented database for European fens and bogs. Phytocoenologia, 45, 291–298. Podani, J. (1984). Spatial processes in the analysis of vegetation: theory and review. Acta Botanica Hungarica, 30, 75–118. Rion, V., Gallandat, J.-D., Gobat, J.-M., & Vittoz, P. (2018). Recent changes in the plant composition of wetlands in the Jura Mountains. Applied Vegetation Science, 21, 121–131. Rodríguez-Rojo, M. P., Jiménez-Alfaro, B., Jandt, U.., Bruelheide, H., Rodwell, J. S., Schaminée, J. H. J., ... & Chytrý, M. (2017). Diversity of lowland hay meadows and pastures in Western and Central Europe. Applied Vegetation Science, 20, 702–719. Rodwell, J. S. (Ed.) (1991). British plant communities. Vol. 2. Mires and heaths. Cambridge, UK: Cambridge University Press. Roleček, J., Chytrý, M., Hájek, M., Lvončík, S., & Tichý L. (2007). Sampling design in large-scale vegetation studies: do not sacrifice ecological thinking to statistical purism! Folia Geobotanica, 42, 199–208.

102

Rybníček, K., Balátová-Tuláčková, E., & Neuhäusl, R. (1984). Přehled rostlinných společenstev rašelinišť a mokřadních luk Československa. Studie ČSAV, 1984/8, 1–124. Sádlo, J., Chytrý, M., & Pyšek, P. (2007). Regional species pools of vascular plants in habitats of the Czech Republic. Preslia, 79, 303–321. Šibík, J. (2012). Slovak Vegetation Database. Biodiversity & Ecology, 4, 429. Steiner, G. M. (1992). Österreichischer Moorschutzkatalog. Graz/Wien, AT: Verlag Ulrich Moser. Storch, D. (2016). The theory of the nested species-area relationship: geometric foundations of biodiversity scaling. Journal of Vegetation Science, 27, 880–891. Swacha, G., Botta-Dukát, Z., Kącki, Z., Pruchniewicz, D., & Żołnierz L. (2017). A performance comparison of sampling methods in the assessment of species composition patterns and environment – vegetation relationships in species-rich grasslands. Acta Societatis Botanicorum Poloniae, 86, DOI: https://doi.org/10.5586/asbp.3561. Tichý, L. (2002). JUICE, software for vegetation classification. Journal of Vegetation Science, 13, 451–453. Tichý, L., Chytrý, M., & Botta-Dukát, Z. (2014). Semi-supervised classification of vegetation: preserving the good old units and searching for new ones. Journal of Vegetation Science, 25, 1504–1512. Tichý, L., Chytrý, M., Hájek, M., Talbot, S. S., & Botta-Dukát, Z. (2010). OptimClass: Using species-to-cluster fidelity to determine the optimal partition in classification of ecological communities. Journal of Vegetation Science, 21, 287–299. Toman, M. (1990). Das Verhältnis zwischen Artenzahl und Aufnahmefläche in der Phytozönologie. Feddes Repertorium, 101, 665–673. Valachovič, M. (Ed.) (2001). Plant communities of Slovakia. 3. Wetland vegetation. Bratislava, SK: Veda. Vittoz, P., & Guisan, A. (2007). How reliable is the monitoring of permanent vegetation plots? A test with multiple observers. Journal of Vegetation Science, 18, 413–422. Vymazalová, M., Tichý, L., & Axmanová, I. (2014). How does vegetation sampling in different parts of the growing season influence classification results and analyses of beta diversity? Applied Vegetation Science, 17, 556–566. Wagner, V. (2016). A review of software tools for spell-checking taxon names in vegetation databases. Journal of Vegetation Science, 27, 1323–1327. Westhoff, V., & van der Maarel, E. (1978). The Braun-Blanquet approach. In R. H. Whittaker (Ed.), Classification of plant communities (pp. 289–399). Hague, NL: W. Junk. Whittaker, R. J. (1998). Island biogeography: Ecology, evolution, and conservation. Oxford, GB: Oxford University Press. Wickham, H. (2009). ggplot2: elegant graphics for data analysis. New York , US: Springer. Wiens, J. A. (1989). Spatial scaling in ecology. Functional Ecology, 3, 385–397. Wildi, O., & Krüsi, B. O. (1992). Long-term monitoring: the function of plot size and sampling design. Abstracta Botanica, 16, 7–14. Willner, W., Jiménez-Alfaro, B., Agrillo, E., Biurrun, I., Campos, J. A., Čarni, A., ... & Chytrý, M. (2017). Classification of European beech forests: a Gordian Knot? Applied Vegetation Science, 20, 494–512. Willner, W., Kuzemko, A., Dengler, J., Chytrý, M., Bauer, N., Becker, T., ... & 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. Wilson, J. B., Steel, J. B., King, W. McG., & Gitay, H. (1999). The effect of spatial scale on evenness. Journal of Vegetation Science, 10, 463–468. Wood, S. N. (2006). Generalized Additive Models: An Introduction with R. Boca Raton, US: Chapman & Hall/CRC.

103

TABLE 1. Subsets of vegetation plots used for the construction of species-area curves. Ecological type of fen Syntaxonomical Region Subset No. of interpretation No. plots calcareous and extremely rich Caricion davallianae, central Europe 1 990 fens with calcicole vascular plants Caricion atrofusco-saxatilis Scandinavia - 214 and without Sphagnum species (only in Scandinavia) rich fens with calcium-tolerant Sphagno warnstorfii- central Europe 2 650 Sphagnum species Tomentypnion nitentis (incl. Scandinavia 3 439 transitions to Stygio-Caricion limosae) poor fens Sphagno-Caricion central Europe 4 1021 canescentis Scandinavia 5 338 dystrophic hollows Scheuchzerion palustris central Europe 6 275 Scandinavia 7 341

TABLE 2. Results of chi-square statistics (χ2) and AIC. The chi-square statistics indicates whether two separate curves (one for specialists, the second one for generalists) represent a better model as compared to one joint curve combining both species groups. AIC estimates, which model is better. m1 = model 1 (joint curve for all species), m2 = model 2 (separate curves for specialists and generalists). Subset No. χ2 Df P m1 m2 1 146.45 2.87 <0.001 13820.87 13679.51 2 75.135 2.79 <0.001 9558.775 9488.604 3 57.496 2.16 <0.001 7393.092 7310.574 4 76.954 2.98 <0.001 12140.76 12069.45 5 25.16 1.29 <0.001 4118.432 4095.868 6 17.326 2.92 <0.001 2432.177 2420.155 7 1.1752 1.04 0.29 3364.556 3365.462

TABLE 3. The congruence among ordinations of plots of different sizes. The congruence among individual ordination is indicated by the sum of squared differences (SS; above diagonal) and Procrustes R2 statistics (below diagonal). Plot size (m2) 0.07 0.25 1 4 16 0.07 – 0.361 0.297 0.327 0.34 0.25 0.827 – 0.113 0.141 0.165 1 0.838 0.941 – 0.064 0.094 4 0.82 0.926 0.967 – 0.09 16 0.812 0.913 0.952 0.954 –

104

FIGURE 1. Species-area curves for individual ecological types of fens. The horizontal axes express plot size in log scale; the vertical axis represents a number of species. (a) calcareous and extremely rich fens with calcicole vascular plants and without Sphagnum species, central Europe; (b) rich fens with calcium-tolerant Sphagnum species, central Europe; (c) rich fens with calcium-tolerant Sphagnum species, Scandinavia; (d) poor fens, central Europe; (e) poor fens, Scandinavia; (f) dystrophic hollows, central Europe; (g) dystrophic hollows, Scandinavia.

105

FIGURE 2. PCoA diagram showing the difference among ordinations of plots of different sizes. Symbol sizes represent the size of plots used within individual ordination. Distances among individual boxes represent dissimilarity among ordinations according to the Procrustes analysis.

FIGURE 3. Representation of plot sizes in clusters produced by unsupervised K-means classification. The horizontal axis represents 50 individual resulting clusters (Appendix S4), symbols in one column represents the sizes of plots in the particular cluster. The vertical axis is plot size in log scale. Individual symbols (crosses, circles) represent quantities of specific plot sizes within the particular cluster.

106

Appendix S1. List of aggregates and species complexes. Name Included taxa Achillea millefolium agg. A. collina, A. millefolium, A. pratensis Agrostis stolonifera agg. A. gigantea, A. stolonifera Alchemilla vulgaris agg. all Alchemilla species except A.glaucescens Anthoxanthum odoratum agg. A. alpinum, A. odoratum Callitriche palustris agg. C. cophocarpa, C. hamulata, C. palustris, C. platycarpa Campylium stellatum agg. C. protensum, C. stellatum Cardamine pratensis agg. C. dentata, C. matthioli, C. pratensis Carex muricata agg. C. divulsa, C. muricata, C. spicata Centaurea jacea agg. C. jacea, C. macroptilon Chiloscyphus polyanthos agg. C. pallescens, C. polyanthos Dactylorhiza maculata agg. D. fuchsii, D. maculata Empetrum nigrum agg. E. hermaphroditum, E. nigrum Festuca rubra agg. F. nigrescens, F. rubra, F. trichophylla Galium mollugo agg. G. album, G. mollugo Galium palustre agg. G. elongatum, G. palustre Gymnadenia conopsea agg. G. conopsea (L.) R. Br., G. densiflora (Wahlenb.) A. Dietrich Hypnum cupressiforme agg. H. andoi, H. cupressiforme, H. jutlandicum Juncus bufonius agg. J. bufonius, J. hybridus, J. ranarius Knautia arvensis agg. K. arvensis, K. kitaibelii Leucanthemum vulgare agg. L. ircutianum, L. vulgare Lophocoela bidentata agg. L. bidentata, L. heterophylla Lotus corniculatus agg. L. corniculatus, L. tenuis Luzula campestris agg. L. campestris, L. multiflora, L. pallescens, L. sudetica Mentha arvensis agg. M. arvensis, M. × verticillata Molinia caerulea agg. M. arundinacea, M. caerulea Myosotis palustris agg. M. nemorosa, M. palustris Palustriella commutata agg. P. falcata, P. commutata Plagiomnium affine agg. P. affine, P. elatum, P. ellipticum, P. medium Plagiothecium laetum agg. P. curvifolium, P. laetum Poa pratensis agg. P. angustifolia, P. pratensis, P. subcaerulea Ranunculus auricomus agg. R. auricomus, R. fallax Rubus fruticosus agg. all Rubus species except R. caesius, R. chamaemorus, R. idaeus and R. saxatilis. Salix repens agg. S. repens, S. rosmarinifolia Scorpidium revolvens agg. S. cossonii, S. revolvens Senecio nemorensis agg. S. hercynicus, S. nemorensis, S. ovatus Sphagnum affine agg. S. affine, S. austinii, S. imbricatum Sphagnum annulatum agg. S. annulatum, S. jensenii Sphagnum auriculatum agg. S. auriculatum, S. inundatum Sphagnum palustre agg. S. centrale, S. palustre Sphagnum recurvum agg. S. angustifolium, S. brevifolium, S. fallax, S. flexuosum Taraxacum sect. xy (sections Taraxacum spp. div. according to Euro+Med) Utricularia intermedia agg. U. intermedia, U. ochroleuca Utricularia minor agg. U. bremii, U. minor Vaccinium oxycoccos agg. V. microcarpum, V. oxycoccos Veratrum album agg. V. album, V. lobelianum Vicia cracca agg. V. cracca, V. tenuifolia

Appendix S2. Ecological types of fens used for species-area curves, and description of individual ecological types of fens and their formal definitions.

Type descriptions: (I) Calcareous and extremely rich fens with calcicole vascular plants and without Sphagnum species The group comprises fen vegetation rich in available mineral (especially calcium) content. The communities occur on both calcareous tufa-forming springs and peat-forming substrates. The herb layer consists mainly of calcicole graminoids (e.g. Carex davalliana, C. hostiana, Eleocharis quinqueflora, Eriophorum latifolium, Schoenus ferrugineus) and herbs (Parnassia palustris, Pinguicula vulgaris, Primula farinosa, Tofieldia calyculata, Valeriana dioica). The bryophyte layer is made up of brown-mosses such as Bryum pseudotriquetrum, Campylium stellatum, Palustriella commutata agg., Scorpidium revolvens

107 agg. Sphagnum species are absent. From a phytosociological point of view, vegetation is classified in the Caricion davallianae and the Caricion atrofusco-saxatilis alliances. The latter, which occurs exclusively in Scandinavia, is further characterised by the presence of species with arcto-alpine distributional ranges (e.g. Carex atrofusca, C. microglochin, Juncus triglumis, Salix reticulata, Thalictrum alpinum). Due to a limited number of vegetation plots of calcareous and extremely rich fens from Scandinavia, the species-area curves for this type were constructed only using plots sampled in central Europe. (II) Rich fens with calcium-tolerant Sphagnum species The type is characterized by generally slightly acidic and sub-neutral pH and a medium level of mineral content. Bryophyte layer comprises calcium-tolerant sphagna, i.e. Sphagnum contortum, S. subnitens, S. teres, S. warnstorfii and S. subfulvum (the latter only in plots from Scandinavia), which are accompanied by other mosses dependently on water regime (e.g. Aulacomnium palustre, Bryum pseudotriquetrum, Campylium stellatum, Paludella squarrosa, Pseudocalliergon trifarium, Scorpidium revolvens agg., S. scorpioides, Tomentypnum nitens). In the analysed dataset, the alliances of Stygio-Caricion limosae and Sphagno warnstorfii-Tomentypnion nitentis, as well as transition stands, are included. The first alliance comprises sedge-brownmoss fens on strongly waterlogged sites with peat accumulation; vegetation is characterised by semi-aquatic brownmosses (most typically by Pseudocalliergon trifarium, Scorpidium scorpioides) and sphagna of the Subsecunda section (mainly Sphagnum contortum, S. subsecundum). The latter comprises rich fens of less waterlogged conditions as compared to the previous alliance; vegetation is characterised by a higher proportion of sphagna (e.g. Sphagnum subnitens, S. teres, S. warnstorfii) and other bryophytes forming small hummocks, such as Aulacomnium palustre and Tomentypnum nitens. (III) Poor fens Poor fens include vegetation of acidic minerotrophic mires. Frequent dominants of the moss layer are Sphagnum recurvum agg., S. sect. Palustria (S. palustre, S. papillosum) and Polytrichum commune. Other non-sphagnaceous mosses are rarely present, with the exception of Straminergon stramineum. The type is classified in the Sphagno-Caricion canescentis alliance. (IV) Dystrophic (bog) hollows The type involves vegetation of dystrophic extremely acidic hollows of the Scheuchzerion palustris alliance. The moss layer is usually formed by Sphagnum cuspidatum, S. lindbergii, S. majus, Warnstorfia fluitans. The herb layer consists of few species, such as Carex limosa, Rhynchospora alba and Scheuchzeria palustris, or can be even absent. The community is traditionally ranked to the Scheuchzerio-Caricetea class; however, bog elements (Andromeda polifolia, Eriophorum vaginatum) frequently occur. In central Europe, the stands are restricted to bog hollows, whereas in northern Europe, the vegetation can cover larger areas, especially in the flat-terrain landscapes.

For further description of above mentioned fen types see for example Hájek, Horsák, Hájková, & Dítě (2006), Chytrý (2011), Moen, Lyngstad, & Øien (2012), Joosten, Tanneberger, & Moen (2017), Peterka et al. (2017).

Type definitions: The definitions consist of functional species groups (#TC), selected species and logical (AND, OR, NOT) plus relational (GR) operators. Brackets are used to group pairs of elements within the definition. Numbers refer to percentage covers. Functional species groups (according to Landucci, Tichý, Šumberová, & Chytrý, 2015) are present in the vegetation plot if the total cover of the member species of the group exceeds the given percentage threshold. Merging of covers of individual species follows the protocol of the JUICE software, recently formally described by Fischer (2015). For further information about the structure and creation of formal definitions see, e.g. Chytrý (2007), Landucci et al. (2015) or Peterka et al. (2017). All the symbols and the structure of the formulas follow the protocol of the expert systems working in the JUICE software (Tichý, 2002; http://www.sci.muni.cz/botany/juice/).

Operators GR = greater than, i.e. the cover of particular functional species group is greater than the cover of given values expressed in percentages or greater than the cover of another functional species group AND = both elements must be present OR = at least one element must be present NOT = element(s) must not be present

Definitions

Calcareous and extremely rich fens ((<#TC Base-rich-brown-mosses GR15>AND<#02 Calcareous-fen-specialists>)OR(<#TC Calcareous-fen-specialists-core GR30>AND<#01 Base-rich-brown-mosses>))NOT(<#TC Sphagnum spp. GR00>OR<#TC Stygio-Caricion-limosae-core- bryophytes GR15>)

108

Rich fens (((<#TC Rich-fen-bryophytes GR25>OR<#TC Base-rich-brown-mosses GR25>)AND<#TC Sphagnum spp. GR00>)AND<#TC rich-fen-core-sphagna GR00>)AND(<#TC Rich-fen-bryophytes GR #TC Poor-fen-bryophytes>NOT<#TC Dystrophic-hollows- bryophytes GR05>)

Poor fens (<#TC Poor-fen-bryophytes GR50>AND<#TC Poor-fen-bryophytes GR #TC Dystrophic-hollows-bryophytes>)NOT(<#TC Base- rich-brown-mosses GR05>OR<#TC Rich-fen-bryophytes GR20>)

Dystrophic hollows (<#TC Dystrophic-hollows-bryophytes GR15>AND(<#TC Dystrophic-hollows-bryophytes GR #TC Poor-fen- bryophytes>AND<#TC Dystrophic-hollows-bryophytes GR #TC Drepanocladion-exannulati-core-bryophytes >))NOT((<#TC Base-rich-brown-mosses GR00>OR<#TC Calcareous-fen-specialists GR00>)OR<#TC Rich-fen-bryophytes GR05>)

Functional species groups

#TC Base-rich-brown-mosses Bryum pseudotriquetrum Palustriella commutata agg. Campyliadelphus elodes Palustriella decipiens Campylium stellatum agg. Philonotis calcarea Catoscopium nigritum Pseudocalliergon lycopodioides Cratoneuron filicinum Scorpidium revolvens agg. Ctenidium molluscum Tomentypnum nitens Fissidens adianthoides

#TC Calcareous-fen-specialists Blysmus compressus Juncus subnodulosus Carex atrofusca Palustriella commutata agg. Carex bicolor Parnassia palustris Carex capillaris Philonotis calcarea Carex davalliana Pinguicula vulgaris Carex distans Polygala amarella Carex hostiana Primula farinosa Carex lepidocarpa Schoenus ferrugineus Carex microglochin Schoenus nigricans Carex pulicaris Sesleria uliginosa Carex viridula Tofieldia calyculata Eleocharis quinqueflora Triglochin maritima Epipactis palustris Triglochin palustris Equisetum variegatum Valeriana dioica Eriophorum latifolium

#TC Calcareous-fen-specialists-core Blysmus compressus Eleocharis quinqueflora Carex atrofusca Schoenus ferrugineus Carex davalliana Schoenus nigricans Carex hostiana Sesleria uliginosa Carex lepidocarpa Tofieldia calyculata Carex microglochin Triglochin maritima Carex viridula

#TC Drepanocladion-exannulati-core-bryophytes Warnstorfia sarmentosa Warnstorfia exannulata

#TC Dystrophic-hollows-bryophytes Gymnocolea inflata Sphagnum majus Sphagnum balticum Sphagnum tenellum Sphagnum cuspidatum Warnstorfia fluitans Sphagnum lindbergii

109

#TC Poor-fen-bryophytes Polytrichum commune Sphagnum recurvum agg. Sphagnum capillifolium Sphagnum riparium Sphagnum obtusum Sphagnum russowii Sphagnum pulchrum Straminergon stramineum

#TC Rich-fen-bryophytes Aulacomnium palustre Paludella squarrosa Breidleria pratensis Philonotis fontana Bryum pseudotriquetrum Pseudocalliergon trifarium Calliergon giganteum Scorpidium revolvens agg. Calliergon richardsonii Scorpidium scorpioides Campylium stellatum agg. Sphagnum contortum Cinclidium stygium Sphagnum platyphyllum Cinclidium subrotundum Sphagnum subfulvum Fissidens adianthoides Sphagnum subnitens Hamatocaulis vernicosus Sphagnum subsecundum Helodium blandowii Sphagnum teres Loeskypnum badium Sphagnum warnstorfii Meesia triquetra Tomentypnum nitens

#TC Stygio-Caricion-limosae-core-bryophytes Pseudocalliergon trifarium Scorpidium scorpioides

#TC rich-fen-core-sphagna Sphagnum contortum Sphagnum warnstorfi

References Chytrý, M. (Ed.) (2007). Vegetation of the Czech Republic 1. Grassland and heathland vegetation. Praha, CZ: Academia. Chytrý, M. (Ed.) (2011). Vegetation of the Czech Republic 3. Aquatic and wetland vegetation. Praha, CZ: Academia. Fischer, H.S. (2015). On the combination of species cover values from different vegetation layers. Applied Vegetation Science, 18, 169–170. Hájek, M., Horsák, M., Hájková, P., & Dítě, D. (2006). Habitat diversity of central European fens in relation to environmental gradients and an effort to standardise fen terminology in ecological studies. Perspectives in Plant Ecology, Evolution and Systematics, 8, 97–114. Joosten, H., Tanneberger, F., & Moen, A. (Eds.) (2017). Mires and peatlands of Europe. Status, distribution and conservation. Schweizerbart Science Publishers, CH: Stuttgart Landucci, F., Tichý, L., Šumberová, K. & Chytrý, M. (2015). Formalized classification of species-poor vegetation: a proposal of a consistent protocol for aquatic vegetation. Journal of Vegetation Science, 26, 791–803. Moen, A., Lyngstad, A., & Øien, D. (2012). Boreal rich fen vegetation formerly used for haymaking. Nordic Journal of Botany, 30, 226–240. Peterka T., Hájek M., Jiroušek M., Jiménez-Alfaro B., Aunina L., Bergamini A., ... & Chytrý M. (2017). Formalized classification of European fen vegetation at the alliance level. Applied Vegetation Science, 20, 124–142. Tichý, L. (2002). JUICE, software for vegetation classification. Journal of Vegetation Science, 13, 451–453.

Appendix S3. List of habitat specialists species. Only species present in initial (i.e. non-stratified) dataset of vegetation plots are listed. An asterisk (*) indicates species not regarded as fen specialists in Mucina et al. (2016).

Agrostis canina Campylium stellatum agg. Allium schoenoprasum Carex aquatilis Aneura pinguis (*) Carex atrofusca Aulacomnium palustre Carex bicolor Blysmus compressus Carex buxbaumii Breidleria pratensis (*) Carex canescens Bryum pseudotriquetrum (*) Carex capitata Calamagrostis neglecta Carex cespitosa Calliergon giganteum Carex chordorrhiza Calliergon richardsonii Carex davalliana Campyliadelphus elodes Carex demissa

110

Carex diandra Juncus castaneus Carex dioica Juncus filiformis Carex echinata Juncus stygius Carex flava Juncus subnodulosus Carex hartmanii Juncus triglumis Carex heleonastes Kobresia simpliciuscula Carex hostiana Ligularia sibirica Carex lasiocarpa Loeskypnum badium Carex laxa Meesia triquetra Carex lepidocarpa Meesia uliginosa Carex limosa Menyanthes trifoliata Carex livida Narthecium ossifragum Carex magellanica Paludella squarrosa Carex maritima Palustriella commutata agg. (*) Carex microglochin Parnassia palustris Carex nigra Pedicularis palustris Carex norvegica Pedicularis sceptrum-carolinum Carex panicea Philonotis calcarea Carex parallela Philonotis fontana Carex pulicaris Pinguicula alpina Carex rariflora (*) Pinguicula vulgaris Carex rostrata Polygala amarella Carex rotundata Primula farinosa Carex saxatilis Pseudocalliergion turgescens Carex vaginata Pseudocalliergon lycopodioides Carex viridula Pseudocalliergon trifarium Catoscopium nigritum Ranunculus flammula Cinclidium stygium Rhizomnium pseudopunctatum (*) Cinclidium subrotundum Rhynchospora alba Cladopodiella fluitans Rhynchospora fusca Comarum palustre Riccardia chamaedryfolia Dactylorhiza maculata agg. Riccardia incurvata Dactylorhiza majalis Riccardia latifrons Dicranum bonjeanii Salix repens agg. Drepanocladus polygamus Scheuchzeria palustris Drepanocladus sendtneri Schoenus ferrugineus Drosera intermedia Schoenus nigricans Drosera longifolia Scorpidium revolvens agg. Drosera rotundifolia Scorpidium scorpioides Dryopteris cristata Selaginella selaginoides (*) Eleocharis quinqueflora Sesleria uliginosa Epilobium palustre Sphagnum annulatum agg. (*) Epipactis palustris Sphagnum auriculatum agg. Equisetum palustre (*) Sphagnum capillifolium Equisetum variegatum Sphagnum contortum Eriophorum angustifolium Sphagnum cuspidatum Eriophorum gracile Sphagnum fimbriatum Eriophorum latifolium Sphagnum lindbergii Eriophorum scheuchzeri Sphagnum majus Euphrasia frigida Sphagnum molle Fissidens adianthoides Sphagnum obtusum Fissidens osmundioides Sphagnum palustre agg. Gymnadenia conopsea agg. Sphagnum papillosum Gymnocolea inflata (*) Sphagnum platyphyllum Hamatocaulis vernicosus Sphagnum pulchrum Hammarbya paludosa Sphagnum recurvum agg. Helodium blandowii (*) Sphagnum riparium Hydrocotyle vulgaris Sphagnum russowii Juncus alpinoarticulatus Sphagnum squarrosum Juncus arcticus Sphagnum subfulvum Juncus articulatus Sphagnum subnitens Juncus biglumis Sphagnum subsecundum Juncus bulbosus Sphagnum teres

111

Sphagnum warnstorfii Triglochin palustris Straminergon stramineum Tritomaria quinquedentata (*) Succisa pratensis Utricularia intermedia agg. Swertia perennis Utricularia minor agg. Tayloria lingulata Vaccinium oxycoccos agg. (*) Tephroseris crispa Valeriana dioica Thalictrum alpinum Veronica scutellata (*) Thelypteris palustris Viola epipsila (*) Tofieldia calyculata Viola palustris Tofieldia pusilla Warnstorfia exannulata Tomentypnum nitens Warnstorfia fluitans Trichophorum alpinum Warnstorfia sarmentosa Trichophorum cespitosum Warnstorfia tundrae Trichophorum pumilum Willemetia stipitata Triglochin maritima

References Mucina, L., Bültmann, H., Dierßen, K., Theurillat, J.-P., Raus, T., Čarni, A., ... & Tichý, L. (2016). Vegetation of Europe: hierarchical floristic classification systemof vascular plant, bryophyte, lichen, and algal communities. Applied Vegetation Science, 19 (Suppl. 1), 3–264.

112

Supporting Information to the paper Peterka, T. et al. Is variable plot size a serious constraint in broad-scale vegetation studies? A case study on fens. Journal of Vegetation Science. Appendix S4. Results of non-hierarchical K-means clustering. Diagnostic species (phi > 0.3) of individual cluster are indicated by backrground shading. The frequency values are shown. The diagnostic species are sorted according to fidelity. Other species are sorted according to decreasing frequency in the dataset (only species with more than 100 occurrences are listed).

Cluster No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 Diagnostic species: Palustriella commutata agg. 99 33 38 22 . 2 11 5 4 1 5 1 . 3 11 1 4 . 1 . 1 . 1 . 2 . . . . 42 1 ...... Tussilago farfara 39 22 11 9 . 1 5 1 13 9 . . . . 1 . 2 . . . . . 2 ...... 10 ...... Calliergonella cuspidata 64 99 50 89 . 34 75 11 57 15 . . . 1 2 24 44 1 47 26 33 53 48 . 62 3 5 40 1 2 3 . 1 1 6 5 . 2 1 1 ...... 1 . 1 . Carex lepidocarpa 24 8 48 2 . 1 31 21 . . 1 2 1 4 10 7 13 1 1 4 . 1 5 . 3 . . . . 4 1 ...... Eleocharis quinqueflora 24 2 51 21 . 2 1 12 . 5 1 23 10 5 31 27 7 . 3 2 1 . 4 . 4 . . 1 . 2 1 1 1 . . 1 ...... Leontodon hispidus 12 10 13 47 . 12 14 1 . 10 . . . . . 11 17 . 10 . 1 8 2 ...... 10 1 . . . 2 1 1 3 . 1 ...... Breidleria pratensis . 2 7 37 . 5 11 ...... 1 5 16 1 21 3 4 18 5 . 3 . 1 1 . 2 ...... Scorzoneroides autumnalis 1 2 5 1 81 2 5 . 2 6 22 . . 3 1 8 4 1 5 5 . 2 1 ...... 1 2 1 . . . . . 1 ...... 1 Sanguisorba officinalis 4 18 7 5 100 18 27 1 . 3 ...... 13 . 10 5 12 23 18 . 4 . 3 2 ...... 7 9 . 3 1 1 ...... Colchicum autumnale 1 4 2 6 49 8 5 . . 3 ...... 2 . . . . 1 1 ...... Polygala amarella 7 . 9 3 46 6 12 9 . 1 ...... 4 ...... 2 ...... Epipactis palustris 27 34 39 53 86 2 17 27 2 3 . . 1 . 4 1 38 1 10 2 2 1 2 . 4 . . . . 1 . . . . 1 1 ...... Lathyrus pratensis 2 18 8 39 70 28 22 . . 9 . . . . 1 . 32 . 6 . 10 10 25 . 9 . . 1 . 1 . . . . 1 1 . 1 ...... Lotus pedunculatus . 1 . 12 43 . 2 ...... 2 . 2 5 4 19 10 . 2 . 2 5 ...... 12 2 ...... Luzula campestris agg. 1 . 1 20 78 22 9 2 11 2 7 . . . 1 3 25 8 27 7 20 59 25 . 2 . . 3 . 7 4 1 . . 19 10 . 8 5 10 ...... 1 . 1 Parnassia palustris 30 6 69 31 100 27 37 55 2 13 28 1 1 18 14 41 42 23 26 14 5 7 14 . 6 12 1 . 1 22 1 1 . . 1 2 ...... 1 . . Galium palustre agg. 1 3 8 35 100 18 25 1 . 9 . 2 2 1 7 16 18 1 20 23 55 43 60 1 69 37 16 68 7 1 4 . 4 6 19 51 1 11 3 1 ...... 2 . . . Eriophorum latifolium 68 55 78 86 100 50 71 33 26 2 2 3 . 17 9 29 56 9 24 9 2 4 7 . 10 . . 1 1 3 . 1 . . 1 1 1 1 1 1 ...... 1 1 . Holcus lanatus 1 13 1 18 62 5 18 1 . 4 . . . . . 3 20 . 22 5 21 37 24 . 12 . 2 8 ...... 9 12 . 12 1 1 ...... Trisetum flavescens . 1 . 1 16 . . . . 4 ...... 1 ...... 1 1 ...... Ranunculus acris 23 46 18 70 95 87 76 2 30 16 11 . . 1 1 7 40 2 25 5 23 51 53 . 16 . . 1 . 12 . . . . 9 9 . 2 . 1 ...... Vicia cracca agg. 3 7 5 20 49 29 18 4 39 7 ...... 10 1 2 . 1 4 7 . 3 . . 1 ...... 1 ...... Rhinanthus minor 1 1 2 11 . 61 5 2 2 4 2 . . 1 1 . 3 1 . . . 4 13 ...... 1 ...... Dactylorhiza maculata agg. 1 . 2 2 . 64 14 . 2 1 1 . . 2 . 2 4 3 5 3 . 13 17 ...... 9 . 1 2 2 6 1 2 1 4 3 ...... 1 1 1 . Trifolium pratense 2 3 3 27 . 59 17 . 43 3 ...... 6 . 2 1 1 5 16 . . . . 1 . 4 . . . . 1 1 ...... Cirsium canum 2 3 1 . . 36 21 4 . 4 ...... 1 . . . . . 3 . 2 ...... Geum rivale 4 1 4 22 . 62 8 1 35 . 1 . . 1 1 . 32 5 6 1 5 7 21 . 9 . . 1 . 5 . . . . 2 1 . 1 ...... Daucus carota 2 5 2 3 . 20 2 . . 1 ...... 1 ...... Carex flava 38 26 28 48 . 78 36 7 9 6 7 7 1 29 9 23 31 6 26 19 2 13 44 . 6 5 1 . 2 26 3 1 . 1 5 2 1 3 . 1 . 1 . . . . 1 . . . Trifolium aureum . . . . . 9 ...... Sesleria uliginosa 2 1 8 . . . 24 . . 1 ...... 1 ...... 1 ...... Mentha aquatica 11 2 2 2 . 21 41 4 . 3 . 1 . . 1 . 2 . 1 1 1 1 6 . 4 . 1 1 ...... 1 ...... Serratula tinctoria 1 . 1 . . . 15 ...... 1 ...... Schoenus ferrugineus . . 4 . . . 2 81 . . 1 . 1 3 1 ...... 1 ...... 1 . . Centaurium littorale ...... 16 . 3 ...... Ctenidium molluscum 4 1 3 . . . 5 11 72 3 ...... 1 ...... Plantago maritima . . . . . 2 . 4 39 7 . . . . 1 ...... Festuca ovina . 1 1 . . . 1 1 57 2 20 . . 4 1 1 2 11 1 3 2 3 2 ...... 1 1 . 2 1 . 1 1 4 ...... 1 . 4 Carex hostiana 9 1 24 1 . 3 43 33 67 . . 2 . 2 3 4 5 . 2 1 . . 1 . 1 ...... Campyliadelphus chrysophyllus. 1 1 . . . . . 24 4 ...... Centaurea jacea agg. 7 14 4 8 22 10 23 . 59 14 . . . . . 1 4 . 3 . . 2 2 . 1 ...... 1 ...... Carex pulicaris . 1 1 1 14 1 9 2 57 . . . . 1 1 29 10 . 16 11 1 9 5 ...... 1 . 3 ...... 2 Succisa pratensis 37 19 40 12 27 23 52 32 93 1 1 1 . 2 3 17 53 2 32 17 8 27 11 . 7 . 1 1 1 . . . 5 . 4 7 . 8 1 2 ...... 1 Lotus corniculatus agg. 6 1 10 14 . 28 25 4 50 10 ...... 7 . 2 . 1 3 3 . 1 . . . . 2 . . . . 1 ...... Salix repens agg. 1 2 12 1 11 4 15 5 48 1 1 2 . . 4 3 9 2 3 1 2 1 4 . 2 . 1 3 1 . . 1 4 . 1 2 1 2 . . . . 2 . . . . 1 . 1 Cluster No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 Prunella vulgaris 29 29 21 65 22 63 45 9 83 6 . . . . 2 10 22 . 12 12 2 18 21 . 2 . 1 . . 13 1 . . . 3 1 ...... Viola canina . . 1 1 . . 5 . 20 ...... 2 . 1 3 . 2 1 . . . . 1 ...... Carex dioica 3 1 32 14 . 14 . 16 74 1 30 15 2 49 24 18 43 52 21 21 2 1 1 . 5 2 . . 2 . 1 4 1 1 . 2 2 6 ...... 5 . 2 Fissidens adianthoides 36 37 35 21 41 2 33 19 67 4 8 . 1 7 3 37 23 1 11 10 1 4 4 . 1 . 1 . . 4 . 1 2 ...... Carex atrofusca ...... 62 1 . 2 1 . . 1 ...... Juncus triglumis ...... 1 57 3 . 5 9 . . 1 ...... 2 ...... Salix reticulata ...... 4 57 1 . 5 6 . . 6 ...... 1 ...... Carex capillaris . . 3 1 . . . . 15 4 66 1 . 11 2 . . 5 ...... Pedicularis oederi ...... 39 . . 2 . . . 3 ...... 1 ...... Bistorta vivipara 1 . 2 1 . 3 . 1 . 22 94 3 . 30 27 1 8 38 . 1 2 . . . . . 3 . . 8 4 15 . . . 1 . . 1 1 ...... Saxifraga aizoides 1 ...... 3 47 1 . 7 3 . . 2 ...... 3 ...... Carex vaginata ...... 55 . . 6 . . . 14 3 1 ...... 3 . . . . 1 1 2 2 ...... 1 . 4 Meesia uliginosa ...... 1 36 1 . 3 3 . . 4 ...... 1 ...... Saussurea alpina ...... 2 55 . . 22 4 1 . 25 ...... 1 1 ...... 2 Tofieldia pusilla ...... 14 . 4 52 2 . 37 4 1 . 25 . 1 ...... 1 ...... 2 . . Carex norvegica ...... 4 22 . . . 1 . . 1 ...... 1 1 ...... Oncophorus virens ...... 4 31 2 . 3 4 . . 3 ...... 1 2 1 3 ...... 1 . . Catoscopium nigritum . . 1 ...... 1 29 . . 10 6 . . 1 ...... Carex parallela ...... 1 22 . . 1 4 . . 1 ...... 2 ...... Salix myrsinites 1 ...... 1 . . 37 2 . 10 10 . . 14 ...... 3 . . . . . 1 ...... 1 Silene acaulis ...... 16 ...... Saxifraga oppositifolia ...... 15 ...... Brachythecium turgidum ...... 1 17 . . 1 1 ...... Distichium capillaceum ...... 2 17 ...... Kobresia simpliciuscula ...... 3 16 . . 1 ...... Bartsia alpina . . 1 . . . . 4 . 1 29 . . 8 1 1 . 10 1 . 1 1 ...... 6 1 . 1 ...... Hylocomiastrum pyrenaicum ...... 1 17 . . 2 . . . 2 ...... 1 ...... Barbilophozia quadriloba ...... 1 18 . . 4 . . . 3 ...... 1 ...... Carex microglochin ...... 1 16 1 1 1 2 ...... Juncus castaneus ...... 15 1 . 1 3 ...... 1 ...... Pinguicula alpina 4 . 1 . . . . 6 . 1 20 1 . 2 1 ...... 2 ...... Equisetum variegatum 4 . 13 10 . 4 3 15 . 7 45 3 . 17 18 1 7 22 1 . 1 . . . 6 . . . . 1 . 3 ...... 1 Carex bigelowii ...... 6 28 1 . 1 3 . . 2 ...... 2 6 12 1 . . . . . 1 6 . . . . 1 . . 3 3 . Euphrasia minima ...... 12 . . 3 ...... Carex saxatilis ...... 2 29 8 . 3 13 . . 4 ...... 1 20 ...... 1 . . . 1 . . Pseudocalliergon trifarium . . 2 . . . . 4 . 1 6 47 35 32 12 8 1 4 1 . . . . . 1 . 1 . 3 . 1 3 ...... Utricularia intermedia agg...... 2 . . . 2 51 3 1 2 . 1 . 2 . . . 4 1 12 8 6 15 . 1 . 2 1 . 1 1 3 . . . 1 1 ...... Carex livida ...... 3 36 4 . . . 1 . . . . . 15 . . . . 4 ...... 1 1 ...... 4 . . 1 . . Utricularia minor agg. 1 . 1 . . . . 1 . . . 13 40 4 10 32 . 1 1 8 1 . . 6 7 2 3 2 7 . 1 . 4 1 . 1 1 1 . . . 1 1 ...... Cinclidium stygium ...... 4 2 . 15 11 10 44 36 . 1 18 ...... 2 6 8 . 3 . 2 13 ...... 1 . . Sphagnum contortum . 1 2 2 . . 1 . . . . 3 7 10 1 64 9 3 36 30 7 5 1 3 20 . 4 3 7 . 1 3 2 . 1 2 1 1 . 1 1 . . . . 2 . 2 . . Trichophorum alpinum . . 1 . . . . 2 . . 2 9 15 40 3 52 2 29 13 15 2 1 . 4 1 . . . 9 10 1 1 2 . . 1 3 4 1 1 ...... 4 . . Juncus bulbosus ...... 1 . 2 . 1 . 37 . . 5 20 2 8 1 1 1 . 3 5 3 . 1 . 19 5 3 2 1 2 1 . . 1 . . . 2 . . . . Tomentypnum nitens 3 10 36 28 30 33 17 7 . 1 30 1 . 11 4 36 95 56 39 15 8 11 7 . 10 . 1 . 1 . . . . . 1 1 . 4 ...... 1 Paludella squarrosa . . 1 . 49 . . . . 1 9 1 1 15 9 3 16 62 13 4 2 . 1 2 1 . 4 1 3 . 1 42 . . . . 2 2 . . 5 . . 3 1 . . 4 . . Betula nana ...... 2 . 2 45 3 3 31 7 1 . 60 . . 1 . . 1 . 2 5 . 11 . 3 15 . 1 . 1 2 3 . 1 12 . 3 7 11 2 . 22 . 24 Sphagnum teres . 1 1 4 32 8 1 . . 5 . 1 4 3 1 11 15 26 58 19 100 41 6 4 27 2 11 13 7 10 4 29 . 1 7 18 4 15 1 2 11 . . . . . 1 5 . . Rhytidiadelphus squarrosus 5 7 1 30 . 2 5 1 2 1 1 . . 1 1 1 8 . 3 1 8 47 7 . 1 . . 3 . 3 . 1 1 . 4 1 . 1 1 1 ...... 3 Myosotis palustris agg. 3 2 1 22 . 41 11 . . 4 . . . . 1 1 13 . 13 4 21 43 60 . 25 . 1 5 . 23 1 . . . 4 11 1 1 . 1 1 ...... Silene flos-cuculi 2 5 1 20 16 42 14 . . 2 ...... 19 . 10 3 22 37 56 . 21 . 2 6 . . 1 . . . 2 6 . . . 1 ...... Cluster No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 Carex diandra . . 3 1 . 7 . 1 . . . 2 2 . 2 7 8 1 3 4 15 . 2 3 60 17 3 4 1 . 1 1 1 1 1 5 1 . 1 . 4 ...... Carex aquatilis ...... 3 1 5 1 3 4 . . 5 ...... 62 . . . . 5 10 ...... 11 . 1 3 3 2 . . . 1 Cicuta virosa ...... 1 28 . 2 ...... 1 ...... Galium trifidum ...... 1 . 1 ...... 2 . . . . 26 2 . . . 1 . . . . 1 ...... Stellaria crassifolia ...... 1 . 1 ...... 1 . . . . 25 3 ...... Calamagrostis neglecta ...... 11 10 2 . 1 9 . . 8 . . 1 . . . 1 48 3 1 . . 9 14 . . . 1 1 . . . 3 ...... Lysimachia thyrsiflora ...... 1 . . . 1 1 . 1 2 . 1 . 3 5 . 1 . 6 48 10 34 2 . 3 . 4 13 . 10 1 6 1 . 11 1 ...... Sparganium natans ...... 14 . 1 ...... 1 ...... Drepanocladus polygamus . . . 1 . . . 1 . 1 . . 1 1 1 1 . . . 1 . 1 . . . 18 2 1 ...... Calla palustris . . . 1 . 5 ...... 1 . . . . . 4 1 1 1 1 25 1 1 . 4 1 . . 2 1 4 2 1 . . 5 ...... Carex lasiocarpa . 1 1 . . . . 16 2 . 3 17 71 39 14 10 6 35 8 6 16 . . 16 21 . 7 28 100 . 9 2 12 6 2 6 10 27 2 . 3 2 2 3 4 . 2 8 1 2 Philonotis seriata 1 ...... 1 ...... 57 9 3 1 . 1 . . . . 1 ...... Epilobium alsinifolium ...... 1 . . 43 1 ...... 1 ...... 2 . . . Ligusticum mutellina ...... 1 . . . . 1 . . . . 1 . 1 1 ...... 53 3 2 . . . . . 1 1 12 ...... Viola biflora 1 ...... 1 21 . . 5 2 . . 4 1 1 . 1 1 ...... 57 3 1 ...... Aconitum napellus 1 ...... 1 . 1 . 1 ...... 33 . . . . . 2 . . . 1 ...... Allium schoenoprasum . . 1 ...... 1 ...... 3 ...... 26 . . 1 ...... Scapania undulata ...... 1 1 ...... 25 2 1 ...... 1 ...... Dichodontium palustre . . 1 ...... 1 . . 1 1 ...... 26 3 1 . . 1 ...... Palustriella decipiens 2 2 1 6 . 2 2 5 . . 1 . . 1 . . 1 1 . . . . 3 . . . 1 . . 32 1 ...... 1 ...... Jacobaea subalpina . . 1 ...... 1 . 1 . 1 1 . 2 2 . 1 . . . . 23 3 . . . 1 1 ...... Alchemilla vulgaris agg. 6 . 2 35 . 40 5 . . 4 14 . . 1 1 . 11 2 3 1 1 13 21 . 1 . . . . 58 . 1 . . 2 1 . 1 ...... Rhizomnium magnifolium ...... 1 1 . . . . 1 2 17 ...... 1 ...... Deschampsia cespitosa 23 14 13 4 11 15 37 5 7 19 38 2 . 3 5 . 5 8 2 2 9 22 31 1 6 . 5 10 . 82 16 9 1 1 16 14 2 5 13 14 8 ...... 1 1 . Luzula alpinopilosa ...... 1 ...... 16 2 1 ...... 2 ...... Cardamine amara 2 1 1 2 . 1 1 ...... 1 1 2 4 7 . 2 . 1 . . 25 1 . . . 1 1 . . . 1 ...... Silene pusilla 1 ...... 1 . . . . . 1 ...... 1 ...... 14 ...... 1 ...... Rumex alpestris ...... 1 . . 1 . . 1 . . . . 13 . . . . 1 . . . . 1 ...... Swertia perennis 2 . 6 . . 7 . . . 1 . . . . 1 . 4 . 1 2 1 1 1 . . . 1 . . 23 2 3 1 1 . . . . 1 ...... Warnstorfia exannulata . . . 1 . . . . . 1 4 16 10 3 7 5 1 11 6 17 7 4 2 37 13 6 26 26 19 18 100 56 9 9 4 5 1 4 1 1 13 . 3 27 40 2 1 12 . . Eriophorum scheuchzeri ...... 7 . . . . 1 ...... 1 18 3 ...... Warnstorfia sarmentosa . . . 1 . 2 . 1 . 1 11 20 1 24 20 2 1 17 1 3 1 . 1 6 . . 3 . 4 9 9 89 2 . 1 1 . . . 3 1 . 1 1 6 5 . 19 1 . Sphagnum auriculatum agg...... 1 . . . . 2 2 1 1 1 1 1 5 1 4 . 8 2 . 8 16 12 3 2 1 53 4 2 1 1 4 . . . 1 1 . 1 2 . 6 . . Sphagnum papillosum ...... 1 1 1 1 . 1 2 5 1 . . 8 1 . 3 1 18 1 1 . 57 15 4 2 8 4 5 3 . 2 18 11 9 2 3 31 11 2 Rhynchospora alba ...... 1 . . . . 2 29 . 1 19 . 1 1 23 . . . 4 . . 3 1 10 . . . 48 4 1 . 3 6 1 . . 31 24 4 . . . 1 1 1 Sphagnum palustre agg. . . . 2 . . 2 . . 1 . . . . . 8 4 6 24 24 4 27 3 2 . . 3 26 1 1 1 . 5 7 32 10 5 100 7 8 1 1 . 1 . . 1 2 . 1 Polytrichum commune . . . 1 . . . . . 2 . . . . 1 1 . 3 2 8 9 12 5 1 2 3 6 21 2 10 14 2 18 73 50 24 10 25 87 62 13 5 2 1 3 5 5 . 16 4 Vaccinium myrtillus . . 1 ...... 1 . . 1 1 2 . 4 . . 2 . 1 1 . 6 1 . . 5 6 1 3 11 28 53 1 1 . . . . . 3 41 27 Sphagnum girgensohnii . . . 1 ...... 1 . . 4 1 3 2 1 2 . 3 1 . 4 2 1 . . 1 1 1 1 5 28 ...... 1 4 2 Avenella flexuosa . . . 1 . . . . . 4 15 . . . . 1 . 3 1 1 1 4 1 . 1 . 1 2 . 12 1 . 2 2 11 2 1 3 26 39 1 ...... 7 2 Sphagnum riparium ...... 1 ...... 1 . . 2 1 . . . . 2 . 1 . 3 3 . 6 . 3 5 . 7 1 100 . . 15 15 2 . . . . Sphagnum cuspidatum ...... 3 . 1 . 3 . . 2 7 4 . 1 1 14 10 1 2 2 5 4 3 1 100 10 3 . 12 50 4 11 1 Sphagnum majus ...... 1 . . 12 . . 3 2 7 . 6 1 5 1 . 1 4 . 1 . . 4 100 37 17 9 7 12 2 . Scheuchzeria palustris ...... 1 17 1 . . . 1 1 2 . . . 11 . . 4 . 5 . 1 . 3 . 1 . 14 . . . . 27 66 33 . 2 13 2 1 1 Sphagnum annulatum agg...... 3 . . . . 1 . 1 1 ...... 1 21 8 . . 1 . . Carex rotundata ...... 1 . 3 1 1 1 . . 1 . . . . . 9 ...... 1 4 ...... 4 . . 1 57 9 . 6 . 1 Eriophorum russeolum ...... 1 ...... 1 . . 1 . . . 1 ...... 1 . . 1 . . . 17 4 . 1 . . Sphagnum compactum ...... 1 . . . . 1 . . 1 . 9 . . . . . 1 3 2 1 . 1 . . . . 17 1 . 1 1 22 . 1 36 2 1 Sphagnum tenellum ...... 1 . . . 2 . 10 . . . 1 . 1 . . 12 6 5 2 12 2 37 11 2 Vaccinium uliginosum ...... 1 34 1 1 17 5 . . 44 1 4 2 2 . . . . 3 . 3 . 1 6 6 3 2 1 3 1 18 21 1 2 2 4 3 7 10 17 78 53 Cluster No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 Eriophorum vaginatum ...... 1 . 1 6 . 1 3 3 1 1 10 . 5 2 2 1 2 . . 6 1 2 3 9 8 13 9 12 3 29 6 63 50 9 34 20 16 5 33 37 31 96 59 Carex pauciflora ...... 2 . 1 . 6 1 5 1 . 1 . . . 2 . 3 . 1 2 5 1 1 1 6 2 11 10 1 1 3 4 2 4 6 12 45 1 Sphagnum russowii . . 1 ...... 1 . . . . 1 1 . 8 1 1 2 4 . 1 1 . 2 1 1 3 3 1 1 1 6 1 3 6 13 26 3 1 . 4 . 7 7 6 40 5 Pleurozium schreberi . 1 ...... 1 1 . 1 . 2 2 8 1 3 . 7 1 . 1 . . . 1 2 . 1 . . 1 1 1 3 4 13 . . . . . 2 . 1 11 77 Rubus chamaemorus ...... 3 . . 1 1 1 . . 19 ...... 2 . . . . . 1 . 1 . . 2 . 1 3 5 . . 8 11 9 . 14 1 74 Racomitrium lanuginosum ...... 1 3 . . . . 1 . . . 1 ...... 1 . . . . 1 . 1 ...... 2 . 33 Empetrum nigrum agg. . 1 . . . . . 1 11 4 12 1 . 3 . . . 30 . 1 ...... 1 . 1 1 2 1 1 . 1 1 2 5 . 2 . 3 1 . 1 14 44 72 Dicranum scoparium ...... 1 ...... 5 . 1 . 1 1 ...... 2 1 ...... 1 3 . . . . 1 . . . 5 21 Hylocomium splendens . 1 . 1 . . . . 2 1 17 . . . . 1 1 10 2 3 . 1 1 ...... 3 ...... 2 . 7 ...... 1 . 30 Ptilidium ciliare ...... 1 6 . 1 1 . . . 10 . . 1 ...... 1 . . 1 ...... 1 . 1 . 1 9 2 1 18 1 30 Eupatorium cannabinum 61 58 18 13 . 1 13 7 . 4 . . . . 1 . 2 . . 1 . . 1 . 1 . . 1 ...... Juncus inflexus 34 34 8 8 . 1 17 . . 5 ...... 1 . . . . . 1 ...... 1 ...... Primula farinosa 12 . 57 2 . 76 18 46 . 4 . 2 . . 3 1 18 . 2 . . . . . 2 ...... Dactylorhiza majalis 22 18 38 74 73 15 31 . . 1 . . . . 1 6 45 . 24 7 17 29 21 . 11 . . . 1 2 . . . . 12 6 1 2 ...... Climacium dendroides 2 12 10 72 73 26 27 . 2 2 7 . . . 1 2 38 3 16 4 15 40 61 . 18 . 2 13 . 9 . . 1 . 2 1 . . 1 ...... Carex demissa 1 3 1 . 95 1 1 . 13 . . 9 . 1 1 58 6 . 18 28 4 14 4 . 4 . . 3 . . . . 2 . 6 1 1 1 ...... 1 Thalictrum alpinum 1 ...... 4 76 . . 58 4 1 . 36 . 1 ...... 1 ...... 1 ...... 2 Selaginella selaginoides 1 . 1 . . . . 11 26 3 59 . 1 62 7 1 . 31 1 4 ...... 1 10 . . 2 . . . . . 1 ...... 9 . 1 Scorpidium scorpioides 1 . 2 . . . 1 25 . . 2 100 99 31 22 15 1 5 . 4 . . . 9 2 6 13 . 27 . 9 16 ...... 1 . 2 . . 2 . 1 . . Trichophorum cespitosum . . 1 . . . 1 13 . 1 18 12 4 92 3 1 . 32 1 1 1 . . 13 . . 7 . 5 1 1 9 3 1 . 1 1 . 3 10 . 3 9 5 14 7 5 96 14 14 Sphagnum warnstorfii . . 4 7 49 . 1 1 . . 9 1 4 19 1 31 50 96 100 6 10 19 2 . 7 2 4 1 2 3 3 15 2 1 5 3 1 12 . 1 1 . . . 1 . 2 6 1 1 Carex limosa . . 5 ...... 38 68 15 9 5 . 8 2 6 5 . 1 95 9 5 8 3 22 . 9 5 3 3 1 3 34 4 2 . 3 27 54 48 17 2 100 17 8 . Sphagnum lindbergii ...... 3 . . 1 . 3 . 8 6 . 1 . . 5 . 1 . 5 1 23 100 100 21 5 32 . . Warnstorfia fluitans . . . 1 . . . . . 1 . 1 3 . . 1 . . 1 5 . 2 5 3 1 35 10 5 8 . . . 2 5 1 1 10 . 7 5 5 21 30 26 26 100 100 15 13 2 Calluna vulgaris . . . . . 2 1 1 2 . . . 1 3 . 5 5 5 6 10 1 5 . . 1 . 2 . 1 2 1 . 21 5 4 . 2 11 14 27 . 3 3 . . . 5 14 79 80 Carex flacca 74 62 41 9 73 14 44 9 93 18 . 1 . . 1 1 10 . 1 . . 2 2 ...... 6 ...... Carex davalliana 44 14 79 23 100 92 92 22 . 1 . 1 . . 4 9 65 . 11 6 2 6 8 . 12 ...... 1 ......

Other species: Eriophorum angustifolium 66 75 63 77 . 52 23 20 2 4 31 79 28 50 57 86 63 42 73 82 62 72 65 74 52 37 39 67 29 28 68 91 76 96 77 58 33 48 18 26 44 23 29 30 67 42 21 57 5 8 Potentilla erecta 69 73 93 94 84 92 82 78 96 7 6 . 1 27 10 81 92 14 96 76 53 96 68 2 29 . 3 9 9 19 1 1 22 14 91 46 7 55 26 40 ...... 3 6 1 4 Carex nigra 18 33 47 85 16 80 64 9 65 7 12 18 4 22 27 40 63 28 59 48 71 90 94 3 68 62 21 59 5 33 31 15 10 33 91 69 19 42 56 54 17 7 1 1 2 14 6 . 9 1 Carex panicea 77 92 91 95 89 82 90 55 93 7 7 26 17 40 31 98 86 15 89 85 44 85 73 2 45 2 2 14 5 8 4 . 8 1 40 24 4 16 1 6 . 1 1 . . . 5 1 2 1 Carex rostrata 8 5 36 12 73 18 14 8 . 1 2 53 46 50 28 58 54 41 58 49 54 19 24 44 82 17 100 44 51 2 16 38 17 48 12 89 76 30 13 8 51 7 39 63 13 14 15 18 4 2 Campylium stellatum agg. 69 67 85 82 89 45 59 94 33 25 84 25 23 89 48 99 52 35 43 40 5 8 10 1 29 5 3 1 20 22 1 3 1 1 1 1 1 . . 1 ...... 4 . . Sphagnum recurvum agg. . 2 1 2 11 ...... 1 1 12 18 17 52 21 30 33 7 5 18 . 5 21 14 2 1 1 21 100 99 100 100 51 100 17 35 10 7 8 1 16 21 7 77 5 Scorpidium revolvens agg. 25 3 99 65 24 80 21 65 26 1 88 46 21 100 100 90 38 27 19 10 . 2 19 8 21 2 8 1 13 4 6 32 2 . 1 1 . 1 . 1 ...... 8 . . Equisetum palustre 54 55 69 77 95 83 42 6 17 7 12 15 6 32 29 26 68 36 40 12 33 23 43 1 53 6 9 8 7 10 4 3 . 1 21 23 2 12 1 1 5 2 . . . . . 2 . 3 Valeriana dioica 53 56 66 65 84 72 82 13 . 1 . 1 . . 4 47 77 . 62 23 48 51 62 . 50 . 2 5 . . . . . 1 8 19 1 11 ...... Bryum pseudotriquetrum 83 51 82 90 68 35 17 18 63 16 35 8 5 21 29 50 62 18 34 22 16 14 32 2 44 12 6 6 2 63 6 6 . 1 1 1 . 1 . 2 ...... 1 . Molinia caerulea agg. 31 24 40 1 22 1 66 88 89 9 14 5 3 48 11 21 27 18 32 52 18 14 13 1 10 . 2 32 19 16 1 2 73 37 6 14 4 38 52 31 1 3 1 1 . 2 5 20 31 6 Straminergon stramineum . . . 1 41 . . . . 9 4 2 3 9 5 4 9 57 58 35 65 32 6 11 24 5 26 38 22 5 31 53 26 34 49 44 28 39 20 8 37 2 2 33 52 16 7 33 14 4 Agrostis canina 3 1 5 30 . 4 7 1 . 4 . 2 . 1 2 38 25 2 55 43 55 70 52 . 44 . 8 75 8 10 6 1 16 39 74 73 3 41 9 14 4 2 1 . . . 1 . . 1 Viola palustris . . 3 6 46 19 3 . . 4 1 3 . 9 5 47 17 14 64 55 63 72 47 2 41 . 14 43 3 5 6 3 10 19 67 75 9 33 12 9 11 . . . . . 3 3 . 1 Vaccinium oxycoccos agg. . . 1 . . . . 6 . . . 7 18 18 6 34 8 52 34 23 18 6 2 15 8 . 17 7 35 . 3 4 33 40 16 17 74 34 44 28 15 33 31 23 11 18 41 34 89 29 Festuca rubra agg. 27 57 23 69 68 49 19 4 20 36 24 . . 9 8 10 52 8 52 11 40 67 27 . 18 3 3 10 . 3 . . . 2 56 36 1 16 3 12 1 ...... Aulacomnium palustre . 1 15 40 68 21 21 2 . 4 7 . 1 4 1 25 80 46 72 47 44 72 24 . 17 2 6 41 4 3 4 1 12 10 23 19 6 35 7 8 ...... 1 14 7 Cirsium palustre 30 21 42 55 . 18 24 8 9 4 1 2 . . 4 46 56 1 58 26 41 81 49 1 33 . 5 23 . 2 1 . 1 1 35 32 1 16 3 4 ...... 3 . . . Carex echinata 2 8 3 54 11 21 6 1 9 . . 2 . 3 1 62 19 1 55 68 33 79 56 2 13 2 3 7 4 30 9 5 19 9 67 33 8 20 14 15 1 . 2 . . . 3 6 1 . Menyanthes trifoliata 2 1 20 12 22 8 10 19 . . 1 32 64 44 16 45 17 31 33 28 34 7 15 44 73 35 23 9 34 . 6 8 7 1 . 30 27 15 2 . 24 1 26 25 12 . 1 12 1 1 Cluster No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 Briza media 44 47 44 91 81 63 68 9 89 14 . . . . 1 22 74 . 47 8 15 52 38 . 5 . . 1 . 6 . . . . 18 5 . 7 . 1 ...... Equisetum fluviatile 16 8 18 37 . 7 22 8 2 6 . 15 57 14 10 36 31 16 39 28 42 31 18 15 48 54 36 38 40 3 9 1 6 1 19 38 10 9 1 3 8 1 2 1 1 2 . 3 . 2 Comarum palustre . 1 . . . . 1 5 . 4 2 10 6 17 21 16 4 34 23 22 52 10 13 6 59 85 54 79 30 1 19 33 11 22 5 64 11 25 2 2 56 1 . 1 1 2 . 2 1 2 Drosera rotundifolia . . 2 3 89 . 4 6 . 1 . 5 6 11 2 71 22 19 54 77 17 10 5 14 11 . 4 15 16 2 1 . 57 38 28 12 25 49 11 6 1 12 13 15 . . 20 24 28 5 Galium uliginosum 2 1 18 26 . 42 30 1 46 1 . 1 . . 4 26 72 . 68 15 60 68 44 . 44 . 7 28 1 2 . . . . 14 31 1 15 1 1 ...... 1 . . . Plagiomnium affine agg. 43 71 54 84 24 52 49 8 9 13 2 2 . 1 5 5 51 11 24 3 27 35 31 . 49 3 7 5 . 5 1 . . . 3 4 . 1 1 1 ...... Lysimachia vulgaris 15 47 6 28 43 3 39 2 . 7 . 1 . . 1 14 23 . 39 28 38 38 25 . 38 . 7 74 2 . 1 . 8 17 23 43 2 38 2 3 3 1 ...... Juncus articulatus 67 49 52 64 22 23 34 28 67 6 . 2 1 1 7 56 34 . 26 34 15 24 27 2 19 . 3 6 . 11 1 . 1 1 5 4 1 6 1 1 ...... Crepis paludosa 29 33 38 73 81 52 21 . . 3 4 . . 1 2 9 58 1 41 8 27 48 27 . 15 . 2 1 . 18 1 . . . 30 12 2 4 1 2 ...... Epilobium palustre 6 7 3 23 . . 5 2 . 10 2 1 1 1 5 7 22 11 29 9 56 49 27 . 36 43 31 47 4 10 14 6 2 2 27 45 2 13 4 3 16 . . . 1 . 2 . . . Andromeda polifolia ...... 7 . 2 13 17 24 61 8 3 . 62 1 6 1 1 . 10 1 . 9 . 30 . 3 13 14 1 1 1 17 6 6 3 5 18 24 41 35 16 13 73 46 76 Anthoxanthum odoratum agg. 1 7 3 33 27 33 16 . . 4 16 . . 1 . 12 26 1 40 10 24 74 42 . 12 . 1 3 . 22 2 . . 1 51 14 1 18 8 14 ...... 1 1 . Carex canescens . . . 1 . 6 . . . 7 . 3 . . 7 2 1 4 8 10 40 15 30 5 32 40 32 44 3 6 36 28 6 33 10 48 20 4 22 8 36 5 2 1 11 16 13 . 4 . Caltha palustris 9 12 9 49 22 60 39 1 11 3 1 2 . . 5 2 16 1 8 5 23 19 53 1 45 32 8 3 . 56 14 4 . . 6 13 2 1 . 1 3 ...... Aneura pinguis 26 12 27 22 . . 3 20 48 5 29 14 30 24 22 58 16 13 25 28 5 6 6 3 10 . 3 3 8 13 1 4 3 1 2 1 1 1 . 1 ...... 2 . . Nardus stricta 2 1 1 8 . 7 1 . . . 13 . . 8 2 13 5 6 18 20 4 59 18 1 . . 1 . . 56 9 2 4 3 63 10 1 20 10 52 ...... 1 1 5 1 Pinguicula vulgaris 14 . 62 16 . 52 12 46 72 4 46 11 . 40 17 10 20 12 6 6 . 2 6 . . . 3 . 2 16 3 2 1 . 1 1 . . 1 ...... 4 . 2 Triglochin palustris 52 20 56 27 11 17 13 16 13 12 5 15 3 17 29 25 25 6 13 12 2 2 7 1 13 5 1 1 1 3 1 1 2 . 1 3 1 1 ...... Filipendula ulmaria 6 8 12 19 59 65 25 1 37 1 3 . . 6 6 5 50 12 25 3 24 15 42 . 23 5 3 7 . 2 . . 1 . 2 9 1 4 ...... 1 Cirsium rivulare 39 54 28 61 54 59 42 . . 1 . . . . 3 3 22 . 4 1 2 12 20 . 4 ...... 1 1 ...... Equisetum arvense 40 51 16 41 41 4 13 2 52 25 15 1 . 2 13 1 6 5 5 3 5 16 8 . 1 . 1 1 . 10 4 8 . . 2 2 . 2 . 1 ...... 1 . . Cardamine pratensis agg. 7 3 8 13 . 47 14 5 . 4 . 1 . . 4 3 26 1 14 1 26 27 35 . 40 2 5 11 . 10 4 1 . . 3 12 . 1 . . 3 . . . . . 1 . . . Angelica sylvestris 4 8 10 13 11 12 14 . 20 3 3 . . 6 3 2 37 14 23 3 20 44 15 . 17 . 2 1 1 1 . . . . 9 9 . 6 . 1 ...... 1 . . . Linum catharticum 28 25 22 48 32 22 27 16 30 15 . 1 . . 1 38 19 . 12 12 . 6 6 . 1 . . . . 5 . . . . . 1 1 1 ...... Juncus effusus 2 2 . 6 . . 8 . . 2 . . . . . 5 3 . 5 9 12 44 24 . 8 . 8 38 . 3 . . 1 16 33 20 2 9 11 5 1 1 . . . 2 . . . . Rumex acetosa 2 3 1 22 . 25 12 1 . 11 1 . . . 1 . 16 3 14 1 26 49 31 1 10 . 2 4 . 3 1 . . . 14 16 . 7 1 2 ...... Juncus filiformis . . 1 ...... 5 1 2 . 4 3 1 . 2 1 2 10 11 15 . 1 2 4 33 . 33 16 7 3 16 16 22 6 3 26 30 12 2 . 1 5 19 3 . 2 . Agrostis stolonifera agg. 26 44 11 13 . 11 14 6 4 15 3 . . 1 4 1 7 . 6 5 7 9 12 . 6 38 3 5 1 12 2 . 2 3 10 14 3 5 3 4 ...... Phragmites australis 6 7 5 1 30 3 21 25 7 6 . 8 13 3 8 17 14 1 10 14 12 1 6 2 12 . 4 15 14 1 1 . 19 10 2 3 1 23 2 2 . 1 . . . . . 2 . . Pedicularis palustris . 1 6 11 . 51 2 13 7 2 . 8 23 13 9 23 5 9 7 10 1 3 21 3 17 40 5 2 8 . 1 1 . . 2 2 . 2 . . 1 ...... 1 . . Mentha arvensis agg. 8 20 12 28 . 6 7 1 20 5 . . 1 . 1 16 8 . 13 9 13 28 21 1 13 2 3 7 . . 1 . . . 4 3 . 1 ...... Polytrichum strictum . . 1 . . 2 ...... 4 7 7 16 12 2 7 5 2 7 . 3 5 4 1 4 3 11 3 8 6 5 28 6 23 . 1 1 1 4 2 3 5 39 5 Equisetum sylvaticum 1 . 1 9 . . . . . 3 4 . . 1 2 6 1 3 14 6 5 34 4 . 14 2 3 3 . 7 3 5 1 1 31 9 1 6 5 8 5 . . . 1 . . 1 . . Picea abies 16 2 11 12 . 2 3 1 . 1 . 1 1 2 1 8 7 1 10 9 4 8 4 . 2 . . . . 2 . . 1 6 10 6 1 9 18 13 . . . . . 2 2 2 26 2 Lythrum salicaria 21 28 5 11 . 2 36 2 2 4 . . . . 1 4 7 . 3 10 7 2 10 . 13 3 5 34 . . . . 1 1 . 5 1 6 ...... Carex chordorrhiza ...... 1 . 26 28 14 16 3 1 26 2 3 4 . 1 30 6 3 9 3 10 . 9 7 . . . 2 9 5 1 . 3 . 3 4 . . . 8 . 1 Scirpus sylvaticus 10 20 2 18 49 15 8 . . 2 . . . . . 1 9 . 7 2 7 15 29 . 7 . 2 7 . 2 1 . . . 9 7 . 2 2 ...... Calliergon giganteum 1 1 6 9 27 3 . . . 1 2 9 6 5 19 18 10 4 9 4 1 1 6 2 26 40 8 6 1 2 11 2 . . 1 1 1 ...... Peucedanum palustre . . 1 1 . . 1 2 . . . . 2 . 2 4 2 . 8 13 14 4 5 . 13 9 8 54 5 . 1 . 5 10 2 14 2 21 3 2 4 . . . . . 1 . . . Trientalis europaea ...... 1 1 . . 2 . 1 . 11 8 9 1 2 1 . 1 2 3 1 2 4 1 . 4 12 20 13 6 7 32 17 7 1 . . . . 2 4 4 2 Sphagnum subsecundum . . . 2 . 2 1 1 . . . 2 4 5 3 4 3 7 7 35 3 7 11 9 4 . 8 22 33 1 6 4 6 1 5 1 2 4 . . . . 3 3 . 2 . 11 . . Bistorta officinalis 1 . 2 3 . 13 1 . . 1 . . . . . 1 4 . 8 3 14 21 12 . 7 . 2 4 . 22 1 1 1 1 14 19 1 5 6 12 1 ...... Pinus sylvestris 6 7 8 . . . 2 9 . 4 . 1 1 3 1 7 3 8 2 10 4 . 1 . 1 . 3 12 2 . 1 . 31 25 1 2 4 15 5 8 . 4 3 4 . . . 1 1 5 Vaccinium vitis-idaea . . 1 . . 1 . . . 1 . . . . 1 1 4 5 4 3 . 2 3 . . . 2 . . 5 1 . 1 4 7 1 4 7 27 37 . 1 ...... 32 23 Drosera longifolia . . 3 . . 1 . 8 . . . 14 43 13 4 13 . 6 2 8 1 . . 16 . . 3 . 10 . 1 1 3 1 . . 1 1 . . . 13 14 15 1 . . 17 . . Salix lapponum ...... 7 13 8 . 11 14 1 . 26 . . 1 . . 1 . 6 15 . 3 1 14 39 . 1 . . 1 . . . 16 . . 3 6 . . 1 . 1 Cruciata glabra 3 18 6 39 . 40 7 . . 16 . . . . . 1 19 . 3 1 1 4 11 . 1 . . . . 2 . . . . 1 1 . 3 ...... Philonotis fontana 2 3 4 25 . 2 1 . 9 7 7 1 . 1 3 10 5 1 12 9 4 14 18 . 4 . 1 1 . 10 6 4 . . 1 2 . . 1 ...... Lycopus europaeus 5 9 3 3 . . 16 4 . 4 . . . . 2 7 5 . 4 14 5 5 9 . 20 . 3 26 1 . . . 1 1 1 6 . 2 ...... 2 . . . Sphagnum capillifolium . . 1 . . 3 2 . . 1 . . . . . 1 11 . 4 9 2 7 5 . 2 . 2 . . 1 5 . 1 1 8 2 . 16 4 32 1 3 1 . . 5 7 1 19 6 Juncus conglomeratus 1 2 1 8 . 11 5 8 2 1 . . . . 1 4 4 . 6 6 4 24 29 . 5 . 1 2 . . 1 . . 1 6 4 . 4 1 1 ...... Cluster No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 Ranunculus flammula . 1 . 2 . 1 8 . . 1 . . 1 . 1 11 2 . 5 14 3 14 29 2 15 5 6 13 . . 4 . . . 4 4 ...... 2 . . . . Salix aurita 1 1 5 2 . . 1 . . . . 1 . . . 5 15 . 16 8 13 9 5 1 10 . . 9 . . 1 . . 3 7 9 2 4 2 3 ...... 2 . . . Hamatocaulis vernicosus 3 . 5 2 . 12 3 ...... 1 19 17 . 18 14 7 2 1 1 29 . . 5 . . 2 . . . 1 ...... Betula pubescens . . 5 1 . . . 2 . 4 4 1 4 15 2 2 4 13 4 8 4 3 1 . 1 8 2 4 2 1 2 1 3 4 3 5 4 9 4 1 1 ...... 1 6 2 Ranunculus auricomus agg. . . . 2 . 17 7 . . 1 . . . . . 3 9 . 10 1 16 26 23 . 5 . 1 1 ...... 1 3 . 1 . 1 ...... Loeskypnum badium ...... 8 . 1 11 4 2 30 1 1 . 28 1 1 . . . . . 2 1 . 4 . 1 9 ...... 3 4 . . 31 . . Alnus glutinosa 7 10 8 3 . . 7 1 7 1 . . . . 1 6 9 . 6 17 6 5 4 . 10 . 1 7 1 . . . 2 2 3 5 . 5 . 1 ...... Betula pendula 4 10 5 6 . . 1 . . 1 . 1 . . . 1 2 1 10 10 2 6 4 . 6 . . 11 2 . . . 3 7 5 4 1 8 6 3 1 . . . . . 1 1 4 . Tephroseris crispa 1 . 1 1 11 7 2 ...... 2 13 . 19 1 15 17 17 . 9 . . . . 5 . . . . 3 4 . 1 ...... Salix cinerea 9 6 5 . . 15 9 2 2 1 . . . . 1 3 4 1 6 4 7 3 4 . 9 . . 18 . . 1 . . 1 3 6 1 6 . 1 . . 1 ...... Sphagnum magellanicum ...... 2 1 1 4 2 1 3 2 . . . 1 1 1 3 2 . 4 5 7 2 10 9 16 12 . 6 2 10 . 2 2 1 19 2 Blysmus compressus 18 7 27 20 . 5 4 6 . 7 . . . . 2 1 11 . . 1 . . 3 ...... 10 ...... Carex magellanica ...... 3 . 4 3 . . 5 . . 2 . . 2 . 3 9 . 2 . 15 21 1 1 1 1 7 . 1 1 17 1 8 19 28 25 . 3 . . Frangula alnus 6 1 9 1 . . 5 1 2 1 . . . . . 6 10 . 8 10 2 4 2 . 2 . . 10 1 . . . 4 4 3 4 . 16 5 4 ...... 1 . Gymnocolea inflata ...... 1 . . . . 1 . . . . 31 . . . . 1 1 6 1 4 1 . . 2 . 3 2 . 5 5 1 27 16 19 31 23 . Achillea millefolium agg. 6 10 2 17 . 11 16 1 9 17 ...... 8 . 3 . 1 11 6 . 1 . . . . 7 . . . . 1 . . 1 1 1 ...... Carex paniculata 17 14 18 3 . 22 6 1 . 1 . . . . 1 . 16 1 2 1 2 . 6 . 3 ...... 1 1 1 ...... Scutellaria galericulata . . . 1 . . 7 1 ...... 1 1 2 . 5 4 16 7 12 . 14 2 3 16 1 . . . 1 . 4 14 . 4 ...... Plantago lanceolata 3 7 3 27 . 16 17 . . 8 ...... 3 . 1 3 1 12 7 ...... 2 . . . . 1 ...... Ranunculus repens 6 14 2 3 . 11 14 2 . 4 ...... 3 1 1 1 4 6 16 . 7 2 1 3 . 9 1 . . . 2 1 . . . 1 ...... Danthonia decumbens 3 7 1 8 8 2 7 1 37 1 . . . . . 5 4 . 7 15 1 9 2 ...... 9 . . 5 . 3 ...... 1 Cratoneuron filicinum 13 29 11 6 . 13 4 1 4 10 1 . . . 1 1 4 . . 1 . . 5 . 2 . 1 . . 8 . . . . . 1 ...... Sphagnum subnitens . . 1 2 . . 2 1 . 1 . . 1 4 . 8 8 8 6 21 1 3 2 . 1 . . 1 11 . 1 . 6 1 1 2 1 2 . . . . . 1 . . . 9 1 1 Juncus alpinoarticulatus 2 . 13 2 . . 1 2 . 8 3 5 1 4 8 18 2 . 4 11 . . 2 1 4 2 2 1 . . 1 1 . . 1 . . 1 ...... Calamagrostis canescens 1 ...... 1 . . . . 1 . . 1 3 1 9 . 2 . 6 . 1 46 3 . . . . 9 2 9 2 4 2 1 ...... Philonotis calcarea 28 3 24 12 . . . 1 . . 1 . . . 4 2 4 . . 1 . . 2 ...... 10 ...... Ajuga reptans 11 15 2 13 . 4 7 . . 10 ...... 5 . 2 1 1 9 17 . 1 ...... 1 ...... Brachythecium rivulare 4 8 1 2 . . 1 . 2 2 ...... 3 . 4 1 7 8 4 . 9 . 3 18 . 19 1 . . . 2 5 . . 1 1 ...... Calamagrostis villosa ...... 1 . 1 1 1 . 1 . 3 2 . 10 1 . 1 4 9 4 6 2 27 14 7 1 ...... 2 . Leucanthemum vulgare agg. 1 2 2 15 32 33 15 . 7 7 ...... 4 . 2 . . 2 5 ...... 4 . . . . 1 ...... Gymnadenia conopsea agg. 13 5 18 16 . 6 5 6 . . 2 . . 2 1 . 5 2 . . . . 1 ...... 1 ...... Salix pentandra 1 . 6 2 . 32 ...... 1 1 1 1 14 1 4 1 2 1 2 . 10 8 . 1 . 1 . . . . 1 3 2 ...... 1 . . . Salix glauca ...... 1 31 3 . 5 7 . . 8 ...... 1 3 7 . . . 6 17 . 1 . . . . 1 . 1 . . 1 1 2 . 1 . . Agrostis capillaris 1 . 1 6 . 2 2 . 2 9 . . . . 1 . 2 . 2 2 4 12 3 . . . 3 . . 3 1 . 1 1 12 2 1 2 5 5 1 ...... Tofieldia calyculata 9 . 27 1 . 9 4 16 . . 1 1 . 1 1 3 1 . 3 1 ...... 4 ...... 1 ...... Poa trivialis 7 9 . 9 . . 3 . . 1 . . . . . 2 1 . 1 1 7 6 7 . 9 2 2 5 . . . . 1 . 1 5 . 4 1 1 ...... Myrica gale ...... 19 2 1 . 3 15 5 1 2 . 3 . 3 . . . 2 1 2 4 1 21 . 1 . 13 . . . 3 4 . 1 1 . 8 . . . . 1 . 1 Melampyrum pratense ...... 1 ...... 1 3 2 1 ...... 1 1 2 2 1 9 2 12 6 ...... 26 3 Mentha longifolia 21 25 8 8 . 1 5 . . 1 . . . . 1 . 1 . . . . . 1 ...... 2 ...... Polygala amara 6 1 19 2 . 8 19 2 . 1 . . . . . 2 2 . 1 . . . 2 . 1 ...... Rhizomnium pseudopunctatum ...... 7 15 1 . 1 1 . . 20 1 1 2 1 . 1 2 8 3 . . 1 4 11 ...... 1 ...... Carex viridula 1 1 5 1 . 2 1 5 . 4 . 9 9 1 7 12 1 . 1 5 . 1 1 2 1 2 3 . 7 . 1 . 1 . . 1 . 1 . . . 1 . . . 2 2 . . . Poa pratensis agg. 1 1 1 2 . 10 11 . . 11 1 . . . 1 . 5 . . . 4 3 13 . 6 2 . 1 . . 1 . . . 1 1 ...... Anemone nemorosa . 1 . 5 . 1 3 . . 1 ...... 8 . 5 . 2 18 3 . 1 ...... 1 . 5 2 . 5 ...... Dicranum bonjeanii 1 . 4 5 . 2 1 . . . 2 1 . 1 . 4 10 8 6 5 1 2 2 ...... 1 . 1 . . 1 . . 1 . 3 . . . . 1 . . 2 . 3 Salix phylicifolia ...... 13 8 1 5 6 5 . . 7 . . 1 . . . 1 17 3 . . . 4 4 . 1 . . . . . 1 5 ...... 2 Trifolium repens 2 1 1 12 . 10 5 . 4 1 ...... 2 . 1 . . 2 22 . 1 . . 1 . 3 . . . . . 1 ...... Chiloscyphus polyanthos agg. 1 . 2 2 . . 2 . 4 1 1 . . . 1 . 3 1 6 4 4 8 4 . 2 2 2 3 . 10 1 . . . 1 2 . . 1 1 ...... 1 . Sphagnum squarrosum . . . 1 . . . . . 3 . . . . 1 . . 3 1 6 2 4 . . 7 5 9 5 . 4 2 4 1 . 2 4 . 1 2 2 9 ...... Cladopodiella fluitans ...... 2 1 . 1 . . . . . 1 . 1 . . . 18 . . 3 . 4 . . . 4 . . . 1 . 1 1 . 8 16 10 3 5 3 19 . . Carex rariflora ...... 5 1 1 . 2 6 . . 6 . . 2 ...... 7 7 ...... 3 . 1 8 9 7 1 11 . 8 Cluster No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 Carex elata . . 1 . . . 12 8 . 1 . 1 2 . 1 1 1 . 1 2 1 . 3 . 7 . 3 15 1 . 1 . . 3 . 2 . 4 . 1 . 1 ...... Viola epipsila . . 1 ...... 1 1 . . 22 2 . 1 11 . 1 . . 1 . . 2 4 2 1 . 1 4 . . . 1 1 2 . . 1 ...... 1 . . Acer pseudoplatanus 16 2 4 8 . . 1 . . 1 ...... 1 5 . 6 1 . . . . . 1 . . 1 . 1 6 1 . 4 ...... Meesia triquetra . . 1 1 . . . . . 1 1 3 . 4 14 10 1 3 1 3 1 . 1 . 7 2 2 . . . 4 3 . . . 1 ...... Scapania irrigua ...... 3 5 . 1 3 . 1 . 2 2 1 2 1 . 3 . . 7 . 3 5 4 11 1 . 1 . . 1 1 . 5 . . 3 3 2 . 7 . . Veronica scutellata . . . 1 . . 1 . . . . 1 . . 1 1 . . 1 2 10 4 8 1 11 . . 12 . . 1 . . 1 1 2 1 1 ...... Polygala vulgaris 1 . 3 13 . 5 . 2 2 ...... 3 4 . 4 3 . 7 3 ...... 1 . 3 . . 1 . 1 ...... Lysimachia nummularia 2 16 . 4 . 3 15 . . 3 ...... 1 . 2 1 7 . 2 . . 2 ...... 1 1 ...... Carex distans 10 10 1 . . 3 21 8 . 10 ...... 1 ...... Appendix S5. Phytosociological interpretation of individual clusters produced in cluster analysis. CE = central Europe, SC = Scandinavia.

No. of plots cluster No. (CE/SC) Vegetation type / cluster interpretation 1 179/1 extremely rich fens with calcium carbonate precipitation of the Caricion davallianae alliance extremely rich fens with calcium carbonate precipitation of the Caricion davallianae alliance, 2 151/2 partially degraded (elements of wet meadows frequently present) 3 252/8 extremely rich fens of Caricion davallianae, initial phases at water-saturated patches 4 179/0 extremely rich fens of Caricion davallianae with meadow elements of Calthion palustris extremely rich fens of Caricion davallianae with meadow elements of Calthion palustris, usual 5 37/0 vegetation of small spring developed and maintained under regular mowing or grazing extremely rich fens of Caricion davallianae, localities concentrated in the Inner Western 6 130/0 Carpathians extremely rich fens of Caricion davallianae, Seslerietum uliginosae and other fen communities 7 190/0 with syntaxonomical relationships to intermittently wet meadows of Molinion caeruleae extremely rich fen of Caricion davallianae, communities with the dominance of Schoenus spp. 8 34/52 (~ Junco subnodulosi-Schoenetum nigricantis Allorge 1921) extremely rich fen of Caricion davallianae with subhalophytic influence, mostly occurring 9 1/45 nearby seashores in Scandinavia 10 72/62 extremely rich fens, syntaxonomically undifferentiated 11 0/157 extremely rich fens of the Caricion atrofusco-saxatilis alliance quaking rich fen of Stygio-Caricion limosae, initial phases at extremely waterlogged sites with 12 14/137 less-developed herb layer quaking rich fen of Stygio-Caricion limosae with boreal bryophytes and boreal sedges (Amblystegio scorpioidis-Caricetum limosae Osvald 1923, Stygio-Caricetum lasiocarpae 13 0/136 Nordhagen 1943) boreal brown-moss non-quaking rich fens (~ Drepanoclado revolventis-Trichophoretum 14 0/227 cespitosi Nordhagen 1928) 15 70/158 rich fens, syntaxonomically undifferentiated rich fens of Sphagno warnstorfii-Tomentypnion nitentis, initial phases of formation (~ 16 138/16 Campylio stellati-Trichophoretum alpini Březina et al. 1963) rich fens of Sphagno warnstorfii-Tomentypnion nitentis in the temperate zone, developed in contact with communities of extremely rich fens (? Sphagno warnstorfiani-Caricetum 17 166/1 davallianae Rybníček 1984) 18 1/206 rich fens of Sphagno warnstorfii-Tomentypnion nitentis in the boreal zone 19 219/12 rich fens of Sphagno warnstorfii-Tomentypnion nitentis 20 130/25 moderately rich fens, syntaxonomically undifferentiated 21 137/27 moderately rich fen grassland (~ Caricetum nigrae Braun 1915, Caricion fuscae) moderately rich fen grassland (~ Caricetum nigrae Braun 1915, Caricion fuscae), transient to 22 147/1 wet meadows of Calthion palustris moderately rich fen grassland (~ Caricetum nigrae Braun 1915, Caricion fuscae), transient to 23 187/5 wet meadows of Calthion palustris 24 11/82 quaking moderately rich fens with boreal elements and without meadow species quaking moderately rich fens with boreal elements, occurring in temperate Europe and thus slightly influenced by nutrient input (higher proportion of meadow species as compared to 25 152/11 cluster 24) quaking moderately rich fens transient to sedge-bed marsh vegetation of Magnocaricion 26 1/64 elatae in the northern boreal zone 27 34/85 quaking moderately rich fens transient to the sedge-bed marsh of Magnocaricion elatae 28 142/7 quaking moderately rich fens 29 8/83 quaking moderately rich fens 30 111/4 moderately rich fens with spring elements arcto-alpine intermediate non-calcareous fens of Drepanocladion exannulati 31 40/120 (Drepanocladetum exannulati Krajina 1933) arcto-alpine intermediate non-calcareous fens of Drepanocladion exannulati (Calliergo 32 10/148 sarmentosi-Eriophoretum angustifolii Hadač et Vápa 1967) moderately rich fens – poor fens with Sphagnum auriculatum agg. and S. papillosum, 33 62/31 localities concentrated in (sub-)oceanic regions

120

No. of plots cluster No. (CE/SC) Vegetation type / cluster interpretation 34 194/11 poor fens of Sphagno-Caricion canescentis poor fens grasslands of Sphagno-Caricion canescentis (~ Carici echinatae-Sphagnetum Soó 35 231/9 1944) quaking poor fens of Sphagno-Caricion canescentis (mostly Sphagno recurvi-Caricetum 36 224/18 rostratae Steffen 1931) quaking poor fens of Sphagno-Caricion canescentis (mostly Sphagno recurvi-Caricetum 37 117/128 rostratae Steffen 1931) poor fens of Sphagno-Caricion canescentis with water level decrease, moss layer frequently 38 115/27 dominated by Sphagnum palustre agg. poor fens of Sphagno-Caricion canescentis with water level decrease, moss layer frequently 39 179/9 dominated by Polytrichum commune 40 135/20 poor fens of Sphagno-Caricion canescentis with water level decrease 41 10/65 quaking poor fens with Sphagnum riparium 42 74/91 dystrophic hollows of Scheuchzerion palustris with the dominance of Sphagnum cuspidatum 43 15/104 dystrophic hollows of Scheuchzerion palustris with the dominance of Sphagnum majus 44 1/72 dystrophic hollows of Scheuchzerion palustris with the dominance of Sphagnum lindbergii 45 0/93 dystrophic hollows of Scheuchzerion palustris with the dominance of Sphagnum lindbergii dystrophic hollows of Scheuchzerion palustris with the dominance of Warnstorfia fluitans, 46 29/28 frequently with low cover of herb layer or without herb layer dystrophic hollows of Scheuchzerion palustris with the dominance of Warnstorfia fluitans and Carex limosa, i.e. Drepanoclado fluitantis-Caricetum limosae (Kästner et Flössner 1933) Krisai 47 81/5 1972 vegetation of lawns (slightly drier margins of dystrophic bog hollows) with Sphagnum 48 19/118 compactum, S. tenellum and Trichophorum cespitosum 49 162/2 the transition vegetation between fens and bogs in the temperate zone 50 7/116 the transition vegetation between fens and bogs in the boreal zone

121

Paper 4 Peterka T., Hájek M., Jiroušek M., Jiménez-Alfaro B., Aunina L., Bergamini A., Dítě D., Felbaba-Klushyna L., Graf U., Hájková P., Hettenbergerová E., Ivchenko T.G., Jansen F., Koroleva N.E., Lapshina E.D., Lazarevid P.M., Moen A., Napreenko M.G., Pawlikowski P., Plesková Z., Sekulová L., Smagin V.A., Tahvanainen T., Thiele A., Biţæ-Nicolae C., Biurrun I., Brisse H., Dušterevska R., De Bie E., Ewald J., FitzPatrick Ú., Font X., Jandt U., Kącki Z., Kuzemko A., Landucci F., Moeslund J.E., Pérez-Haase A., Rašomavičius V., Rodwell J.S., Schaminée J.H.J., Šilc U., Stančid Z. & Chytrý M. (2017): Formalized classification of European fen vegetation at the alliance level. – Applied Vegetation Science 20: 124–142.

122

Applied Vegetation Science 20 (2017) 124–142 Formalized classiﬁcation of European fen vegetation at the alliance level Tomas Peterka, Michal Hajek, Martin Jirousek, Borja Jimenez-Alfaro, Liene Aunina, Ariel Bergamini, Daniel Dıte, Ljuba Felbaba-Klushyna, Ulrich Graf, Petra Hajkov a, Eva Hettenbergerova, Tatiana G. Ivchenko, Florian Jansen, Natalia E. Koroleva, Elena D. Lapshina, Predrag M. Lazarevic, Asbjørn Moen, Maxim G. Napreenko, Paweł Pawlikowski, Zuzana Pleskova, Lucia Sekulova, Viktor A. Smagin, Teemu Tahvanainen, Annett Thiele, Claudia Bita-Nicolae, Idoia Biurrun, Henry Brisse, Renata Cu sterevska, Els De Bie, Jorg€ Ewald, Una FitzPatrick, Xavier Font, Ute Jandt, Zygmunt Kazcki, Anna Kuzemko, Flavia Landucci, Jesper E. Moeslund, Aaron Perez-Haase, Valerijus Rasomavicius, John S. Rodwell, Joop H.J. Schaminee, Urban Silc, Zvjezdana Stanci c & Milan Chytry

Keywords Abstract Biogeography; Ecological gradients; Endangered habitats; Mires; Releves; Aims: Phytosociological classification of fen vegetation (Scheuchzerio palustris-Carice- Supervised vegetation classification; tea fuscae class) differs among European countries. Here we propose a unified vegeta- Unsupervised vegetation classification; tion classification of European fens at the alliance level, provide unequivocal Vegetation plots; Wetlands assignment rules for individual vegetation plots, identify diagnostic species of fen alliances, and map their distribution. Abbreviations EVA = European Vegetation Archive; GIVD = Location: Europe, western Siberia and SE Greenland. Global Index of Vegetation-Plot Databases. Methods: 29 049 vegetation-plot records of fens were selected from databases using Nomenclature a list of specialist fen species. Formal definitions of alliances were created using the Tutin et al. (1968–1993) for vascular plants; presence, absence and abundance of Cocktail-based species groups and indicator spe- Frey et al. (2006) for bryophytes cies. DCA visualized the similarities among the alliances in an ordination space. The ISOPAM classification algorithm was applied to regional subsets with homogeneous Received 27 February 2016 plot size to check whether the classification based on formal definitions matches the Accepted 10 August 2016 results of unsupervised classifications. Co-ordinating Editor: Angelika Schwabe-Kratochwil Results: The following alliances were defined: Caricion viridulo-trinervis (sub-halophytic Atlantic dune-slack fens), Caricion davallianae (temperate calcareous fens), Caricion atrofusco-saxatilis (arcto-alpine calcareous fens), Stygio-Caricion limosae (boreal Peterka, T. (corresponding author, topogenic brown-moss fens), Sphagno warnstorfii-Tomentypnion nitentis (Sphagnum- [email protected])1, brown-moss rich fens), Saxifrago-Tomentypnion (continental to boreo-continental Hajek, M. ([email protected])1, nitrogen-limited brown-moss rich fens), Narthecion scardici (alpine fens with Balkan Jirousek, M. ([email protected])1,2, Jimenez-Alfaro, B. ([email protected])1, endemics), Caricion stantis (arctic brown-moss rich fens), Anagallido tenellae-Juncion Aunina, L. ([email protected])3, bulbosi (Ibero-Atlantic moderately rich fens), Drepanocladion exannulati (arcto-boreal- Bergamini, A. ([email protected])4, alpine non-calcareous fens), Caricion fuscae (temperate moderately rich fens), Dıte, D. ([email protected])5, Sphagno-Caricion canescentis (poor fens) and Scheuchzerion palustris (dystrophic hol- Felbaba-Klushyna, L. ([email protected])6, lows). The main variation in the species composition of European fens reflected site Graf, U. ([email protected])4, chemistry (pH, mineral richness) and sorted the plots from calcareous and extremely 1,7 Hajkova, P. ([email protected]) , rich fens, through rich and moderately rich fens, to poor fens and dystrophic hollows. Hettenbergerova, E. ISOPAM classified regional subsets according to this gradient, supporting the ecologi- ([email protected])1, cal meaningfulness of this classification concept on both the regional and continental Ivchenko, T.G. ([email protected])8, scale. Geographic/macroclimatic variation was reflected in the second most impor- Jansen, F. ([email protected])9, Koroleva, N.E. (fl[email protected])10, tant gradient. 11 Lapshina, E.D. ([email protected]) , Conclusions: The pan-European classification of fen vegetation was proposed and Lazarevic, P.M. supported by the data for the first time. Formal definitions developed here allow con- ([email protected])12, sistent and unequivocal assignment of individual vegetation plots to fen alliances at Moen, A. ([email protected])13, Napreenko, M.G. ([email protected])14, the continental scale.

Pawlikowski, P. ([email protected])15, 6Uzhhorod National University, Tolstoy Str. 44, 21Research Institute for Nature and Forest Pleskova, Z. ([email protected])1, 88018 Uzhhorod, Ukraine; (INBO), Kliniekstraat 25, 1070 Brussels, Sekulova, L. ([email protected])1, 7Department of Vegetation Ecology, Belgium; Smagin, V.A. ([email protected])8, Institute of Botany, The Czech Academy of 22Faculty of Forestry, University of Applied Tahvanainen, T. Sciences, Lidicka 25/27, 60200 Brno, Czech Sciences Weihenstephan-Triesdorf, Hans-Carl- (teemu.tahvanainen@uef.fi)16, Republic; von-Carlowitz-Platz 3, D-85354 Freising, Thiele, A. ([email protected])9, 8Komarov Botanical Institute, Russian Germany; Bita-Nicolae, C. ([email protected])17, Academy of Sciences, Prof. Popova 2, 197376 23National Biodiversity Data Centre, Biurrun, I. ([email protected])18, Sankt-Petersburg, Russia; Carriganore WIT West Campus, Carriganore, Brisse, H. ([email protected])19, 9Institute of Botany and Landscape Ecology, County Waterford, Ireland; Cu sterevska, R. ([email protected])20, University of Greifswald, Soldmannstr. 15, 24Facultat de Biologia, Universitat de De Bie, E. ([email protected])21, 17489 Greifswald, Germany; Barcelona, Av. Diagonal 643, 08028 Barcelona, Ewald, J. ([email protected])22, 10Polar-Alpine Botanical Garden-Institute, Kola Spain; FitzPatrick, U. Science Center, Russian Academy of Sciences, 25Geobotany and Botanical Garden, Institute (ufi[email protected])23, Kirovsk 6, Murmansk Province, 184256 Russia; for Biology, Martin-Luther University Halle- Font, X. ([email protected])24, 11Yugra State University, 628012 Khanty- Wittenberg, Am Kirchtor 1, 06108 Halle, Jandt, U. ([email protected])25,26, Mansiysk, Khanty-Mansiysk Autonomous Germany; Kazcki, Z. ([email protected])27, District, Russia; 26German Centre for Integrative Biodiversity Kuzemko, A. ([email protected])28, 12Institute for Nature Conservation of Serbia, Research (iDiv) Halle-Jena-Leipzig, Deutscher Landucci, F. (fl[email protected])1, Dr. Ivana Ribara 91, 11070 Belgrade, Serbia; Platz 5e, 04103 Leipzig, Germany; Moeslund, J.E. 13Museum of Natural History and Archaeology, 27Department of Vegetation Ecology, ([email protected])29, Norwegian University of Science and University of Wroclaw, Kanonia 6/8, 50-328 Perez-Haase, A. ([email protected])24, Technology, 7491 Trondheim, Norway; Wroclaw, Poland; Rasomavi cius, V. 14Institute of Chemistry and Biology, Immanuel 28National Dendrological Park ‘Sofievka’, ([email protected])30, Kant Baltic Federal University, Universitetskaya National Academy of Sciences of Ukraine, 12a Rodwell, J.S. ([email protected])31, 2, 236040 Kaliningrad, Russia; Kyivska St., 20300 Uman, Ukraine; Schaminee, J.H.J. 15Department of Plant Ecology and 29Department of Bioscience, Aarhus ([email protected])32, Environmental Conservation, Biological and University, Grenavej 14, 8410 Rønde, Denmark; Silc, U. ([email protected])33, Chemical Research Centre, Faculty of Biology, 30Institute of Botany, Nature Research Centre, Stanci c, Z. ([email protected])34, University of Warsaw, Zwirki_ i Wigury 101, Zaliuzjuz Ezeruz 49, 08406 Vilnius, Lithuania; Chytry, M. ([email protected])1 02096 Warsaw, Poland; 317 Derwent Road, Lancaster LA1 3ES, UK; 16Department of Environmental and Biological 32Alterra Wageningen UR, P.O. Box 47, 6700 1Department of Botany and Zoology, Masaryk Sciences, University of Eastern Finland, AA Wageningen, The Netherlands; University, Kotlarska 2, 61137 Brno, Czech Yliopistokatu 7, 80101 Joensuu, Finland; 33Institute of Biology, ZRC SAZU, Novi trg 2, Republic; 17Institute of Biology Bucharest, Romanian 1000 Ljubljana, Slovenia; 2Department of Plant Biology, Faculty of Academy, 296 Spl. Independentei, 060031 34Faculty of Geotechnical Engineering, Agronomy, Mendel University in Brno, Bucharest, Romania; University of Zagreb, Hallerova aleja 7, 42000 Zemed elsk a 1, 61300 Brno, Czech Republic; 18Department of Plant Biology and Ecology, Varazdin, Croatia 3Laboratory of Geobotany, Institute of Biology, University of the Basque Country UPV/EHU, University of Latvia, 3 Miera Street, 2169 P.O. Box 644, 48080 Bilbao, Spain; Salaspils, Latvia; 19 36 rue Henri Dunant, 13700 Marignane, This paper is dedicated to the memory of Kamil 4WSL Swiss Federal Institute for Forest, Snow € France; Rybnıcek (1933–2014), who established the and Landscape Research, Zurcherstr. 111, 20 Institute of Biology, Faculty of Natural first modern classification system of fens in 8903 Birmensdorf, Switzerland; Sciences and Mathematics, University of Ss. Central Europe, and Emil Hadac (1914–2003), 5Institute of Botany, Slovak Academy of Cyril and Methodius, Arhimedova 3, 1000, who contributed to unification of the Zurich-€ Sciences, Dubravsk a cesta 9, 84523 Bratislava, Skopje, Republic of Macedonia; Montpellier and Uppsala phytosociological Slovakia; traditions.

forming mosses; Udd et al. 2015) or both. From a syntaxo- Introduction nomic point of view, Eurosiberian fens are traditionally Fens (minerotrophic mires) are natural or semi-natural assigned to the class Scheuchzerio palustris-Caricetea fuscae ecosystems with a unique species composition. They can Tuxen€ 1937. be deﬁned as groundwater-fed wetlands poor in available In many parts of Europe, fens are currently endangered macronutrients whose herb layer is mostly dominated by habitats with great importance for biodiversity protection. Cyperaceae species and whose bryophyte layer is usually A large number of fens were destroyed by fertilizer applica- well developed and consists of Sphagnum species or so tion, drainage, abandonment of traditional uses and conse- called ‘brown mosses’ (i.e. non-sphagnaceous weft- quent successional changes in the second half of the 20th

Applied Vegetation Science Doi: 10.1111/avsc.12271 © 2016 International Association for Vegetation Science 125 Classification of European fens T. Peterka et al. century (Topic&Stancic 2006; Malson€ et al. 2008; Ber- Ruuhijarvi€ 1960; Persson 1961; Heikkila€ 1987), gives bryo- gamini et al. 2009; Koch & Jurasinski 2014; Grootjans phytes and vascular plants equal importance. It was fol- et al. 2015; Hajek et al. 2015). Therefore, selected types lowed, with various modifications, by several national or of fens have recently also been protected by the Euro- regional vegetation surveys from outside Fennoscandia pean Community (Habitats Directive, 92/43/EEC, from (e.g. Succow 1974; Rybnıcek et al. 1984; Sanda et al. 1992) as Natural Habitat Types of Community Interest: 2001; Kuznetsov 2003; Dıte et al. 2007; Tzonev et al. depressions on peat substrates of the Rhynchosporion 2009; Lapshina 2010; Hajek & Hajkov a 2011). (code 7150), Fennoscandian mineral-rich springs and Furthermore, some transitional classification systems spring fens (7160), and several types of Calcareous fens were introduced, reflecting ecologically delimited units but (7210, 7230, 7240) and Boreal mires (7310). However, keeping some broadly defined alliances frequently charac- the effective protection of individual habitats and corre- terized by selected vascular plant species (e.g. Kojicetal. sponding vegetation types at the continental scale is 1998; Koska & Timmermann 2004; Felbaba-Klushyna only possible on the basis of a harmonized classification 2010a,b; Rivas-Martınez 2011; Ermakov 2012). The dis- system with clearly defined units that would enable crepancies among classification systems for fen communi- unequivocal assignment of plant communities to higher ties have resulted in different perception and delimitation syntaxa such as alliances and classes. Therefore, it is nec- of alliances in individual European countries, leading to essary to establish a consistent vegetation classification confusion. The most problematic issues concern the Rhyn- system useful for communication among scientists from chosporion albae and Caricion lasiocarpae alliances, which different countries and for supporting conservation man- were originally delimited narrowly (Koch 1926; Lebrun agement (De Caceres et al. 2015). et al. 1949), but thereafter interpreted in different ways Classification of fen vegetation in various European (cf. Dierssen 1982; Rybnıcek et al. 1984). countries is different due to diverse classification concepts. The need for a consistent classification system in Europe Despite some regional differences, two main approaches to has recently driven vegetation scientists to elaborate fen vegetation classification are generally applied. The sys- broad-scale syntheses integrating national classification tems defining particular alliances on the basis of hydrologi- systems (De Caceres et al. 2015). One of the first steps is a cal conditions and vegetation physiognomy were synopsis of nomenclaturally valid high-rank syntaxa in introduced by Vanden Berghen (in Lebrun et al. 1949), Europe (EuroVegChecklist; Mucina et al. 2016). Simulta- Vanden Berghen (1952), Oberdorfer (1957), Pop (1960) neously with the construction of EuroVegChecklist, which and Dierssen (1982), and more or less accepted in some shares several authors with our study, we gathered avail- other vegetation surveys (e.g. Steiner 1992; Martincic able vegetation-plot records of fen vegetation in Europe, 1995; Coldea et al. 1997; Gerdol & Tomaselli 1997; Lajer aiming to test the quality of delimitation of the major fen 1998; Oberdorfer 1998; Jermacane & Laivins 2001; Lawes- alliances reported from Europe in terms of their floristic son 2004; Matuszkiewicz 2007; Graf et al. 2010). In these composition and reproducibility using formal definitions classification systems, topogenic waterlogged fens (usually (i.e. supervised classification; De Caceres & Wiser 2012). called Caricion lasiocarpae and Rhynchosporion albae)are Because some syntaxonomic aspects of our study have generally distinguished as opposed to spring fens and already been reflected in the final version of fen meadows (Caricion davallianae and Caricion fuscae). EuroVegChecklist, this study does not therefore concen- The dominance of different species of vascular plants trate on nomenclature, but deepens the classification (e.g. Carex davalliana, C. lasiocarpa, C. limosa, C. nigra and scheme of EuroVegChecklist by (1) formally delimiting Rhynchospora alba) is usually used as the chief alliance- individual alliances using a large set of primary data, i.e. delimiting criterion. This concept results in broadly defined individual vegetation-plot records, (2) identifying diagnos- alliances spanning a range of habitats of different, or even tic species and distribution patterns of individual alliances, contrasting, ecological features and entails the subsequent and (3) testing the robustness of the presented supervised delimitation of numerous associations, subassociations and pan-European classification by comparing it with regional varieties. unsupervised classifications and unconstrained gradient The second classification approach distinguishes individ- analysis. ual vegetation types along the main compositional and environmental gradient within fens that generally coin- Methods cides with pH and calcium concentration (the poor–rich Data collection and filtering gradient; Du Rietz 1949; Sjors€ 1952; Malmer 1986; Sjors€ & Gunnarsson 2002; Tahvanainen 2004; Hajek et al. 2006). The data sources were vegetation plots (phytosociological The vegetation classification based on the poor–rich gradi- releves) stored within national or regional vegetation data- ent, as introduced by Fennoscandian botanists (Dahl 1956; bases, mostly registered in the Global Index of Vegetation-

Plot Databases (GIVD; Dengler et al. 2011) and available were omitted as well as taxa determined only to the genus through the European Vegetation Archive (EVA; Chytry level. Subspecies records were merged to the level of spe- et al. 2016), and also private data of European mire cies. Taxa of problematic, unstable or ambiguous status researchers (Appendix S1). For comparative purposes, the (usually not equally differentiated in all the data sources) data set was enlarged with vegetation plots of E. Lapshina were merged to aggregates or species sensu lato from western Siberia, the locus classicus of the Saxifrago- (Appendix S2) to minimize the taxonomic bias (Jansen & Tomentypnion alliance (Lapshina 2010), although there was Dengler 2010). a certain geographic gap between the European and Siber- Since the original assignment of plots to the Scheuchze- ian data (Fig. 1). All the vegetation plots were combined rio palustris-Caricetea fuscae class was inconsistent among into a single file using the TURBOVEG 3 software and data sets and was even absent in some data sets, we had thereafter imported to the JUICE 7.0 program (Tichy to eliminate plots of vegetation types other than fens on 2002), in which the subsequent analyses, apart from DCA, the basis of habitat specialists (further referred to as ‘typi- were performed. cal fen species’; Appendix S3). First, plots containing at As in our previous study of European fens (Jimenez- least four typical fen species were selected. However, Alfaro et al. 2014), we had to deal with data collected plots of very species-poor communities, such as high- using different sampling designs (see also Michalcova mountain fens dominated by Drepanocladus exannulatus or et al. 2011). Therefore, several steps had to be carried out dystrophic hollows with Sphagnum cuspidatum or S. lind- to homogenize and balance the data set. First, only geo- bergii did not match this simple criterion. Therefore, plots referenced plots of a size of 1–100 m2 were selected for with at least one typical fen species reaching cover values analyses. Although Chytry&Ot ypkov a(2003)recom- over 25% or 50% (for details see Appendix S3) were mended 16 m2 as a standard plot size for sampling fens, added. restriction to a narrower plot size range would have Since bryophytes are extremely important organisms in resulted in a large loss of important data from several mires (Jones et al. 1994; Bergamini et al. 2001; Udd et al. regions. The possible effect of different plot sizes was 2015), all plots with no or insufficiently identified bryo- assessed through applying unsupervised classification to phytes (e.g. with non-identified species of sphagna) were subsets of plots of equal sizes (see the section Unsuper- excluded. Nevertheless, the data set still contained some vised classification). As suggested by Dengler et al. plots of other vegetation types harbouring some fen species (2009), means and variation in plot sizes within individ- (mostly wet meadows of the Calthion palustris or Molinion ual clusters were presented. caeruleae alliances). The following criteria were used to The nomenclature was harmonized following Tutin exclude these non-fen plots: (1) total cover of non-fen spe- et al. (1968–1993) for vascular plants and Frey et al. cies exceeding 25% for plots in which covers of individual (2006) for bryophytes. Algae, fungi, lichens and hybrids species were indicated, and (2) the presence of at least six non-fen species in plots with species presences/absences only. The list of non-fen species (Appendix S3) was partially adopted from Jimenez-Alfaro et al. (2014) and extended according to the authors0 experience. Covers of individual species were merged following the protocol of the JUICE software, recently formally described by Fischer (2015). Moreover, plots with a cover of 25% or more of selected woody species (Appendix S3) were excluded to avoid forests and scrub with fen species in the herb layer. As there was an overlap between some databases, dupli- cates were searched for and eliminated. Finally, it was necessary to stratify the data set geographically to reduce oversampling of some countries (Knollova et al. 2005), especially those of Western and Central Europe where thousands of digitized plots were available in contrast to other regions (see Schaminee et al. 2009; Peterka et al. 2015; Chytry et al. 2016). Therefore, we geographically stratified the data from Western and Central Europe using a maximum of ten plots randomly selected from each grid cell of 1.25 min longitude 9 0.75 min latitude (approxi- Fig. 1. Distribution of fen vegetation plots compiled in the initial data set. mately 1.5 9 1.4 km).

generalists in less numerous plot groups. Plots that reached Supervised classification lower index values remained unclassified. To identify the core set of plots representing particular alli- The diagnostic species of particular plot groups (al- ances, we created formal definitions consisting of a priori liances) were calculated using the phi coefficient of associa- defined species groups based on the geographically tion (Tichy & Chytry 2006) for the equalized size of all stratified data set. The sociological groups (Appendix S4), groups. Species with a fidelity to a particular alliance of i.e. groups of species with a statistical tendency of co- / > 0.3 were considered as diagnostic (Table 1). The sig- occurrence in vegetation plots (Kocı et al. 2003), were nificance of fidelity was tested using Fisher’s exact test developed using the Cocktail method (Bruelheide 1995, (P < 0.001). Diagnostic species were calculated twice, first 2000) with the phi coefficient as a measure of interspecific for core plots only (Table 1, Appendix S6) and then for association (Chytry et al. 2002). The resulting sociological core plots plus plots assigned on the basis of FPFI groups, supplemented by covers of selected species or total (Appendix S7). covers of some sociological groups (‘functional groups with The data set of core plots of alliances was subjected to selection based on the total cover’ in Landucci et al. 2015), DCA with three pseudo-species cut levels for species cover were combined using the logical operators AND, OR (0%, 5% and 25%). Centroids were calculated for each and NOT (Bruelheide 1997). The formal definitions alliance that was successfully reproduced by the formal (Appendix S5) were applied to the entire data set except definition. This analysis was performed in R software the plots containing presence–absence data only. A socio- (v 2.9.0; www.r-project.org, package vegan). The original logical group was considered as being present in a plot if at material of Rhynchosporetum albae Koch 1926 (valid least half of its member species occurred within this plot. nomenclature type of Rhynchosporion albae Koch 1926) The groups of selected plots were considered cores of indi- consisting of two vegetation plots was also shown in the vidual alliances after exclusion of a few transitional plots ordination space in order to demonstrate the original matching two definitions at the same time. meaning of this alliance. Using this method, we defined 13 alliances. In addition, we tried to define the Caricion lasiocarpae and Rhynchospo- Unsupervised classification rion albae alliances in their narrower ‘ecological’ concepts, i.e. to delimit groups of plots corresponding composition- To assess whether the classification based on formal defini- ally to the nomenclature type releves of Caricetum lasio- tions mirrors the main vegetation gradients, and to check carpae Koch 1926 and Rhynchosporetum albae Koch 1926 for possible effects of different plot sizes on the classifica- (see Dengler et al. 2004). Two oromediterranean alliances tion results, we performed several unsupervised classifica- with a narrow geographic range (Caricion intricatae Quezel tions using subsets of plots of different sizes from different 1953, Festucion frigidae Rivas-Martınez et al. 2002), which regions. Ten plot subsets were chosen from six regions: (1) are also listed in EuroVegChecklist, could not be SE Greenland and arctic Europe (plots of 1 m2); (2) south- distinguished due to the absence of plots with identified ern and central Scandinavia (plots of 1 and 16 m2); (3) NE bryophytes. Europe (plots of 1 and 100 m2); (4) Eastern-Central Eur- To assess the distributional ranges of the alliances (in- ope (plots of 1, 15–16 and 50–100 m2); (5) the Alps (plots cluding their non-core plots), the remaining plots, which of 10–20 m2); (6) southern and central Balkans (plots of were not ranked to any alliance or met the conditions of 15–16 m2). All these subsets were selected from the geo- two formal definitions, were assigned to the most similar graphically non-stratified data set and then stratified in the alliance based on the Frequency-Positive Fidelity Index same way as the main data set used for the supervised clas- (FPFI; Tichy 2005). This index expresses the similarity of sification (i.e. taking a maximum of ten plots from each species composition of individual vegetation plots to the grid cell of 1.25 9 0.75 min). We applied the unsuper- predefined groups of plots assigned to given vegetation vised non-hierarchical classification algorithm ISOPAM types, in our case to cores of particular alliances. This step (Schmidtlein et al. 2010) at the level of six clusters with also enabled the classification of plots with presences/ the Jaccard coefficient as the dissimilarity measure. ISO- absences. The minimum FPFI value for assignment was PAM is based on the classification of ordination scores from arbitrarily set to 0.15 for widespread fen vegetation types isometric feature mapping. Ordination and classification and 0.30 for more local types with a small number of are repeated in a search for groups rich in diagnostic spe- core plots (Anagallido tenellae-Juncion bulbosi, Caricion viri- cies and high overall fidelities of species to particular clus- dulo-trinervis, Caricion stantis, Narthecion scardici, Saxifrago- ters. The number of clusters was arbitrarily set to six, Tomentypnion). Preliminary results suggested that such a which corresponds to the number of major ecological types two-level threshold made sense, since the value of the sim- of fens usually recognized in Europe (Hajek et al. 2006). In ilarity index could be biased due to the higher constancy of the case of southern and central Scandinavia, the first run

Table 1. Shortened synoptic table of fen alliances in Europe based on core plots (i.e. plots formally assigned to alliances).

Alliance CvT Cd CaS SCl SwT SaT Ns Cs AJ De Cf SCc Sp Group no. 1 2 3 4 5 6 7 8 9 10 11 12 13 Number of plots 138 2456 213 604 1059 101 37 17 97 449 1594 2001 542 Plot size range (m2) 1–100 1–100 1–16 1–100 1–100 1–100 2–80 1–25 1–100 1–100 1–100 1–100 1–100 Mean plot size (m2) 15 18.4 2.5 2.8 18.8 21.4 18.3 4.0 13.2 4.4 20.2 18.8 23.8

Caricion viridulo-trinervis Salix repens 99 4. 1 5 4.. 1 . 5 1 . Carex trinervis 64 ...... 1.. Juncus anceps 60 ...... Hydrocotyle vulgaris 87 2. 1 2 ...24.173 1 Schoenus nigricans 54 3. 1 1 ...13. 1 . . Juncus articulatus 72 32 1 1 22 . 19 . 15 1 18 1 . Campylium polygamum 281. 1 1 1.. . . 1 . 1 Ranunculus flammula 59 2 . 2 4 . . . 40 . 24 1 1 Calliergonella cuspidata 74 48 . 3 27 11 . . 13 1 34 1 2 Dactylorhiza incarnata 41 9 . 3 6 28 . . . 1 2 1 . Liparis loeselii 26 3 . 2 2 14 . . . . 1 1 . Caricion davallianae Carex davalliana . 61 8118.... .21 . Carex hostiana .34114 ...2.2 . . Valeriana dioica s. l. 1 50 . 1 35 1 . . 3 1 25 2 . Schoenus ferrugineus .21122.... .1 . . Carex panicea 29 78 7 21 53 . 3 . 45 1 44 6 1 Tofieldia calyculata .268 . 7 . . . . 1 1 1 . Succisa pratensis . 41 . 1 21 . . . 20 1 17 2 . Palustriella commutata s.l..271911.5. .11 . . Caricion atrofusco-saxatilis Carex capillaris .269 13...... Juncus triglumis .271 11 .5. .11 . . Salix reticulata .163 11...... Polygonum viviparum .485 112. .6 . 6 1 1 . Carex atrofusca .162 11...... Saxifraga aizoides .362 11....1... Equisetum variegatum 5970 17 ....11 . . Thalictrum alpinum .149 28 ....11 . . Catoscopium nigritum .140 11...... Tofieldia pusilla .148 111. . . . . 1 . . Meesia uliginosa .134 11....1... Carex vaginata .136 18 ....111 . Carex parallela .127.1...... Carex microglochin .12711...... Saussurea alpina .135111...... Carex bicolor .120...... Pinguicula alpina .32511....11 . . Oncophorus virens .12311....11 . . Selaginella selaginoides .946318.19.1131 . Carex saxatilis .12311....61 . . Drepanocladus revolvens agg.452723341.565.183 1 . Pinguicula vulgaris .3437317...11311 Stygio-Caricion limosae Scorpidium scorpioides 44292 8..6511111 Calliergon trifarium .18363 ....11 .1 Cinclidium stygium . 1 12 32 11 . . 6 . 8 1 1 . Utricularia minor agg. .3.27438.41112 Sphagno warnstorfii-Tomentypnion nitentis Sphagnum warnstorfii ...185 211. . 3 3 2 1 Tomentypnum nitens . 19 19 1 66 41 . 29 . 1 4 1 .

Table 1. (Continued). Alliance CvT Cd CaS SCl SwT SaT Ns Cs AJ De Cf SCc Sp Group no. 1 2 3 4 5 6 7 8 9 10 11 12 13 Number of plots 138 2456 213 604 1059 101 37 17 97 449 1594 2001 542 Plot size range (m2) 1–100 1–100 1–16 1–100 1–100 1–100 2–80 1–25 1–100 1–100 1–100 1–100 1–100 Mean plot size (m2) 15 18.4 2.5 2.8 18.8 21.4 18.3 4.0 13.2 4.4 20.2 18.8 23.8

Aulacomnium palustre 1 10 1 2 73 35 41 6 7 2 32 20 2 Sphagnum contortum . ..629.3.5231 . Paludella squarrosa . 1 5 3 34 6 . 6 . 13 1 1 1 Sphagnum teres ...23628..71751 Saxifrago-Tomentypnion Hamatocaulis vernicosus .4161086 .. . . 4 1 1 Brachythecium mildeanum 121.269 .. . . 1 1 . Stellaria crassifolia ...1162 ...... Drepanocladus aduncus agg. 14 1 . 1 1 77 .. . . 2 1 1 Saxifraga hirculus .1..469 .6 . . . 1 . Carex diandra .3.121085 .. . 1113 1 Cicuta virosa .1.1155 .. . 1 1 1 1 Triglochin maritima 12..153 ...... Epilobium palustre 44161782 .. 410248 1 Eriophorum gracile .1.3238 ..51111 Lysimachia thyrsiflora .1.3442...2861 Carex appropinquata .2.1732... .211 Helodium blandowii .111724...111 . Bryum pseudotriquetrum agg.4433716356932. 6 7131 1 Narthecion scardici Pinguicula balcanica .1..1 .100 ..13.. Plantago gentianoides .1..1 .68 ...1.. Pseudorchis frivaldii .1..1 .62 ..111. Gentiana pyrenaica .1....51 ...1.. Scapania irrigua .1112.49 ..921. Primula deorum ...... 32 ...1.. Philonotis seriata .11.1.41 .2521 . Sphagnum platyphyllum ...11.30..22.1 Saxifraga stellaris .11.1.24..421. Juncus filiformis .1..1.27..71271 Sphagnum subsecundum . . . 4 10 . 35 . 10 3 28 2 1 Calliergon stramineum . 1 1 6 31 4 57 24 2 25 28 42 2 Caricion stantis Dupontia fisheri ...... 94 .1... Calliergon turgescens ..1....35 .1... Ranunculus hyperboreus ....1..29.3... Calliergon richardsoni . ..111.29.111 . Eriophorum scheuchzeri .11.1..35.131.1 Aulacomnium turgidum ..5....24.1... Sphagnum squarrosum . ..131.2411631 Anagallido tenellae-Juncion bulbosi Anagallis tenella 12 1 . 2 1 . . . 90 .11 . Juncus bulbosus 11.15 ...89 1714 Hypericum elodes ...1....61 .1 .1 Carum verticillatum ...... 60 .21 . Sphagnum denticulatum agg. . . . 1 2 . . . 64 1141 3 Potamogeton polygonifolius 11.11 ...41 .21 . Narthecium ossifragum .1.11 ...37 .111 Eleocharis multicaulis 11.11 ...36 .111 Scutellaria minor ...... 32 .11 . Wahlenbergia hederacea ...... 24.11. Juncus acutiflorus .2.11 ...31.71 . Erica tetralix 91.11 ...31.153

Drepanocladion exannulati Drepanocladus exannulatus .1277.4329.89 16 2 2 Calliergon sarmentosum .1553.1935.4621 . Caricion fuscae Viola palustris . 4 . 2 30 . . . 19 2 60 18 1 Agrostis canina 12 7 . 1 27 2 46 . 13 1 61 23 4 Sphagno-Caricion canescentis Sphagnum recurvum agg. . . . 1 28 . . . 2 1 22 100 1 Vaccinium oxycoccos agg. . 2 . 24 40 35 . . . 2 8 67 38 Polytrichum commune . 1 . . 2 . 16 12 3 5 14 34 1 Scheuchzerion palustris Sphagnum cuspidatum ....1...814464 Scheuchzeria palustris .1.511...211467 Sphagnum majus ....1....31141 Drepanocladus fluitans 41.11 ..18213542 Rhynchospora alba . 1 . 9 6 . . . 11 1 4 10 31 Diagnotis species for two or more alliances Epipactis palustris 63 34 . 2 17 48 . . . 1 2 1 . Eriophorum latifolium .592 434651. 1 1 3 1 . Primula farinosa s. l. . 42 14 1 8 . 68 ..111. Carex limosa .2.901369..91562452 Carex chordorrhiza .1.571476 .. .132 7 1 Carex lasiocarpa .5.582550..3613137 Menyanthes trifoliata 1 18 1 77 38 83 . . 25 6 26 27 9 Carex dioica . 11 22 7 47 50 . . . 1 3 1 1 Carex nigra 38 39 15 12 47 2 95 . 9 23 65 29 2 Carex echinata . 12 . 1 31 1 70 . 66 5 67 15 1 Other species reaching a frequency higher than 25% in at least one cluster Eriophorum angustifolium 12 34 29 41 51 46 11 6 33 64 60 46 22 Carex rostrata . 18 1 45 50 51 . . 3 32 35 52 22 Campylium stellatum s. l. 27 68 69 31 57 14 16 24 14 2 5 1 . Carex flava agg. 64 59 5 20 45 1 22 . 48 1 20 2 1 Parnassia palustris 55 59 27 4 39 29 22 . 12 2 13 1 . Drosera rotundifolia 1 4 . 13 42 12 . . 39 1 19 39 31 Potentilla palustris 12 2 . 21 25 36 . . 8 23 37 23 5 Equisetum fluviatile . 13 . 42 35 19 . . 2 6 26 13 2 Andromeda polifolia . 1826258 . . . 7 32032 Carex curta .1.15 .3.22625214 Eriophorum vaginatum .1617514. .652927 Fissidens adianthoides 12 28 6 1 15 . . . 3 1 3 1 . Aneura pinguis 41327242348 . 7 2 5 1 1 Eleocharis quinqueflora 33 22 16 6 7 . 3 . 3 1 1 1 . Peucedanum palustre .4.9730...114111 Utricularia intermedia agg. . 1 . 29 3 24 . . 9 1 2 1 1 The percentage occurrence frequency values are shown. Species are sorted by decreasing fidelity within alliances. The background shading indicates diagnostic species of alliances in cases when / > 0.3, bold numbers indicate diagnostic species when / > 0.5. Species with a clear optimum outside fen vegetation and species reaching a frequency lower than 20% within any cluster are shown only in the full version of the table (Appendix S6). of ISOPAM classification led to clusters that did not corre- alliances in Scandinavia as compared to other regions, this spond to the clusters resulting from other regional classifi- result may have been caused by an insufficient number of cations. Because the ISOPAM algorithm is non- clusters. We therefore ran ISOPAM again, with seven hierarchical and because we expected a higher number of resulting clusters. Diagnostic species of each cluster were

Applied Vegetation Science Doi: 10.1111/avsc.12271 © 2016 International Association for Vegetation Science 131 Classification of European fens T. Peterka et al. calculated in the same way as for the pan-European data springs and peat substrates developed on limestone, cal- set. The species with fidelity to a particular cluster of careous sedimentary or metamorphic rocks and ultrabasic / > 0.3 were considered diagnostic (Appendix S8) and the crystalline rocks. The herb layer consists mainly of calcicole significance of fidelity was tested using Fisher’s exact test graminoids (e.g. Carex davalliana, C. hostiana, Eleocharis (P < 0.01). quinqueflora, Eriophorum latifolium, Schoenus ferrugineus) and herbs such as Parnassia palustris, Pinguicula vulgaris, Results Primula farinosa subsp. farinosa and Tofieldia calyculata.The bryophyte layer is composed of brown mosses such as Cam- Vegetation units (supervised classification) pylium stellatum s. l., Drepanocladus revolvens agg. (D. cossonii A total number of 29 049 vegetation plots were identified in this case), Palustriella commutata s. l. and Philonotis cal- as representing fen vegetation (Fig. 1), of which 24 091 carea. Plots are scattered throughout almost the whole of plots remained in the data set after geographic stratification Europe, though concentrated in the Alps and the Carpathi- that removed some plots from oversampled regions. In ans. In Iceland and northern Scandinavia, the alliance is total, 9308 plots (38.6% of the geographically stratified represented by stands dominated by Eleocharis quinqueflora data set) were assigned unequivocally to alliances (i.e. they only. met the assumption of exactly one formal definition). They Caricion atrofusco-saxatilis Nordhagen 1943 (group 3) are referred to as ‘core plots’. Only 71 plots matched two includes low-productive communities of calcareous min- definitions and were therefore moved to the unclassified eral substrates, with initial successional stages occurring subset. In the next step, 7629 plots (31.7%) were associ- even on gravel, in arctic or alpine climates. The alliance ated with groups of the core plots of particular alliances on shares some typical species with the previous alliances of the basis of the similarity expressed by the FPFI values; we extremely rich fens, but is differentiated by the presence of refer to these as ‘non-core plots’. The remaining 7154 plots the arcto-alpine species (e.g. Carex atrofusca, C. microglochin, (29.7%) were unclassified (see Appendix S9 for their char- Juncus triglumis, Kobresia simpliciuscula, Salix reticulata, Sax- acterization). Fen alliances were formally defined ifraga aizoides, Thalictrum alpinum) that also frequently (Appendix S5) and their diagnostic species were identified occur in contact habitats such as snow-beds or alpine grass- (Table 1, Appendix S6). The effort to delimit Caricion lasio- lands. It occurs in the Alps (mostly above 2000 m a.s.l.), carpae and Rhynchosporion albae was not successful due to the Scandinavian mountains, Iceland and Greenland. the overlap with other alliances (Appendix S10). As a Stygio-Caricion limosae Nordhagen 1943 (group 4) repre- result, 13 alliances were defined formally. According to the sents fens with sedges and brown mosses occurring mostly ecological classification of fens (Hajek et al. 2006), they in topogenic, strongly waterlogged wetlands with peat can be interpreted as follows: calcareous fens and extremely accumulation. The vegetation is composed of boreal sedges rich fens (groups 1–3), rich fens (groups 4–6), moderately rich (Carex chordorrhiza, C. lasiocarpa, C. limosa and occasionally fens (groups 9–11), poor fens (group 12) and dystrophic (bog) C. livida) and weft-forming bryophytes (Calliergon trifarium, hollows (group 13). Groups 7 and 8 represent geographi- Scorpidium scorpioides) with sporadic occurrence of Sphag- cally restricted types of rich fens transitional to moderately num species such as S. contortum and S. platyphyllum.The rich fens. A generalized classification scheme with distin- alliance is widespread in N Europe, extending southwards guishing features of the fen alliances is given in to Britain, Ireland, the Baltic states, the Alps and, rarely, Appendix S11 and the geographic distribution of these alli- the Carpathians and the Balkans. ances is presented in Fig. 2. Sphagno warnstorfii-Tomentypnion nitentis Dahl 1956 Caricion viridulo-trinervis Julve ex Hajek et Mucina in (group 5) is characterized by calcium-tolerant sphagna, Theurillat et al. 2015 (group 1) includes the vegetation of i.e. Sphagnum contortum, S. subnitens, S. teres, S. warnstorfii the dune-slacks of the Atlantic coast of W Europe. Typical and S. subfulvum (the last in N Europe only), which are taxa are Carex trinervis, Juncus anceps and Salix repens.The accompanied by other mosses depending on microtopog- alliance is further characterized by a peculiar combination raphy (e.g. Aulacomnium palustre, Paludella squarrosa and of alkaline fen specialists (Dactylorhiza incarnata, Eleocharis Tomentypnum nitens). Typical bryophytes of extremely quinqueflora, Epipactis palustris), sub-halophytic and halo- rich fens (Campylium stellatum s. l., Drepanocladus revolvens phytic species (Centaurium littorale s. l., Glaux maritima, agg.) or calcicole vascular plants (Carex davalliana, Eleo- Samolus valerandi) and species of disturbed wetlands or charis quinqueflora, Eriophorum latifolium, Parnassia palus- generalist taxa (Agrostis stolonifera agg., Calamagrostis epige- tris) still occur frequently. Some Central European jos, Mentha aquatica). Most plots occur in the Netherlands, vegetation types dominated by Carex davalliana in the with further occurrences in France, Ireland and Denmark. herb layer and sphagna in the bryophyte layer, formerly Caricion davallianae Klika 1934 (group 2) comprises min- often classified to Caricion davallianae,havealsobeen eral-rich fen vegetation on both calcareous tufa-forming included in this alliance. Aulacomnium palustre, Sphagnum

Fig. 2. Distribution of core and non-core plots of individual alliances within the non-stratified data set. teres, S. warnstorfii and Tomentypnum nitens form small some typical species with the previous rich-fen alliances. It hummocks on which shallow-rooting acidophytes (Dro- further contains nutrient-demanding bryophytes of aqua- sera rotundifolia) or species preferring drier (i.e. oxic) con- tic and semi-aquatic habitats, either generalists (Brachythe- ditions can grow. Due to the superficial structure and cium mildeanum, Drepanocladus aduncus agg., Marchantia water chemistry representing the niche margins of both polymorpha) or fen specialists with higher phosphorus calcicole and calcifuge plants, this vegetation type demands (Hamatocaulis vernicosus;compareHajek et al. belongs to the most species-rich fen communities. The 2014). The co-existence of nutrient-demanding wetland alliance is distributed across Europe, being concentrated generalists and reed-bed species (Cicuta virosa, Ranunculus in mountain or highland areas. lingua, Thelypteris palustris), phosphorus-demanding grass- Saxifrago-Tomentypnion Lapshina 2010 (group 6) land species (Poa pratensis agg., Rumex acetosa s. l.) and includes calcium-rich (but not tufa-forming) fens sharing species of low-productive boreo-continental habitats

(Saxifraga hirculus, Stellaria crassifolia, subhalophytic Triglo- stellatum s. l. The alliance had been previously reported chin maritima) is characteristic of the herb layer. The group only from the Iberian Peninsula, but we also detected of nutrient-demanding grassland and reed-bed species occurrences in France, Britain and Ireland. delimits this alliance against other fen types. Sphagnum Drepanocladion exannulati Krajina 1933 (group 10) com- species are usually absent. In addition to W Siberia, where prises species-poor low-productive communities domi- it was first described (Lapshina 2010), the alliance has a nated by the pleurocarpous mosses Drepanocladus scattered distribution across the NE European lowlands exannulatus and Calliergon sarmentosum. The sparse herb with an isolated occurrence in the Romanian Carpathians. layer consists of cotton grasses (Eriophorum angustifolium, Non-core plots are also distributed in NE Germany and the E. scheuchzeri) accompanied by sedges (Carex curta, Jura Mountains. In contrast to the previous alliance, Sax- C. lachenalii, C. rariflora). In contrast to the following alli- ifrago-Tomentypnion more frequently occupies topogenic ance, Drepanocladion exannulati usually lacks grasses (Agros- habitats or waterlogged springs. tis canina, Anthoxanthum odoratum agg., Festuca rubra agg.) Narthecion scardici Horvat ex Lakusic1968(group7)isa and moderately nutrient-demanding dicots of mire mead- low-productive, rich to moderately rich fen community ows (e.g. Cirsium palustre, Lysimachia vulgaris, Ranunculus sharply differentiated from other fen alliances by the pres- flammula). The vegetation develops on waterlogged non- ence of Balkan endemics such as Narthecium scardicum, Pin- calcareous sites in boreal-arctic regions and in high moun- guicula balcanica, Primula deorum, P. farinosa subsp. exigua tains in Central and S Europe. and Pseudorchis frivaldii. Spring species of the Montio-Carda- Caricion fuscae Koch 1926 (group 11) includes slightly minetea class (Epilobium nutans, Saxifraga stellaris, Soldanella acidic sedge-moss fens with intermediate to low calcium pindicola agg.) and mountain grassland species (Ligusticum supply. The alliance comprises mesotrophic fens of water- mutellina, Nardus stricta) are also typical for this alliance. logged sites characterized by Carex diandra, Menyanthes trifo- The bryophyte layer is mostly formed by Drepanocladus liata and Potentilla palustris, young mire meadows as well as exannulatus, Philonotis seriata and Sphagnum subsecundum. initial stages of mire succession on shallow peat layers. Cal- The alliance is restricted to the high mountains of the cicole species are mostly absent. The bryophyte layer fre- southern Balkan Peninsula (Bulgaria, Kosovo, Macedonia, quently contains nutrient-demanding peat-mosses Montenegro and probably also Albania and Greece) and is (Sphagnum denticulatum agg., S. subsecundum, S. teres)and developed near streams and springs, usually above the tim- otherbryophytessuchasAulacomnium palustre, Bryum berline. pseudotriquetrum agg., Drepanocladus exannulatus and Caricion stantis Matveyeva 1994 (group 8) comprises Philonotis fontana agg. In contrast to the following alliance, brown-moss-sedge vegetation of high-arctic areas. The Sphagnum species of the Cuspidata section do not prevail. herb layer is typically formed of Carex aquatilis subsp. The alliance is not sharply differentiated in terms of species stans or Dupontia fisheri, accompanied by other species with composition because of a high proportion of pH generalists. Holarctic distribution in the arctic or sub-arctic zones (Erio- It occurs throughout Europe, but only a few plots were phorum scheuchzeri, Juncus biglumis, Ranunculus hyperboreus, recorded in the boreal and arctic zones. Salix polaris, Saxifraga foliolosa). The bryophyte layer con- Sphagno-Caricion canescentis Passarge (1964) 1978 (group tains, for example, Aulacomnium turgidum, Calliergon gigan- 12) represents vegetation of acidic minerotrophic mires teum, C. sarmentosum, C. turgescens and Campylium stellatum that are poor with respect to both species richness and s. l. All core plots come from Svalbard, none-core plots mineral supply. Frequent dominants of the bryophyte were recorded in Greenland, Iceland and the arctic coast of layer are Sphagnum recurvum agg., S. sect. Sphagnum northern Norway. The community occupies the wettest (S. palustre s. l., S. papillosum)andPolytrichum commune. habitats within the arctic tundra, e.g. stream valleys inun- There are frequent transitions to initial stages of ombro- dated by flowing water after snowmelt or permanently trophic mires (bogs) characterized by the presence of Carex waterlogged depressions (Matveyeva 1994). pauciflora, Eriophorum vaginatum or Sphagnum magella- Anagallido tenellae-Juncion bulbosi Br.-Bl. 1967 (group 9) nicum. Nevertheless, a minerotrophic water regime is still represents Ibero-Atlantic moderately rich fens character- indicated by species that do not enter pristine bogs, e.g. ized by diagnostic species with Atlantic or sub-Atlantic dis- Agrostis canina, Carex echinata, C. nigra, Menyanthes trifoliata, tribution ranges (Anagallis tenella, Eleocharis multicaulis, Potentilla palustris or Viola palustris. The community is dis- Juncus acutiflorus, J. bulbosus, Hydrocotyle vulgaris, Hypericum tributed throughout Europe, especially in the temperate elodes, Narthecium ossifragum, Potamogeton polygonifolius, and boreal zones. Scutellaria minor) which make its differentiation from other Scheuchzerion palustris Nordhagen ex Tx. 1937 (group fen types rather pronounced. The moss layer is most fre- 13) involves vegetation of dystrophic, extremely acidic quently formed of Sphagnum denticulatum agg. or other and species-poor hollows. The bryophyte layer is usually bryophytes such as Calliergonella cuspidata and Campylium formed of Drepanocladus fluitans, Sphagnum cuspidatum,

S. majus or S. lindbergii (the latter only in the boreal and were shifted to the arcto-alpine end of the second gradi- arctic zones). The herb layer consists of few species such ent. The gradient of the third axis cannot be easily defined as Carex limosa, Rhynchospora alba and Scheuchzeria palus- ecologically, but it delimits Caricion stantis from both Sty- tris. The alliance is traditionally included to the Scheuchz- gio-Caricion limosae and Caricion atrofusco-saxatilis and erio palustris-Caricetea fuscae class, although bog elements Narthecion scardici from Sphagno warnstorfii-Tomentypnion (Carex pauciflora, Eriophorum vaginatum) occur frequently. nitentis. The plots from the original diagnosis of Rhyn- In Central Europe, this vegetation is restricted to small chosporetum albae Koch 1926 were close to the core of patches in bog hollows, having a limited extent, whereas Caricion fuscae. in N Europe it can cover larger areas, especially in flat landscapes. Unsupervised classification The ISOPAM classification into six or seven clusters could Ordination be interpreted analogously to the supervised classification The first axis of DCA based on the data set of core plots of presented above (Appendix S8), reflecting the main eco- particular alliances (Fig. 3) runs from calcareous and sub- logical types (i.e. calcareous and extremely rich fens, rich halophytic fens (Caricion atrofusco-saxatilis, Caricion daval- fens, moderately rich fens, poor fens and dystrophic hol- lianae, Caricion viridulo-trinervis), through rich and moder- lows). Although there were some differences in diagnostic ately rich fens (Anagallido tenellae-Juncion bulbosi, Caricion species and in the representation of plant communities fuscae, Caricion stantis, Drepanocladion exannulati, Narthecion among particular areas, the main vegetation pattern and scardici, Sphagno warnstorfii-Tomentypnion nitentis, Saxifrago- units were generally consistent across regions and plot Tomentypnion), to poor fens (Sphagno-Caricion canescentis) sizes. The high consistency in main diagnostic species and dystrophic hollows (Scheuchzerion palustris). The sec- among the study areas was observed for clusters related to ond axis reflects the geographic–macroclimatic gradient Sphagno warnstorfii-Tomentypnion nitentis (Sphagnum warn- from the Atlantic alliances (Anagallido tenellae-Juncion bul- storfii, Tomentypnum nitens), Sphagno-Caricion canescentis bosi, Caricion viridulo-trinervis) through the widespread alli- (Sphagnum recurvum agg.), Scheuchzerion palustris (Drepa- ances, to arcto-alpine and arcto-boreo-alpine alliances nocladus fluitans, Scheuchzeria palustris, Sphagnum cuspida- (Caricion atrofusco-saxatilis, Caricion stantis, Drepanocladion tum, S. majus)andStygio-Caricion limosae (Calliergon exannulati). Narthecion scardici and Stygio-Caricion limosae trifarium, Carex limosa, Scorpidium scorpioides, Utricularia

Fig. 3. DCA of core plots (i.e. plots formally assigned to alliances) with centroids of particular clusters (alliances) along the ﬁrst three ordination axes. Eigenvalues: 1st axis (DCA1) 0.595, 2nd axis (DCA2) 0.430, 3rd axis (DCA 3) 0.378. CvT = Caricion viridulo-trinervis,Cd= Caricion davallianae, CaS = Caricion atrofusco-saxatilis,SCl= Stygio-Caricion limosae,SwT= Sphagno warnstorﬁi-Tomentypnion nitentis,SaT= Saxifrago-Tomentypnion, Ns = Narthecion scardici,Cs= Caricion stantis,AJ= Anagallido tenellae-Juncion bulbosi,De= Drepanocladion exannulati,Cf= Caricion fuscae, SCc = Sphagno-Caricion canescentis,Sp= Scheuchzerion palustris. Black points refer to plots from the original diagnosis of the association Rhynchosporetum albae Koch 1926 (valid nomenclature type of Rhynchosporion albae Koch 1926).

Applied Vegetation Science Doi: 10.1111/avsc.12271 © 2016 International Association for Vegetation Science 135 Classification of European fens T. Peterka et al. intermedia agg., U. minor agg.), the latter in boreal and arc- The large set of vegetation plots used in this study com- tic regions. On the Balkan Peninsula, the Sphagno warnstor- prised primary data from different sources, regions and fii-Tomentypnion nitentis alliance was mostly characterized time periods, which maximized its representativeness. The by meadow species, though the typical peat moss of this pre-selection of fen data allowed us to create relatively community (Sphagnum contortum) also displayed high fide- straightforward Cocktail groups that adequately reflected lity to the corresponding cluster. In the NE European sub- internal variation within the class Scheuchzerio palustris-Car- set, Sphagno warnstorfii-Tomentypnion nitentis was separated icetea fuscae and enabled the development of definitions of from Saxifrago-Tomentypnion, though only in the data set of biogeographically or ecologically unique alliances within plots of 1 m2. For the 100-m2 plot data set, two rich-fen fens. Finally, 38.6% of plots met the criteria of the formal clusters each containing plots of both alliances emerged. definitions, 33.7% were assigned using a similarity index, Likewise, Caricion davallianae was repeatedly identified and 27.7% remained unclassified. These figures corre- as a separate type and characterized, for example, by Carex spond to analogous studies across different vegetation davalliana, Drepanocladus revolvens agg., Eleocharis quinque- types (Kocı et al. 2003; Rolecek 2007; Rodrıguez-Rojo flora, Palustriella commutata s. l. and the absence of sphagna, et al. 2014; Douda et al. 2016). It is important to note that though more clusters were established in the Alps, the Bal- several alliances were defined by simple and therefore kans and Central Europe (for the plot size of 16 m2). In easily comprehensible and robust definitions. This holds both Scandinavian subsets, Caricion davallianae was well true for both widespread alliances (Sphagno-Caricion canes- delimited against Caricion atrofusco-saxatilis. Higher variabil- centis, Sphagno warnstorfii-Tomentypnion nitentis) and regio- ity in sets of diagnostic species was apparent for moderately nal alliances (Anagallido tenellae-Juncion bulbosi, Narthecion rich fens of the Caricion fuscae alliance, which are character- scardici). Construction of more complicated definitions ized by either species of mire meadows (Epilobium palustre, could result in more classified plots, but such definitions Galium uliginosum, Ranunculus flammula) or specialists of might be too complex and hard to comprehend (Rolecek waterlogged mires (Carex diandra, Drosera intermedia, Rhyn- 2007). In some cases, more complex definitions were how- chospora fusca) in Central and NE Europe subsets. In the ever necessary, most apparently in the case of Caricion fus- Balkan subset, the Caricion fuscae-related cluster is charac- cae, which is a traditionally recognized, widely distributed terized by Sphagnum subsecundum and Carex nigra.Separate alliance unequivocally understood as vegetation of moder- clusters corresponding to the Drepanocladion exannulati alli- ately rich fens. In these terms, it corresponds to the ‘central ance appeared in the Scandinavian and high-arctic subsets, syntaxon’ which has no diagnostic species (Willner 2006) while in the Alps Caricion fuscae and Drepanocladion exannu- or perhaps very few diagnostic species with a weak diag- lati formed a joint cluster. As a geographically constrained nostic value. Caricion fuscae is characterized by the occur- vegetation type, a cluster corresponding to Narthecion scar- rence of moderately rich-fen species with a broad dici appeared in the Balkan subset. ecological niche and the scarcity of species indicating extreme pH values. In contrast to Anagallido tenellae-Jun- Discussion cion bulbosi and Narthecion scardici, it also lacks geographically restricted species. This fact results in indistinct Methodological aspects and constraints of a broad-scale differentiation from other alliances (Table 1) and variable fen classification diagnostic species in the Caricion fuscae-related clusters in The most prominent purpose of vegetation classification unsupervised regional classifications (Appendix S8). is defining distinct objects for habitat conservation, mon- An important issue related to broad-scale vegetation itoring and ecological research. From this perspective, analyses is variation in plot size (Chytry&Ot ypkov a 2003; distinguished units are useful only if they reflect differ- Dengler et al. 2009). As demonstrated in a previous study ent habitat conditions, vegetation history or geographic on European fens (Jimenez-Alfaro et al. 2014), a restric- distribution (Willner 2006). This fact advocates, in our tion to a narrow range of plot sizes would lead to a loss of a opinion, application of supervised methods and prevents vast majority of data from important regions. Following a the application of fully automated procedures and the suggestion from Dengler et al. (2009), we made a posteriori classification of all available plots over large geographic assessment of the potential influence of different plot sizes extents. This study is the first attempt to synthesize phy- via presentation of their means and ranges for particular tosociological data and to create a unified classification clusters. The ranges of plot sizes among all alliances of the Scheuchzerio palustris-Caricetea fuscae class at the (Table 1) were almost equal, with the exception of Caricion continental scale based on individual vegetation plots. It atrofusco-saxatilis and Caricion stantis, which, however, could therefore serve as a state-of-the-art baseline for occupy naturally small areas. Drepanocladion exannulati and further development of the pan-European fen typology Stygio-Caricion limosae were also sampled using smaller sizes on various hierarchical levels. in their centre of distribution in Scandinavia than in other

Applied Vegetation Science 136 Doi: 10.1111/avsc.12271 © 2016 International Association for Vegetation Science T. Peterka et al. Classification of European fens regions due to the regional sampling tradition. Generally, The gradient of water table depth was not evident in the our formal definitions were able to classify plots across dif- pan-European DCA, although it has been repeatedly ferent sizes. Unsupervised classification (Appendix S8) reported to be important locally (Bragazza & Gerdol 1996; reproduced identical vegetation types, largely correspond- Jabłonska et al. 2011; Moeslund et al. 2013) or regionally ing to alliances, even when different plot sizes were used. (Schenkova et al. 2014; Horsakov a et al. 2015; Perez- These findings indicate that the quality of individual vege- Haase 2015). This fact suggests that the variation in species tation-plot records (i.e. precision in plant identification or composition related to water table should not be used as degree of plot homogeneity) might have a greater potential themainclassificationcriterionwithinfensonabroadscale to confound classification than plot size, at least for wet- without also taking into account shifts in species composi- land vegetation. tion driven by site chemistry. The second most important gradient in the species composition of European fens followed macroclimatic and biogeographic differences, advo- Vegetation units and classification criteria cating acceptance of geographically constrained alliances. There has been a long-standing debate about the ‘ideal’ Since base richness and biogeographic influences were fen classification without reaching any general consensus identified as the most important determinants of species (Rybnıcek 1981, 1985; Dierssen 1982; Dierssen & Dierssen composition, they should be used as the key criteria to 1985; Malmer 1985; Hajek et al. 2002; Dengler et al. define fen alliances. The reproduced alliances show eco- 2004; Lapshina 2010; Peterka et al. 2014). The system as logically and biogeographically meaningful diagnostic proposed here is similar to the classification established by species, underlying the applicability of the presented Fennoscandian authors (Sjors€ 1948; Dahl 1956; Ruu- classification. hijarvi€ 1960; Fransson 1972; Elveland 1976; Moen et al. 2012) and deepened further by Czech and Slovak Less well-known alliances (Rybnıcek et al. 1984; Dıte et al. 2007; Hajek & Hajkov a 2011) and some German (Passarge 1964; Succow 1974), Some less well-known alliances described from specific Polish (Pałczynski 1975), Russian (Koroleva 2001), Bul- parts of Eurasia proved to be clearly defined composition- garian (Tzonev et al. 2009) and, to a certain extent, Irish ally on the European scale, which was supported by DCA. (Crıodain & Doyle 1994) authors. In these classification Some geographically constrained fen alliances delimited in systems, major vegetation units differ in particular in their this study also show distinct environmental conditions or a position on the poor-rich gradient. Our pan-European specific history. study has confirmed that this gradient is the principal one The Caricion viridulo-trinervis alliance represents sub- even at the continental scale. An alternative approach halophytic fens with a shallow peat layer occurring in uses the dominance of selected vascular plants along with coastal dune slacks. This well-defined habitat was a subject hydrological characteristics (i.e. water table depth) as the of ecological studies in W Europe (Grootjans et al. 1991; main alliance-delimiting criteria (e.g. Oberdorfer 1957; Lammerts et al. 1999) and served as a postglacial refugium Dierssen 1982; Steiner 1992; Matuszkiewicz 2007). A of some continental wetland species such as Blysmus rufus main disadvantage of this approach is that vascular plants (Hajkov aetal.2015). whose dominance was used as the chief classification The Caricion atrofusco-saxatilis alliance comprises arctic criterion (Carex lasiocarpa, C. limosa, C. nigra, C. rostrata or and high-mountain calcareous fens delimited from the Rhynchospora alba) occur across a broad range of base rich- widespread Caricion davallianae by species with an arcto- ness (Gerdol 1995; Martincic 1997), for which reason the alpine distribution. We confirmed the occurrence of this resulting classification fails to mirror the major composi- alliance in the Alps and in N Europe, though it has also tional gradient within fens. Furthermore, this approach been reported in Scotland (Rodwell 1991), Spain (Rivas- does not consider bryophytes in the delimitation of alli- Martınez 2011) and the Romanian Carpathians (Coldea ances, thus ignoring the general knowledge that bryo- et al. 2008). Further research is therefore needed to com- phytes precisely indicate habitat conditions (Malmer et al. pare communities in these regions with those from the 1994; Peterka et al. 2014) and have a crucial importance Alps, Scandinavia and the European Arctic. for the functioning of mire habitats (Jones et al. 1994). The Narthecion scardici alliance is an alpine relict alliance The classification proposed here refutes the concepts of of the Balkans similar to rich and moderately rich fens of the extremely wide Caricion lasiocarpae and Rhynchosporion Sphagno warnstorfii-Tomentypnion nitentis and Caricion fuscae, albae alliances and accepts the Stygio-Caricion limosae and though virtually lacking boreal species of fens and har- Sphagno warnstorfii-Tomentypnion nitentis alliances. This bouring specific ecotypes of temperate fen plants instead solution was also supported by the results of a set of (Hajkov a et al. 2008). Lakusic (1968) originally considered regional unsupervised classifications. this alliance to be restricted to Kosovo and Macedonia.

Probably due to the absence of Narthecium scardicum,the of Saxifrago-Tomentypnion to the boreal-continental regions corresponding communities in Bulgaria were previously is probably shaped by improved phosphorus availability assigned to Caricion fuscae (Roussakova 2000; Tzonev et al. that would lead to the development of grassland or reed 2009). However, the community is well defined by Balkan vegetation in warmer regions with a longer growing sea- endemics and relicts (Lakusic&Grgic 1971; Lakusic 1973; son. In Finland, occurrence of this vegetation has been Roussakova 2000) and displays sympatric distribution with connected to high phosphorus availability and vivianite Caricion fuscae on the Balkan Peninsula (Kojic et al. 1998), deposits (Kotilainen 1944), although this pattern has not the latter occurring below the timberline in managed fen been studied in detail. The combination of a boreal-conti- grasslands (Hajek et al. 2008). nental climate and good phosphorus availability leads to The Anagallido tenellae-Juncion bulbosi alliance is con- unique species combinations, such as the co-existence of fined to a strongly oceanic climate without significant bor- Saxifraga hirculus with species of productive grasslands in eal or continental influences, together with intermediate pristine, undisturbed fens. levels of pH and mineral richness. Nevertheless, classification of fens in Atlantic Europe deserves further research, Acknowledgements since fen communities in this region frequently share some species with the Oxycocco-Sphagnetea and Littoreletea uniflorae We are grateful to Ilona Knollova for the preparation of classes (Fernandez Prieto et al. 1987; Rodwell 1991; Heras EVA data and technical help, to Lubomır Tichy for help et al. 2011). with the JUICE software, and to all our colleagues who The Caricion stantis alliance occurs in Europe as part of its helped with data compilation (namely Olga Galanina, broader circumpolar distribution range. It was described Raimo Heikkila,€ Oleg Kuznetsov, Corrado Marcenoand in northern Siberia (Matveyeva 1994) as vegetation of Vladimir Ranđelovic) or assisted with the fieldwork. Our sedge-brown-moss fens with scattered sphagna dominated special thanks go to several generations of European vege- mostly by Carex aquatilis subsp. stans or Dupontia fisheri. tation scientists who gathered the original vegetation-plot Analogous communities were documented throughout data in the field. Angelika Schwabe-Kratochwil and the circumpolar arctic zone (Thannheiser 1976; Hadac anonymous referees provided useful comments. The 1989; Lavrinenko et al. 2016). We consider this alliance as research of the leading authors was funded by the Czech the high-arctic vicariant of the rich to moderately rich fens Science Foundation (Centre of Excellence Pladias; 14- found in the south. However, the internal variability of the 36079G) and Masaryk University (MUNI/A/1048/2015). alliance and the relationships to other syntaxa deserve fur- The research of P.H. was supported by The Czech Academy ther research. of Sciences (project RVO 67985939). The Saxifrago-Tomentypnion alliance was described in western Siberia (Lapshina 2010) and has been distin- References guished for Europe on the basis of data from NE Europe and Poland. An analogous vegetation type is included in Bergamini, A., Pauli, D., Peintinger, M. & Schmid, B. 2001. Rela- the Finnish mire site type classification (Eurola et al. 2015) tionships between productivity, number of shoots and num- as ‘Eutrophic diandra-hirculus birch fen’. Its species com- ber of species in bryophytes and vascular plants. Journal of position is similar to Caricion davallianae, Sphagno warnstor- Ecology 89: 920–929. fii-Tomentypnion nitentis and tall-sedge vegetation, but it Bergamini, A., Peintinger, M., Fakheran, S., Moradi, H., Schmid, includes mostly brown-moss vegetation (as opposed to B. & Joshi, J. 2009. Loss of habitat specialists despite conser- Sphagno warnstorfii-Tomentypnion nitentis) and contains true vation management in fen remnants 1995–2006. Perspectives fen species (as opposed to tall-sedge vegetation). Although in Plant Ecology, Evolution and Systematics 11: 65–79. overall macronutrient availability is very low in all fens, Boyer, M.L.H. & Wheeler, B.D. 1989. Vegetation patterns in the nutrient ratios account for the compositional unique- spring-fed calcareous fens: calcite precipitation and con- – ness of Saxifrago-Tomentypnion. The species composition of straints on fertility. Journal of Ecology 77: 597 609. most calcium-rich fens is shaped by a shortage of phospho- Bragazza, L. & Gerdol, R. 1996. Response surfaces of plant species along water-table depth and water pH gradients in a poor rus caused by its immobilization during carbonate precipi- mire on the Southern Alps. Annales Botanici Fennici 33: 11–20. tation (Boyer & Wheeler 1989). In contrast, Saxifrago- Bruelheide, H. 1995. Die Grunlandgesellschaften€ des Harzes und Tomentypnion fens show better phosphorus availability, ihre Standortsbedingungen. Mit einem Beitrag zum Glieder- and therefore a lower N:P ratio in the biomass, and their ungsprinzip auf der Basis von statistisch ermittelten Arten- productivity seems to be N-limited rather than P-limited gruppen. Dissertationes Botanicae 244: 1–338. (Pawlikowski et al. 2013). Consequently, most of the Cari- Bruelheide, H. 1997. Using formal logic to classify vegetation. cion davallianae species are rare, while nutrient-demanding Folia Geobotanica 32: 41–46. species occur quite frequently. The geographic restriction

Bruelheide, H. 2000. A new measure of fidelity and its applica- Douda, J., Boublık, K., Slezak, M., Biurrun, I., Nociar, J., Havr- tion to defining species groups. Journal of Vegetation Science dova, A., Doudova, J., Acic, S., Brisse, H. (...) & Zimmer- 11: 167–178. mann, N.E. 2016. Vegetation classification and Chytry, M. & Otypkov a, Z. 2003. Plot sizes used for phytosocio- biogeography of European floodplain forests and alder carrs. logical sampling of European vegetation. Journal of Vegetation Applied Vegetation Science 19: 147–163. Science 14: 563–570. Du Rietz, G.E. 1949. Huvudenheter och huvudgranser€ i svensk Chytry, M., Tichy, L., Holt, J. & Botta-Dukat, Z. 2002. Determi- myrvegetation. Svensk Botanisk Tidskrift 43: 279–304. nation of diagnostic species with statistical fidelity measures. Elveland, J. 1976. Myrar pa Storon€ vid norrbottenskusten. Journal of Vegetation Science 13: 79–90. Wahlenbergia 3: 1–274. Chytry, M., Hennekens, S.M., Jimenez-Alfaro, B., Knollova, I., Ermakov, N. 2012. Prodromus vysshikh edinits rastitel’nosti Ros- Dengler, J., Jansen, F., Landucci, F., Schaminee, J.H.J., Acic, sii. In: Mirkin, B.M. & Naumova, L.G. (eds.) Sovremennoye sos- S. (...) & Yamalov, S. 2016. European Vegetation Archive toyanie osnovnykh kontseptsii nauki o rastitel’nosti, pp. 377–483. (EVA): an integrated database of European vegetation plots. Gilem, Ufa, RU [In Russian]. Applied Vegetation Science 19: 173–180. Eurola, S., Huttunen, A., Kaakinen, E., Kukko-oja, K., Saari, V. Coldea, G., Sanda, V., Popescu, A. & Stefan, N. 1997. Les associa- &Salonen,V.2015.Sata suotyyppia€ – Opas Suomen suokasvil- tions vegetales de Roumanie. Tome 1. Les associations herbacees nat- lisuuden tuntemiseen. Thule Institute, University of Oulu, urelles. Presses Universitaires, Cluj, RO. Oulu, FI. Coldea, G., Filipas, L. & Stoica, I.-A. 2008. Contributions to Felbaba-Klushyna, L. 2010a. Prodromus syntaksoniv roslynnosti Romanian vegetation studies (IV). Contributii Botanice 43: bolit i kholodnykh dzherel Ukrayinskykh Karpat (Klasy 45–52. Scheuchzerio-Caricetea fuscae Tx. 1937, Oxycocco-Sphagnetea Br.- Dahl, E. 1956. Rondane. Mountain vegetation in south Norway Bl. et Tx. ex Westhoff et al. 1946, Montio-Cardaminetea Br.-Bl. and its relation to the environment. Skrifter utgitt av Det et Tx. ex Klika et Hadac1944).Naukovyi visnyk Uzhhorodskoho Norske Videnskaps-Akademi i Oslo, Mat.-Naturvidensk. Klasse 3: universytetu, ser. Biolohiya 28: 73–82. [In Ukrainian.] 1–374. Felbaba-Klushyna, L. 2010b. Roslynnyi pokryv bolit i vodoim verkhi- De Caceres, M. & Wiser, S.K. 2012. Towards consistency in vya baseinu r. Tysa (Ukrayinski Karpaty) ta flyuvialna kontsept- vegetation classification. Journal of Vegetation Science 23: siya yoho okhorony. Lira, Uzhhorod, UA [In Ukrainian]. 387–393. Fernandez Prieto, J.A., Fernandez Ordonez,~ C. & Collado, M.A. De Caceres, M., Chytry, M., Agrillo, E., Attorre, F., Botta-Dukat, 1987. Datos sobre la vegetacion de las «turberas de Z., Capelo, J., Czucz, B., Dengler, J., Ewald, J. (...)&Wiser, esfagnos» galaico-asturianas y orocantabricas. Lazaroa 7: S.K. 2015. A comparative framework for broad-scale plot- 443–471. based vegetation classification. Applied Vegetation Science 18: Fischer, H.S. 2015. On the combination of species cover values 543–560. from different vegetation layers. Applied Vegetation Science 18: Dengler, J., Koska, I., Timmermann, T., Berg, C., Clausnitzer, U., 169–170. Isermann, M., Linke, C., Pazolt,€ J., Polte, T. & Spangenberg, Fransson, S. 1972. Myrvegetation i sydvastra€ Varmland.€ Acta A. 2004. New descriptions and typifications of syntaxa Phytogeographica Suecica 57: 1–133. within the project ‘Plant communities of Mecklenburg-Vor- Frey, W., Frahm, J.-P., Fischer, E. & Lobin, W. 2006. The liver- pommern and their vulnerability’ – part II. Feddes Repertorium worts, mosses and ferns of Europe. Harley Books, Colchester, 115: 343–392. UK. Dengler, J., Lobel,€ S. & Dolnik, C. 2009. Species constancy Gerdol, R. 1995. Community and species-performance patterns depends on plot size – a problem for vegetation classification along an alpine poor–rich mire gradient. Journal of Vegetation and how it can be solved. Journal of Vegetation Science 20: Science 6: 175–182. 754–766. Gerdol, R. & Tomaselli, M. 1997. Vegetation of wetlands in the Dengler, J., Jansen, F., Glockler,€ F., Peet, R.K., De Caceres, M., Dolomites. Dissertationes Botanicae 281: 1–197. Chytry, M., Ewald, J., Oldeland, J., Lopez-Gonzalez, G. (...) Graf, U., Wildi, O., Feldmeyer-Christe, E. & Kuchler,€ M. 2010. A & Spencer, N. 2011. The Global Index of Vegetation-Plot phytosociological classification of Swiss mire vegetation. Databases (GIVD): a new resource for vegetation science. Botanica Helvetica 120: 1–13. Journal of Vegetation Science 22: 582–597. Grootjans, A.P., Hartog, P.S., Fresco, L.F.M. & Esselink, H. 1991. Dierssen, K. 1982. Die wichtigsten Pflanzengesellschaften der Moore Succession and fluctuation in a wet dune slack in relation NW-Europas. Conservatoire et Jardin botaniques, Geneve, CH. to hydrological changes. Journal of Vegetation Science 2: Dierssen, K. & Dierssen, B. 1985. Suggestions for common 545–554. approach in phytosociology for Scandinavian and Central Grootjans, A.P., Bulte, M., Wołejko, L., Pakalne, M., Dullo, B., European mire ecologists. Aquilo Series Botanica 21: 33–44. Eck, N. & Fritz, C. 2015. Prospects of damaged calcareous Dıte, D., Hajek, M. & Hajkov a, P. 2007. Formal definitions of Slo- spring systems in temperate Europe: can we restore traver- vakian mire plant associations and their application in regio- tine-marl deposition? Folia Geobotanica 50: 1–11. nal research. Biologia 62: 400–408.

Hadac, E. 1989. Notes on plant communities of Spitsbergen. Folia Jimenez-Alfaro, B., Hajek, M., Ejrnæs, R., Rodwell, J., Paw- Geobotanica et Phytotaxonomica 24: 131–169. likowski, P., Weeda, E.J., Laitinen, J., Moen, A., Bergamini, Hajek, M. & Hajkov a, P. 2011. Vegetace slatinist, prechodovych A. (...)&Dıaz, T.E. 2014. Biogeographic patterns of base- raselinistavrchovistnıch slenku(Scheuchzerio palustris-Carice- rich fen vegetation across Europe. Applied Vegetation Science tea nigrae). In: Chytry, M. (ed.) Vegetace Cesk e republiky 3. 17: 367–380. Vodnı a mokradnıvegetace, pp. 614–704. Academia, Praha, CZ. Jones, C.G., Lawton, J.H. & Shachak, M. 1994. Organisms as Hajek, M., Hekera, P. & Hajkov a, P. 2002. Spring fen vegetation ecosystem engineers. Oikos 69: 373–386. and water chemistry in the Western Carpathian flysch zone. Knollova, I., Chytry, M., Tichy, L. & Hajek, O. 2005. Stratified Folia Geobotanica 37: 205–224. resampling of phytosociological databases: some strategies Hajek, M., Horsak, M., Hajkov a, P. & Dıte, D. 2006. Habitat for obtaining more representative data sets for classification diversity of central European fens in relation to environmen- studies. Journal of Vegetation Science 16: 479–486. tal gradients and an effort to standardise fen terminology in Koch, W. 1926. Die Vegetationseinheiten der Linthebene ecological studies. Perspectives in Plant Ecology, Evolution and unter Berucksichtigung€ der Verhaltnisse€ in der Nord- Systematics 8: 97–114. ostschweiz. Systematisch-kritische Studie. Jahrbuch Hajek, M., Hajkov a, P. & Apostolova, I. 2008. New plant associa- der St. Gallischen Naturwissenschaftlichen Gesellschaft 61/2: tions from Bulgarian mires. Phytologia Balcanica 14: 377–399. 1–144. Hajek, M., Pleskova, Z., Syrovatka, V., Peterka, T., Laburdova, J., Koch, M. & Jurasinski, G. 2014. Four decades of vegetation devel- Kintrova, K., Jirousek,M.&Hajek, T. 2014. Patterns in moss opment in a percolation mire complex following intensive element concentrations in fens across species, habitats, and drainage and abandonment. Plant Ecology & Diversity 8: 49–60. regions. Perspectives in Plant Ecology, Evolution and Systematics Kocı, M., Chytry, M. & Tichy, L. 2003. Formalized reproduction 16: 203–218. of an expert-based phytosociological classification: a case Hajek, M., Jirousek, M., Navratilov a, J., Horodyska, E., Peterka, study of subalpine tall-forb vegetation. Journal of Vegetation T., Pleskova, Z., Navratil, J., Hajkov a, P. & Hajek, T. 2015. Science 14: 601–610. Changes in the moss layer in Czech fens indicate early suc- Kojic, M., Popovic, R. & Karazic, B. 1998. Sintaksonomski pregled cession triggered by nutrient enrichment. Preslia 87: 279– vegetacije Srbije.Inst.Biol.Res.“Sinisa Stankovic”, Beograd, YU. 301. Koroleva, N.E. 2001. Sintaksonomicheskii obzor bolot tun- Hajkov a, P., Hajek, M., Apostolova, I., Zeleny, D. & Dıte, D. drovogo poyasa Chibinskikh gor (Murmanskaya oblast’). 2008. Shifts in the ecological behaviour of plant species Rastitel’nost’ Rossii 2: 49–57 [In Russian.]. between two distant regions: evidence from the base rich- Koska, I. & Timmermann, T. 2004. Parvo-Caricetea den Held & ness gradient in mires. Journal of Biogeography 35: 282–294. Westhoff in Westhoff & den Held 1969 nom. cons. propos. – Hajkov a, P., Petr, L., Horsak, M., Rohovec, J. & Hajek, M. 2015. Riede und Rohrichte€ maßig€ nahrstoffarmer€ Niedermoore Interstadial inland dune slacks in south-west Slovakia: a und Ufer. In: Berg, C., Dengler, J., Abdank, A. & Isermann, multi-proxy vegetation and landscape reconstruction. Qua- M. (eds.) Die Pflanzengesellschaften Mecklenburg-Vorpommerns ternary International 357: 314–328. und ihre Gefahrdung€ , pp. 163–195. Weissdorn, Jena, DE. Heikkila,€ H. 1987. The vegetation and ecology of mesotrophic Kotilainen, M. 1944. Kasvit erikoislaatuisen substraatin and eutrophic fens in western Finland. Annales Botanici Fen- indikaattoreina. Societas Scientiarum Fennica 22: 1–18. nici 24: 155–175. Kuznetsov, O. 2003. Topological–ecological classification of mire Heras, P., Infante, M., Biurrun, I., Campos, J.A. & Berastegi, A. vegetation in the Republic of Karelia (Russia). The Finnish 2011. Tipologıa, vegetacion y estado de conservacion de los Environment 485: 117–123. habitats hidroturbosos del noroeste de Navarra. Acta Botanica Lajer, K. 1998. Bevezetes a magyarorszagi Lapok vegetaci o- Barcinonensia 53: 27–45. okol€ ogi aj aba. Tilia 6: 84–238. Horsakov a, V., Horsak, M., Hajek, M., Hajkov a, P. & Dıte, D. Lakusic, R. 1968. Planinska vegetacija jugoistocnih Dinarida. 2015. Mollusc assemblages of Scandinavian fens: species Glasnik Republickog zavoda za zastitu prirode i prirodnjacke zbirke composition in relation to environmental gradients and veg- 1: 9–75. etation. Annales Zoologici Fennici 52: 1–16. Lakusic, R. 1973. Narthecietalia ordo novus der Scheuchzerio-Cari- Jabłonska, E., Pawlikowski, P., Jarzombkowski, F., Chormanski, cetea fuscae Nordh. 1936 in den sudeurop€ aischen€ Gebirgen. J., Okruszko, T. & Kłosowski, S. 2011. Importance of water Ver€offentlichungen des Geobotanischen Institutes der ETH, Stiftung level dynamics for vegetation patterns in a natural percolation Rubel€ 51: 158–162. mire (Rospuda fen, NE Poland). Hydrobiologia 674: 105–117. Lakusic, R. & Grgic, P. 1971. Ekologija i rasprostiranje ende- Jansen, F. & Dengler, J. 2010. Plant names in vegetation data- micnih vrsta Narthecium scardicum Kos., Pinguicula balcanica bases – a neglected source of bias. Journal of Vegetation Science Cas., Gymnadenia friwaldii Hampe i Silene asterias Grsb. Ekolo- 21: 1179–1186. gija 6: 337–350. Jermacane, S. & Laivins, M. 2001. Latvija aprakstıto Lammerts, E.J., Pegtel, D.M., Grootjans, A.P. & Veen, A.V.D. augu sabiedrıbu sintaksonu saraksts. Latvijas Veģetacija 4: 1999. Nutrient limitation and vegetation changes in a coastal 115–132. dune slack. Journal of Vegetation Science 10: 111–122.

Landucci, F., Tichy, L., Sumberov a, K. & Chytry, M. 2015. For- and algal communities. Applied Vegetation Science 19, Suppl. 1: malized classification of species-poor vegetation: a proposal 3–264. of a consistent protocol for aquatic vegetation. Journal of O’Crıódaín, C. & Doyle, G.J. 1994. An overview of Irish small- Vegetation Science 26: 791–803. sedge vegetation: syntaxonomy and a key to communities Lapshina, E. 2010. Rastitel’nost’ bolot yugo–vostoka Zapadnoi Sibiri. belonging to the Scheuchzerio-Caricetea nigrae (Nordh. Izdatel’stvo NGU, Novosibirsk, RU [In Russian]. 1936) Tx. 1937. Biology and Environment: Proceedings of the Lavrinenko, O.V., Matveyeva, N.V. & Lavrinenko, I.A. 2016. Royal Irish Academy 94B: 127–144. Soobshchestva klassa Scheuchzerio-Caricetea nigrae (Nordh. Oberdorfer, E. 1957. Suddeutsche€ Pflanzengesellschaften. 1936) Tx. 1937 v vostochnoevropeiskikh tundrakh. Rastitel’- Pflanzensoziologie 10: 1–564. nost’ Rossii 28: 63–96 [In Russian.]. Oberdorfer, E. (ed.) 1998. Suddeutsche€ Pflanzengesellschaften. Teil I: Lawesson, J.E. 2004. A tentative annotated checklist of Danish Fels- und Mauergesellschaften, alpine Fluren, Wasser-, Verlan- syntaxa. Folia Geobotanica 39: 73–95. dungs- und Moorgesellschaften, 4th ed. Gustav Fischer, Jena/ Lebrun, J., Noirfalise, A., Heinemann, P. & Vanden Berghen, C. Stuttgart, DE. 1949. Les Associations veg etales de Belgique. Bulletin de la Pałczynski, A. 1975. Bagna Jacwieskie (pradolina Biebrzy). Rocz- Societe Royale de Botanique de Belgique 82: 105–199. niki nauk rolniczych. Ser. D - Monografie 145: 1–232. Malmer, N. 1985. Remarks to the classification of mires and mire Passarge, H. 1964. Pflanzengesellschaften des nordostdeutschen vegetation – Scandinavian arguments. Aquilo Series Botanica Flachlandes I. Pflanzensoziologie 13: 1–324. 21: 9–17. Pawlikowski, P., Abramczyk, K., Szczepaniuk, A. & Kozub, Ł. Malmer, N. 1986. Vegetation gradients in relation to environ- 2013. Nitrogen:phosphorus ratio as the main ecological mental conditions in northwestern European mires. Cana- determinant of the differences in the species composition of dian Journal of Botany 64: 375–383. brown-moss rich fens in north-eastern Poland. Preslia 85: Malmer, N., Svensson, B.M. & Wallen, B. 1994. Interactions 349–367. between Sphagnum mosses and field layer vascular plants in Perez-Haase, A. 2015. Patrons estructurales, ecologics i biogeografics the development of peat-forming systems. Folia Geobotanica en vegetacio de molleres i de torberes d’esfagnes. PhD Thesis, et Phytotaxonomica 29: 483–496. University of Barcelona, Barcelona, ES. Malson,€ K., Backeus, I. & Rydin, H. 2008. Long-term Persson, A. 1961. Mire and spring vegetation in an area north of effects of drainage and initial effects of hydrological restora- lake Tornetrask,€ Torne Lappmark, Sweden. I. Description of tion on rich fen vegetation. Applied Vegetation Science 11: the vegetation. Opera Botanica 6: 1–187. 99–106. Peterka, T., Pleskova, Z., Jirousek,M.&Hajek, M. 2014. Testing Martincic, A. 1995. Vegetacija razreda Scheuchzerio-Caricetea fuscae floristic and environmental differentiation of rich fens on the (Nordh. 36) R. Tx. 37 v Sloveniji. Bioloski vestnik 40: 101– Bohemian Massif. Preslia 86: 337–366. 111. Peterka, T., Jirousek, M., Hajek, M. & Jimenez-Alfaro, B. 2015. Martincic, A. 1997. Ekolosko-fitocenoloske dvozivke. Acta Bio- European Mire Vegetation Database: a gap-oriented data- logica Slovenica 41: 43–60. base for European fens and bogs. Phytocoenologia 45: 291– Matuszkiewicz, W. 2007. Przewodnik do oznaczania zbiorowisk 298. roslinnych Polski, 3rd ed. Panstwove Wydawnictwo Nau- Pop, E. 1960. Mlastinile de turba din R.P. Romana. Editura Acade- kowe, Warszawa, PL. miei Republicii Populare Romane, Bucharest, RO. Matveyeva, N.V. 1994. Floristic classification and ecology of tun- Rivas-Martınez, S. (ed.) 2011. Mapa de series, geoseries y geop- dra vegetation of the Taymyr Peninsula, northern Siberia. ermaseries de vegetacion de Espana.~ Parte II. Itinera Geob- Journal of Vegetation Science 5: 813–828. otanica 18: 5–424. Michalcova, D., Lvoncık, S., Chytry, M. & Hajek,O.2011.Biasin Rodrıguez-Rojo, M.P., Fernandez-Gonz alez, F., Tichy, L. & vegetation databases? A comparison of stratified–random Chytry, M. 2014. Vegetation diversity of mesic grasslands and preferential sampling. Journal of Vegetation Science 22: (Arrhenatheretalia) in the Iberian Peninsula. Applied Vegetation 281–291. Science 17: 780–796. Moen, A., Lyngstad, A. & Øien, D. 2012. Boreal rich fen vegeta- Rodwell, J.S. (ed.) 1991. British plant communities, Vol. 2. Mires tion formerly used for haymaking. Nordic Journal of Botany and heaths. Cambridge University Press, Cambridge, UK. 30: 226–240. Rolecek, J. 2007. Formalized classification of thermophilous oak Moeslund, J.E., Arge, L., Bøcher, P.K., Dalgaard, T., Odgaard, forests in the Czech Republic: what brings the Cocktail M.V., Nygaard, B. & Svenning, J.-C. 2013. Topographically method? Preslia 79: 1–21. controlled soil moisture is the primary driver of local Roussakova, V. 2000. Veg etation alpine et sous alpine superiore vegetation patterns across a lowland region. Ecosphere 4: 1–26. de la montagne de Rila (Bulgarie). Braun-Blanquetia 25: 3– Mucina, L., Bu¨ ltmann, H., Dierßen, K., Theurillat, J.-P., Raus, T., 132. Cˇ arni, A., Sˇumberova´, K., Willner, W., Dengler, J. & Tichy´, Ruuhijarvi,€ R. 1960. Uber€ die regionale Einteilung der L. 2016. Vegetation of Europe: hierarchical floristic Nordfinnischen Moore. Annales Botanici Societatis Zoologicæ classification system of vascular plant, bryophyte, lichen, Botanicæ Fennicæ ‘Vanamo’ 31/1: 1–360.

Rybnıcek, K. 1981. Problematika klasifikace raselinnych Topic, J. & Stancic, Z. 2006. Extinction of fen and bog plants and spolecenstev. Zpravy Ceskoslovensk e botanickespolecnosti, Mate- their habitats in Croatia. Biodiversity and Conservation 15: rialy 2: 65–70. 3371–3381. Rybnıcek, K. 1985. A Central European approach to the Tutin, T.G., Heywood, V.H., Burges, N.A., Moore, D.M., Valen- classification of mire vegetation. Aquilo, Series Botanica 21: tine, D.H., Walters, S.M. & Webb, D.A. (eds.) 1968–1993. 19–31. Flora Europaea,Vols2–5, 2nd ed. of Vol. 1. Cambridge Rybnıcek, K., Balatov a-Tul ackova, E. & Neuhausl,€ R. 1984. University Press, Cambridge, UK. Prehled rostlinnych spolecenstev raselinist a mokradnıch Tzonev, R.T., Dimitrov, M.A. & Roussakova, V.H. 2009. Syntaxa luk Ceskoslovenska. Studie CSAV 1984: 1–124. according to the Braun-Blanquet approach in Bulgaria. Phy- Sanda, V., Popescu, A. & Stancu, D.I. 2001. Structura cenotica si tologia Balcanica 15: 209–233. caracterizarea ecologica a fitocenozelor din Romania. Editura Con- Udd, D., Malson,€ K., Sundberg, S. & Rydin, H. 2015. Explaining phis, Pitesti, RO. species distributions by traits of bryophytes and vascular Schaminee, J.H.J., Hennekens, S.M., Chytry, M. & Rodwell, J.S. plants in a patchy landscape. Folia Geobotanica 50: 161–174. 2009. Vegetation-plot data and databases in Europe: an over- Vanden Berghen, C. 1952. Contribution a l’etude des bas-marais view. Preslia 81: 173–185. de Belgique. Bulletin du Jardin Botanique de l’Etat aBruxelles Schenkova, V., Horsak, M., Hajek, M., Pleskova, Z., Dıte, D. & 22: 1–64. Pawlikowski, P. 2014. Mollusc and plant assemblages con- Willner, W. 2006. The association concept revisited. Phytocoenolo- trolled by different ecological gradients at Eastern European gia 36: 67–76. fens. Acta Oecologica 56: 66–73. Schmidtlein, S., Tichy, L., Feilhauer, H. & Faude, U. 2010. A brute-force approach to vegetation classification. Journal of Supporting Information Vegetation Science 21: 1162–1171. Sjors,€ H. 1948. Myrvegetation i Bergslagen. Acta Phytogeographica Additional Supporting Information may be found in the Suecica 21: 1–340. online version of this article: Sjors,€ H. 1952. On the relation between vegetation and electrolytes in north Swedish mire waters. Oikos 2: 241–258. Appendix S1. Data sources. Sjors,€ H. & Gunnarsson, U. 2002. Calcium and pH in north Appendix S2. List of aggregates, species complexes and and central Swedish mire waters. Journal of Ecology 90: problematic taxa merged at the species level. 650–657. Appendix S3. Lists of fen specialists (‘typical fen species’) Steiner, G.M. 1992. Osterreichischer€ Moorschutzkatalog, 4th ed. and non-fen species. Ulrich Moser, Graz/Wien, AT. Appendix S4. List of species groups based on the Cocktail Succow, M. 1974. Vorschlag einer systematischen Neugliede- method. rung der mineralbodenwasserbeeinflussten wachsenden Appendix S5. Formal definitions of alliances. Moorvegetation Mitteleuropas unter Ausklammerung des Appendix S6. Full version of the synoptic table with core Gebirgsraumes. Feddes Repertorium 85: 57–113. plots assigned to alliances on the basis of formal defini- Tahvanainen, T. 2004. Water chemistry of mires in relation to tions. – the poor rich vegetation gradient and contrasting geochemi- Appendix S7. Synoptic table with all plots assigned to the cal zones of north-eastern Fennoscandian shield. Folia Geo- alliances on the basis of both formal definitions and simi- – botanica 39: 353 369. larity in species composition established using the Fre- Thannheiser, D. 1976. Ufer-und Sumpfvegetation auf dem west- quency-Positive Fidelity Index (FPFI). lichen kanadischen Arktis-Archipel und Spitzbergen. Polar- Appendix S8. Results of the ISOPAM analyses. forschung 46: 71–82. Appendix S9. Description of unclassified group of plots. Tichy, L. 2002. JUICE, software for vegetation classification. Appendix S10. Reasons for the abandonment of the Journal of Vegetation Science 13: 451–453. Rhynchosporion albae and Caricion lasiocarpae alliances. Tichy, L. 2005. New similarity indices for the assignment of Appendix S11. releves to the vegetation units of an existing phytosociologi- Generalized classification scheme with cal classification. Plant Ecology 179: 67–72. distinguishing features of fen alliances. Tichy, L. & Chytry, M. 2006. Statistical determination of diagnostic species for site groups of unequal size. Journal of Vegetation Science 17: 809–818.

Applied Vegetation Science 142 Doi: 10.1111/avsc.12271 © 2016 International Association for Vegetation Science Appendix S1. Data sources. The following vegetation databases, with IDs according to the Global Index of Vegetation Plot Databases (Dengler et al. 2011), were made available for the present study through the European Vegetation Archive (Chytrý et al. 2016): Austrian Vegetation Database (EU-AT-001; Willner et al. 2012), Czech National Phytosociological Database (EU-CZ-001; Chytrý & Rafajová 2003), Dutch National Vegetation Database (EU-NL- 001; Schaminée et al. 2012), German Vegetation Reference Database – GVRD (EU-DE-014; Jandt & Bruelheide 2012), INBOVEG (EU-BE-002), Irish Vegetation Database (EU-IE-001; Weekes & FitzPatrick 2010), Lithuanian vegetation Database (EU-LT-001), Mire Vegetation Database of Latvia (EU-LV-002; Aunina 2012), National Vegetation Database of Denmark (EU-DK-002), Phytosociological Database of Non-Forest Vegetation in Croatia (EU-HR-001; Stančid 2012), Polish Vegetation Database (EU-PL-001; Kącki & Śliwioski 2012), Slovak Vegetation Database (EU-SK-001; Šibík 2012), SOPHY (EU-FR-003; Garbolino et al. 2012), UK National Vegetation Classification Database (EU-GB-001; Rodwell 2012), Ukrainian Grasslands Database (EU-UA-001; Kuzemko 2012), Vegetation Database of Slovenia (EU-SI-001; Šilc 2006), Vegetation Database of the Republic of Macedonia (EU-MK-001), Vegetation-Plot Database of the University of the Basque Country (EU-00-011; Biurrun et al. 2012), VegetWeb (EU-DE-013; Ewald et al. 2012), VegItaly (EU-IT-001; Landucci et al. 2012) and VegMV (EU-DE-001; Jansen et al. 2012). A large proportion of the vegetation plots comes from the gap- oriented European Mire Vegetation Database (EU-00-022; Peterka et al. 2015) which was created specifically for the purpose of this project. Other GIVD-registered data sources comprise the mire monitoring programme of the Swiss Federal Office for the Environment (Graf et al. 2010a,b; Ecker et al. 2010), the Belarus Peatland Restoration Project Database (EU-BY-001; Thiele et al. 2015) and the Iberian and Macaronesian Vegetation Information System – SIVIM (EU-ES-001). These data were supplemented with vegetation plots sampled by the authors during their own fieldwork, the private database of mire vegetation plots from Romania of C. Biţa- Nicolae and relevés which had been gathered by the late Kamil Rybníček and Emil Hadač and stored in the private archive of Michal Hájek. These data sets were accompanied by selected vegetation plots published by Lájer (1998) and Koch (1926). The latter paper contains two relevés on the basis of which the Rhynchosporion albae alliance was described. Data sources of plots from individual countries and regions, numbers of available fen plots before and after geographic stratification.

Country Data sources No. of fen plots No. of fen plots before geographic after geographic stratif. stratif. Austria EU-AT-001; private data (M. Hájek, P. Hájková, D. 1926 1524 Dítě) Belarus EU-BY-001 49 49 Belgium EU-BE-002, EU-FR-003 107 106 Bosnia and EU-00-022 41 41 Herzegovina Britain EU-GB-001 882 869 Bulgaria private data (M. Hájek, P. Hájková et al.) 227 227 Croatia EU-HR-001, EU-00-022 18 18 Czech Republic EU-CZ-001; private data (T. Peterka, M. Jiroušek, M. 2475 2045 Hájek, P. Hájková, Z. Plesková) Denmark EU-DK-002, EU-00-022 22 22 Estonia EU-00-022; private data (V. Smagin, M. Hájek, P. 101 101 Hájková, D. Dítě) Faroe Islands EU-00-022 8 8 Finland EU-00-022; private data (T. Tahvanainen) 1038 1038 France EU-FR-003 1170 914 Germany EU-DE-001, EU-DE-013, EU-DE-014 3724 2437 Greece private data (M. Hájek, P. Lazarevid) 15 15 Greenland EU-00-022 49 49 Hungary EU-SK-001; Lájer (1998) 13 13 Iceland EU-00-022 310 310 Ireland EU-IE-001; EU-00-022 3217 1398 Italy EU-IT-001, EU-00-022 245 235 Latvia EU-LV-002, EU-00-022, private data (L. Aunina, D. 1361 1361 Dítě, M. Hájek, P. Hájková)

142

Country Data sources No. of fen plots No. of fen plots before geographic after geographic stratif. stratif. Lithuania EU-LT-001; private data (D. Dítě, M. Hájek, P. 12 12 Hájková) Macedonia EU-MK-001; private data (T. Peterka et al.) 11 11 Montenegro EU-00-022 14 14 Netherlands EU-NL-001 1652 1311 Norway (except EU-00-022; private data (A. Moen, D. Dítě, M. Hájek, 2064 2064 Svalbard) P. Hájková, T. Peterka et al.) Poland EU-PL-001, EU-00-022; private data (P. Pawlikowski, 1303 1303 D. Dítě, M. Hájek, P. Hájková) Romania EU-00-022; private database of C. Bita-Nicolae; 201 201 private data (M. Hájek, P. Hájková, D. Dítě et al.) Russian EU-00-022; private data (T. Ivchenko, M. Jiroušek, N. 1322 1322 Federation Koroleva, E. Lapshina, M. Napreenko, V. Smagin) Serbia (incl. EU-00-022; private data (P. Lazarevid) 122 122 Kosovo) Slovak Republic EU-SK-001; private data (K. Rybníček, D. Dítě, M. 1828 1719 Hájek, P. Hájková) Slovenia EU-SI-001; private data (M. Hájek, P. Hájková, E. 227 227 Hettenbergerová, M. Jiroušek, T. Peterka) Spain + Andorra EU-00-011, EU-ES-001, EU-FR-003; private data (B. 803 600 Jiménez-Alfaro) Svalbard EU-00-022 134 134 Sweden EU-00-022, private data (T. Peterka, M. Jiroušek, E. 1577 1577 Hettenbergerová, Z. Plesková, D. Dítě, M. Hájek, P. Hájková) Switzerland EU-FR-003; mire monitoring programme of the Swiss 630 543 Federal Office for the Environment; private data (A. Bergamini, L. Sekulová, D. Dítě, M. Hájek, P. Hájková); Koch (1926) Ukraine EU-UA-001, EU-00-022; private data (L. Felbaba- 151 151 Klushyna, E. Hadač)

References Aunina, L. 2012. Mire Vegetation Database of Latvia. Biodiversity & Ecology 4: 410. Biurrun, I., García-Mijangos, I., Campos, J.A., Herrera, M. & Loidi, J. 2012. Vegetation-Plot Database of the University of the Basque Country (BIOVEG). Biodiversity & Ecology 4: 328. Chytrý, M. & Rafajová, M. 2003. Czech National Phytosociological Database: basic statistics of the available vegetation-plot data. Preslia 75: 1–15. Chytrý, M., Hennekens, S.M., Jiménez-Alfaro, B., Knollová, I., Dengler, J., Jansen, F., Landucci, F., Schaminée, J.H.J., (...) & Yamalov, S. 2016. European Vegetation Archive (EVA): an integrated database of European vegetation plots. Applied Vegetation Science 19: 173–180. Dengler, J., Jansen, F., Glöckler, F., Peet, R.K., De Cáceres, M., Chytrý, M., Ewald, J., Oldeland, J., Lopez-Gonzalez, G., (...) & Spencer, N. 2011. The Global Index of Vegetation-Plot Databases (GIVD): a new resource for vegetation science. Journal of Vegetation Science 22: 582–597. Ecker, K., Waser, L.T. & Küchler, M. 2010. Contribution of multi-source remote sensing data to predictive mapping of plant- indicator gradients within Swiss mire habitats. Botanica Helvetica 120: 29–42. Ewald, J., May, R. & Kleikamp, M. 2012. VegetWeb – the national online-repository of vegetation plots from Germany. Biodiversity & Ecology 4: 173–175. Garbolino, E., De Ruffray, P., Brisse, H. & Grandjouan, G. 2012. The phytosociological database SOPHY as the basis of plant socio-ecology and phytoclimatology in France. Biodiversity & Ecology 4: 177–184. Graf, U., Wildi, O., Feldmeyer-Christe, E. & Küchler, M. 2010a. A phytosociological classification of Swiss mire vegetation. Botanica Helvetica 120: 1–13. Graf, U., Wildi, O., Küchler, M. & Ecker, K. 2010b. Five-year changes in Swiss mire vegetation. Botanica Helvetica 120: 15– 27. Jandt, U. & Bruelheide, H. 2012. German Vegetation Reference Database (GVRD). Biodiversity & Ecology 4: 355. Jansen, F., Dengler, J. & Berg, C. 2012. VegMV – the vegetation database of Mecklenburg-Vorpommern. Biodiversity & Ecology 4: 149–160. Kącki, Z. & Śliwioski, M. 2012. The Polish Vegetation Database: structure, resources and development. Acta Societatis Botanicorum Poloniae 81: 75–79.

143

Koch, W. 1926. Die Vegetationseinheiten der Linthebene unter Berücksichtigung der Verhältnisse in der Nordostschweiz. Systematisch-kritische Studie. Jahrbuch der St. Gallischen Naturwissenschaftlichen Gesellschaft 61/2: 1–144. Kuzemko, A. 2012. Ukrainian Grasslands Database. Biodiversity & Ecology 4: 430. Lájer, K. 1998. Bevezetés a magyarországi Lápok vegetáció-ökológiájába. Tilia 6: 84–238. Landucci, F., Acosta, A.T.R., Agrillo, E., Attorre, F., Biondi, E., Cambria, V.M., Chiarucci, A., Del Vico, E., De Sanctis, M., Facioni, L., Geri, F., Gigante, D., Guarino, R., Landi, S., Lucarini, D., Panfili, E., Pesaresi, S., Prisco, I., Rosati, L., Spada, F. & Venanzoni, R. 2012. VegItaly: The Italian collaborative project for a national vegetation database. Plant Biosystems 146: 756–763 Peterka, T., Jiroušek, M., Hájek, M. & Jiménez-Alfaro, B. 2015. European Mire Vegetation Database: a gap-oriented database for European fens and bogs. Phytocoenologia 45: 291–298. Rodwell, J. 2012. UK National Vegetation Classification Database. Biodiversity & Ecology 4: 381. Schaminée, J.H.J., Hennekens, S.M. & Ozinga, W.A. 2012. The Dutch National Vegetation Database. Biodiversity & Ecology 4: 201–209. Šibík, J. 2012. Slovak Vegetation Database. Biodiversity & Ecology 4: 429. Šilc, U. 2006. Slovenian Phytosociology in a Database: state of the art, basic statistics and perspectives. Hladnikia 19: 27–34. Stančid, Z. 2012. Phytosociological Database of Non-Forest Vegetation in Croatia. Biodiversity & Ecology 4: 391. Thiele, A., Liaščynskaya, N., Broska, T., Bärisch, S., Skuratovič, A., Dubovik, D., Stepanovič, J., Ermolenko, G., Sozinov, O. & Sakovič, A. 2015. Belarus Peatland Database. Phytocoenologia 45: 399–400. Weekes, L.C. & FitzPatrick, Ú. 2010. The National Vegetation Database: Guidelines and Standards for the Collection and Storage of Vegetation Data in Ireland. Version 1.0. Irish Wildlife Manuals, No. 49. National Parks and Wildlife Service, Department of Environment, Heritage and Local Government, Dublin, IR. Willner, W., Berg, C. & Heiselmayer, P. 2012. Austrian Vegetation Database. Biodiversity & Ecology 4: 333.

Appendix S2. List of aggregates, species complexes and problematic taxa merged at the species level. The nomenclature used in this list generally follows that of Tutin et al. (1968–1993) and Frey et al. (2006). Aggregates (agg.) are used when individual species are merged. Sensu lato species (s.l.) are used mostly when subspecies are merged to the level of species. Note that these taxa are frequently treated at different taxonomic levels in different sources. Although they are listed as subspecies by Tutin et al. (1968–1993) and Frey et al. (2006), they are frequently treated as separate species. Therefore, the concept of sensu lato species is used. Name Included taxa Achillea millefolium agg. A. inundata, A. millefolium, A. pannonica, A. roseo-alba, A. stricta Agrostis stolonifera agg. A. gigantea, A. stolonifera Alchemilla alpina agg. A. alpina, A. basaltica Alchemilla vulgaris agg. all Alchemilla species with the exception of A. alpina agg. and A. glaucescens Anthoxanthum odoratum agg. A. alpinum, A. odoratum Arabis hirsuta agg. A. hirsuta, A. planisiliqua Arabis soyeri s.l. A. soyeri subsp. jacquinii, A. soyeri subsp. subcoriacea Brachypodium pinnatum s.l. B. pinnatum subsp. pinnatum, B. pinnatum subsp. rupestre Bryum capillare agg. B. capillare, B. torquescens Bryum pseudotriquetrum agg. B. bimum, B. pseudotriquetrum, B. subneodamense Bryum rubens agg. B. klinggraeffii, B. rubens Callitriche palustris agg. C. cophocarpa, C. hamulata, C. palustris, C. platycarpa Campylium calcareum agg. C. calcareum, C. sommerfeltii Campylium stellatum s.l. C. stellatum var. protensum, C. stellatum var. stellatum Cardamine pratensis s.l. C. pratensis subsp. dentata, C. pratensis subsp. matthioli, C. pratensis subsp. polemonioides, C. pratensis subsp. pratensis, C. pratensis subsp. rivularis Carex buxbaumii agg. C. buxbaumii, C. hartmanii Carex flava agg. C. demissa, C. flava, C. jemtlandica, C. lepidocarpa, C. serotina Carex muricata agg. C. divulsa, C. muricata, C. spicata Carex nigra incl. C. juncella Centaurea jacea agg. C. jacea, C. macroptilon, C. pannonica, C. subjacea Centaurium littorale s.l. C. littorale subsp. littorale, C. littorale subsp. uliginosum Cephalozia lunulifolia incl. C. affinis Cerastium arvense s.l. C. arvense subsp. arvense, C. arvense subsp. glandulosum, C. arvense subsp. molle, C. arvense subsp. strictum, C. arvense subsp. suffruticosum Cerastium fontanum s.l. C. fontanum subsp. fontanum, C. fontanum subsp. lucorum, C. fontanum subsp. vulgare Chiloscyphus polyanthos agg. C. pallescens, C. polyanthos Dactylorhiza maculata agg. D. fuchsii, D. maculata, D. saccifera Dactylorhiza majalis agg. D. baltica, D. majalis Drepanocladus aduncus agg. D. aduncus, D. polycarpus Drepanocladus revolvens agg. D. cossonii, D. revolvens

144

Name Included taxa Eleocharis palustris agg. E. austriaca, E. mamillata, E. palustris Empetrum nigrum s.l. E. nigrum subsp. hermaphroditum, E. nigrum subsp. nigrum Equisetum arvense incl. E. boreale Euphrasia rostkoviana agg. E. hirtella, E. rostkoviana Festuca rubra agg. F. nigrescens, F. rivularis, F. rubra, F. trichophylla Galium mollugo agg. G. album, G. mollugo Galium palustre agg. G. elongatum, G. palustre Gentianella germanica agg. G. austriaca, G. bulgarica, G. germanica, G. praecox Gymnadenia conopsea agg. G. conopsea (L.) R. Br., G. densiflora (Wahlenb.) A. Dietrich Hypnum cupressiforme agg. H. andoi, H. cupressiforme, H. jutlandicum, H. lacunosum Juncus bufonius agg. J. bufonius, J. hybridus, J. ranarius Juniperus communis s.l. J. communis subsp. alpina, J. communis subsp. communis Knautia arvensis agg. K. arvensis, K. kitaibelii Leucanthemum vulgare agg. L. adustum, L. cuneifolium, L. praecox, L. vulgare Lotus corniculatus agg. L. alpinus, L. corniculatus, L. tenuis Luzula arcuata agg. L. arcuata, L. confusa Luzula campestris agg. L. campestris, L. multiflora, L. pallescens, L. sudetica Mentha arvensis s.l. incl. Mentha x verticillata Molinia caerulea agg. M. arundinacea, M. caerulea Myosotis scorpioides agg. M. nemorosa, M. scorpioides Palustriella commutata s.l. P. commutata var. commutata, P. commutata var. falcata Pedicularis palustris s.l. P. palustris subsp. borealis, P. palustris subsp. opsiantha (incl. P. karoi), P. palustris subsp. palustris Philonotis fontana agg. P. fontana, P. tomentella Plagiomnium affine agg. P. affine, P. elatum, P. ellipticum, P. medium, P. rostratum Plagiothecium laetum agg. P. curvifolium, P. laetum Poa pratensis agg. P. angustifolia, P. pratensis, P. subcaerulea Polygala vulgaris agg. P. oxyptera, P. vulgaris Primula farinosa s.l. P. farinosa subsp. exigua, P. farinosa subsp. farinosa Racomitrium canescens agg. R. canescens, R. elongatum Ranunculus aquatilis s.l. R. aquatilis, R. peltatus Ranunculus auricomus agg. R. auricomus, R. fallax Ranunculus montanus agg. R. carinthiacus, R. montanus, R. oreophilus Rubus fruticosus agg. all Rubus species with the exception of R. arcticus, R. caesius, R. chamaemorus, R. idaeus and R. saxatilis. Rumex acetosa s.l. incl. R. fontanopaludosus Salix arbuscula agg. S. arbuscula, S. foetida, S. waldsteiniana Salix myrsinifolia agg. S. borealis, S. myrsinifolia Salix myrsinites agg. S. alpina, S. breviserrata, S. myrsinites Saxifraga umbrosa agg. S. hirsuta, S. spathularis, S. umbrosa Scirpus lacustris s.l. S. lacustris subsp. lacustris, S. lacustris subsp. tabernaemontani Senecio nemorensis agg. S. nemorensis subsp. fuchsii, S. nemorensis susbp. nemorensis, S. hercynicus Soldanella pindicola agg. S. chrysostricta, S. pindicola Sphagnum sp. Sphagnum spp. div. Sphagnum sect. Subsecunda S. contortum, S. denticulatum agg., S. platyphyllum, S. subsecundum Sphagnum not sect. Subsecunda all Sphagnum species with the exception of the Subsecunda section (for the purpose of formal definitions; Appendix S5) Sphagnum annulatum agg. S. annulatum, S. jensenii Sphagnum denticulatum agg. S. denticulatum, S. inundatum Sphagnum imbricatum s.l. S. imbricatum subsp. affine, S. imbricatum subsp. austinii Sphagnum palustre s.l. S. palustre var. centrale, S. palustre var. palustre Sphagnum recurvum agg. S. angustifolium, S. brevifolium, S. fallax, S. flexuosum Stellaria palustris agg. S. fennica, S. palustris Taraxacum sect. xy (sections Taraxacum spp. div. according to Flora Europaea) Utricularia intermedia agg. U. intermedia, U. ochroleuca Utricularia minor agg. U. bremii, U. minor Vaccinium oxycoccos agg. V. microcarpum, V. oxycoccos Valeriana dioica s.l. V. dioica subsp. simplicifolia, V. dioica subsp. dioica Valeriana officinalis s.l. V. officinalis subsp. collina, V. officinalis subsp. officinalis, V. officinalis subsp. sambucifolia Vicia cracca agg. V. cracca, V. tenuifolia

145

References Frey, W, Frahm J.-P., Fischer, E., Lobin, W. 2006. The liverworts, mosses and ferns of Europe. English edition. Harley Books, Colchester, UK. Tutin, T.G., Heywood, V.H., Burges, N.A., Moore, D.M., Valentine, D.H., Walters, S.M. & Webb, D.A. (eds) 1968–1993. Flora Europaea, Vols 2–5, 2 nd edn. of Vol. 1. Cambridge University Press, Cambridge, UK.

Appendix S3. Lists of fen specialists (“typical fen species”) and non-fen species. 1. List of fen specialists (“typical fen species”) whose presence or dominance in vegetation plots was considered in plot selection. Vegetation plots (i.e. fen relevés) were selected if (1) they contained at least 4 typical fen species or (2) the cover of at least one typical fen species reached 25% or 50% (see the values in brackets). Cover of 50% is used for bryophytes of dystrophic hollows (see below) and for Sphagnum recurvum agg. to avoid inclusion of bog vegetation. Aggregates or species complexes as well as subordinate taxa (see Appendix S2) were used during the selection procedure. Several fen species with a broader ecological niche occurring frequently in other vegetation types (such as Aulacomnium palustre or Carex rostrata) were not included.

Blysmus compressus >25% Cinclidium subrotudum >25% Pinguicula nevadensis >25% Calliergon giganteum >25% Dactylorhiza incarnata >25% Pinguicula vulgaris >25% Calliergon richardsonii >25% Drepanocladus badius >25% Primula deorum >25% Calliergon sarmentosum >25% Drepanocladus exannulatus >25% Primula farinosa s.l. >25% Calliergon stramineum >25% Drepanocladus lycopodioides >25% Primula stricta >25% Calliergon trifarium >25% Drepanocladus pseudostramineus >25% Pseudorchis frivaldii >25% Calliergon turgescens >25% Drepanocladus revolvens agg. >25% Rhynchospora alba >25% Calycocorsus stipitatus >25% Drepanocladus sendtneri >25% Rhynchospora fusca >25% Campylium polygamum >25% Drepanocladus trichophyllus >25% Saxifraga hirculus >25% Campylium stellatum s.l. >25% Drepanocladus tundrae >25% Scheuchzeria palustris >25% Carex atrofusca >25% Drosera anglica >25% Schoenus ferrugineus >25% Carex bicolor >25% Drosera intermedia >25% Schoenus nigricans >25% Carex buxbaumii agg. >25% Eleocharis quinqueflora >25% Scirpus cespitosus >25% Carex capitata >25% Epipactis palustris >25% Scirpus hudsonianus >25% Carex chordorrhiza >25% Equisetum variegatum >25% Scirpus pumilus >25% Carex curta >25% Eriophorum angustifolium >25% Scorpidium scorpiodes >25% Carex davalliana >25% Eriophorum gracile >25% Silene asterias >25% Carex diandra >25% Eriophorum latifolium >25% Sphagnum contortum >25% Carex dioica >25% Eriophorum scheuchzeri >25% Sphagnum denticulatum agg. >25% Carex echinata >25% Fissidens adianthoides >25% Sphagnum obtusum >25% Carex flava agg. >25% Hamatocaulis lapponicus >25% Sphagnum platyphyllum >25% Carex frigida >25% Hamatocaulis vernicosus >25% Sphagnum pulchrum >25% Carex heleonastes >25% Hammarbya paludosa >25% Sphagnum recurvum agg. (> 50%)* Carex hostiana >25% Juncus alpinus >25% Sphagnum subfulvum >25% Carex lasiocarpa >25% Juncus arcticus >25% Sphagnum subnitens >25% Carex laxa >25% Juncus biglumis >25% Sphagnum subsecundum >25% Carex limosa >25% Juncus castaneus >25% Sphagnum teres >25% Carex livida >25% Juncus triglumis >25% Sphagnum warnstorfii >25% Carex magellanica >25% Meesia triquetra >25% Spiranthes aestivalis >25% Carex maritima >25% Narthecium scardicum >25% Swertia perennis >25% Carex microglochin >25% Paludella squarrosa >25% Tofieldia calyculata >25% Carex norvegica >25% Parnassia palustris >25% Tofieldia pusilla >25% Carex panicea >25% Pedicularis mixta >25% Tomentypnum nitens >25% Carex parallela >25% Pedicularis palustris >25% Triglochin palustris >25% Carex pulicaris >25% Pedicularis sceptrum-carolinum >25% Valeriana dioica s.l. >25% Carex saxatilis >25% Pedicularis verticillata >25% Wahlenbergia hederacea >25% Carex trinervis >25% Pinguicula balcanica >25% Cinclidium stygium >25% Pinguicula grandiflora >25%

146

Bryophytes of dystrophic hollows: Drepanocladus fluitans >50% Sphagnum cuspidatum >50% Sphagnum lindbergii >50% Sphagnum majus >50% * The plot could not contain Eriophorum vaginatum > 25% at the same time.

2. List of non-fen species (i.e. negative-preferential fen species) used as a criterion for exclusion of vegetation plots from the dataset. A. Nutrient demanding herbs and graminoids of wet habitats (mostly of wet or intermittently wet Calthion palustris or Molinion caeruleae meadows) or species of disturbed wetlands. The criteria for the exclusion of plots were the following: (1) the presence of at least six negative-preferential fen species in the case of presence-absence data and (2) summed cover of meadow species of at least 25% for plots with recorded covers of individual species.

Aegopodium podagraria Festuca pratensis Phalaris arundinacea Alopecurus pratensis Festuca rubra agg. Phleum pratense Angelica sylvestris Filipendula ulmaria Poa pratensis agg. Bellis perennis Galium uliginosum Poa trivialis Berula erecta Geranium palustre Polygonum amphibium Calamagrostis epigejos Geum rivale Polygonum bistorta Caltha palustris Holcus lanatus Pulicaria dysenterica Carex disticha Hypericum tetrapterum Ranunculus acris Carex hirta Chaerophyllum hirsutum Ranunculus auricomus agg. Cerastium fontanum s.l. Juncus articulatus Ranunculus repens Cirsium canum Juncus conglomeratus Rumex acetosa Cirsium dissectum Juncus effusus Rumex conglomeratus Cirsium oleraceum Laserpitium prutenicum Rumex crispus Cirsium rivulare Lathyrus pratensis Sanguisorba officinalis Cynosurus cristatus Lolium perenne Scirpus sylvaticus Dactylis glomerata Lychnis flos-cuculi Stachys officinalis Deschampsia cespitosa Lysimachia nummularia Urtica dioica Epilobium hirsutum Mentha arvensis Veronica beccabunga Epilobium parviflorum Mentha longifolia Vicia cracca agg. Festuca arundinacea Myosotis scorpioides agg.

B. Woody species. Plots with a cover reaching 25% or more of these woody species were excluded to eliminate forests or scrub with fen species in the herb layer.

Alnus glutinosa Picea abies Pinus mugo Salix aurita Salix cinerea Salix cinerea

147

Appendix S4. List of species groups based on the COCKTAIL method. Species groups Species Anagallis tenella gr. Anagallis tenella, Carum verticillatum, Hypericum elodes, Juncus bulbosus, Scutellaria minor, Wahlenbergia hederacea Calliergon sarmentosum gr. Calliergon sarmentosum, Carex rariflora, Drepanocladus exannulatus Carex atrofusca gr. Carex atrofusca, Carex capillaris, Carex parallela, Catoscopium nigritum, Juncus triglumis, Salix reticulata Carex bicolor gr. Carex bicolor, Carex microglochin, Equisetum variegatum Carex davalliana gr. Carex davalliana, Eriophorum latifolium, Parnassia palustris, Primula farinosa, Tofieldia calyculata, Valeriana dioica s.l. Carex elata gr. Calamagrostis canescens, Carex elata, Lycopus europaeus, Peucedanum palustre Carex limosa gr. Carex limosa, Carex lasiocarpa, Carex chordorrhiza Carex trinervis gr. Calamagrostis epigejos, Carex trinervis, Juncus anceps, Juncus gerardi, Salix repens Drepanocladus revolvens gr. Drepanocladus revolvens agg., Carex flava agg., Campylium stellatum s.l., Pinguicula vulgaris Erica tetralix gr. Erica tetralix, Narthecium ossifragum, Odontoschisma sphagni, Sphagnum papillosum Eriophorum vaginatum gr. Carex pauciflora, Eriophorum vaginatum, Sphagnum fuscum, Sphagnum magellanicum, Sphagnum rubellum Mentha aquatica gr. Eupatorium cannabinum, Juncus subnodulosus, Lythrum salicaria, Mentha aquatica Menyanthes trifoliata gr. Carex rostrata, Equisetum fluviatile, Menyanthes trifoliata, Potentilla palustris Narthecium scardicum gr. Gentiana pyrenaica, Narthecium scardicum, Pinguicula balcanica, Plantago gentianoides, Primula deorum, Pseudorchis frivaldii Philonotis calcarea gr. Blysmus compressus, Eleocharis quinqueflora, Palustriella commutata s.l., Philonotis calcarea, Triglochin palustris Saxifraga hirculus gr. Drepanocladus aduncus agg., Hamatocaulis vernicosus, Saxifraga hirculus, Stellaria crassifolia, Triglochin maritima Saxifraga rivularis gr. Alopecurus alpinus, Ranunculus hyperboreus, Saxifraga foliolosa, Saxifraga rivularis Scorpidium scorpioides gr. Calliergon trifarium, Cinclidium stygium, Scorpidium scorpioides, Utricularia minor agg. Schoenus ferrugineus gr. Carex hostiana, Epipactis palustris, Sesleria caerulea, Schoenus ferrugineus Sphagnum cuspidatum gr. Drepanocladus fluitans, Scheuchzeria palustris, Sphagnum cuspidatum, Sphagnum majus Sphagnum recurvum gr. Calliergon stramineum, Polytrichum commune, Sphagnum recurvum agg., Vaccinium oxycoccos agg. Sphagnum subsecundum gr. Agrostis canina, Carex echinata, Sphagnum subsecundum, Viola palustris Sphagnum warnstorfii gr. Aulacomnium palustre, Carex dioica, Paludella squarrosa, Sphagnum contortum, Sphagnum teres, Sphagnum warnstorfii, Tomentypnum nitens Veronica scutellata gr. Galium palustre agg., Hydrocotyle vulgaris, Ranunculus flammula, Veronica scutellata

Appendix S5. Formal definitions of alliances. The definitions consist of species groups (sociological or functional species groups), selected species and logical (AND, OR, NOT) plus relational (GR) operators. Brackets are used to group pairs of elements within the definition. Numbers refer to percentage covers. For further information about the structure and creation of formal definitions see Chytrý et al. (2007) and Landucci et al. (2015). All the symbols and the structure of the formulas follow the protocol of the JUICE software (Tichý 2002; http://www.sci.muni.cz/botany/juice/).

Species groups ### = sociological group; plot is selected if at least half of the species of the group occur there #TC = functional species group (according to Landucci et al. 2015); the plot is selected if the total cover of the member species of the group exceeds the given percentage threshold; the merging of covers of individual species follows the protocol of the JUICE software, recently formally described by Fischer (2015)

Operators

GR = greater than (followed by threshold cover values expressed in percentages) AND = both elements must be present OR = at least one element must be present NOT = element(s) must not be present

Caricion viridulo-trinervis <### Carex trinervis>

148

Caricion davallianae ((<### Carex davalliana>OR<### Philonotis calcarea>)OR(<### Schoenus ferrugineus>OR(<#TC Philonotis calcarea GR50>OR)))NOT((OR)OR<### Carex trinervis>)

Caricion atrofusco-saxatilis (<### Carex atrofusca>OR<### Carex bicolor>)NOT(OR<### Carex limosa>)

Stygio-Caricion limosae ((<#TC Scorpidium scorpioides GR05>OR<### Scorpidium scorpioides>)AND(<### Carex limosa>OR))NOT(OR(<### Carex davalliana>OR(((<### Philonotis calcarea>OR)OR(<#TC Schoenus ferrugineus GR05>OR<#TC Philonotis calcarea GR05>))OR(<### Mentha aquatica>OR(<### Erica tetralix>OR<### Carex elata>)))))

Sphagno warnstorfii-Tomentypnion nitentis <### Sphagnum warnstorfii>OR((<### Drepanocladus revolvens>OR<### Carex davalliana>)AND(OR))

Saxifrago-Tomentypnion <### Saxifraga hirculus>

Narthecion scardici <### Narthecium scardicum>

Caricion stantis OR(AND(<### Saxifraga rivularis>OR<### Drepanocladus revolvens>))

Anagallido tenellae-Juncion bulbosi <### Anagallis tenella>NOT

Drepanocladion exannulati ((<### Calliergon sarmentosum>AND(OR(OR)))OR<#TC Calliergon sarmentosum GR50>)NOT(<### Sphagnum subsecundum>OR(<### Sphagnum warnstorfii>OR<### Veronica scutellata>))

Caricion fuscae (<### Sphagnum subsecundum>OR(<### Menyanthes trifoliata> AND<### Veronica scutellata>))NOT(<### Erica tetralix>OR(OR(<### Carex davalliana>OR(<### Drepanocladus revolvens>OR(<### Sphagnum warnstorfii>OR(<### Narthecium scardicum>OR(OR(<### Anagallis tenella>OR(OR(OR(<#TC Scorpidium scorpioides GR05>OR<#TC Eriophorum vaginatum GR25>)))))))))))

Sphagno-Caricion canescentis (((<### Sphagnum recurvum>OR<### Carex limosa>)OR<#TC Menyanthes trifoliata GR50>)AND)NOT<#TC Sphagnum cuspidatum GR25>

Scheuchzerion palustris (<### Sphagnum cuspidatum>OR<#TC Sphagnum cuspidatum GR75>)NOT((<### Sphagnum recurvum>OR<### Carex elata>)OR((<### Sphagnum subsecundum>OR<### Menyanthes trifoliata>)OR(OR)))

The above listed formal definitions are valid only within the data set of fen plots. To avoid misclassification of related vegetation types (e.g. bogs, wet meadows, heathlands, tundra or salt marsh vegetation), more complex but universally valid definitions each consisting of the current definition combined with the criteria used for the fen dataset selection (see section Data collection and filtering) should be applied. The criteria used for fen dataset selection can also be used as the definition for the Scheuchzerio palustris-Caricetea fuscae class. The definition of this class is as follows:

(((<#SC Typical fen species GR25 EXCEPT #SC Bryophytes of dystrophic hollows|Sphagnum recurvum agg. >)OR((<#SC Bryophytes of dystrophic hollows GR50>)OR(NOT)))OR(<#03 Typical fen species>))NOT(<#TC Non-fen species GR25>OR<#SC Woody species GR25>)

For the lists of Typical fen species, Bryophytes of dystrophic hollows, Non-fen species and Woody species see Appendix S3.

149

#SC = the cover of any member of the species group is greater than the given percentage threshold #03 = the plot is selected if more than three members of the species group occurs there

References Chytrý, M. (ed.) 2007. Vegetace České republiky, 1. Travinná a keříčková vegetace. Academia, Praha, CZ. Fischer, H.S. 2015. On the combination of species cover values from different vegetation layers. Applied Vegetation Science 18: 169–170. Landucci, F., Tichý, L., Šumberová, K. & Chytrý, M. 2015. Formalized classification of species-poor vegetation: a proposal of a consistent protocol for aquatic vegetation. Journal of Vegetation Science 26: 791–803. Tichý, L. 2002. JUICE, software for vegetation classification. Journal of Vegetation Science 13: 451–453.

Appendix S6. Full version of the synoptic table with core plots assigned to alliances on the basis of formal definitions. The frequency values are shown, the species are sorted according to fidelity. The background shading indicates diagnostic species of alliances in cases when phi > 0.3, the bold underlined numbers indicate diagnostic species when phi > 0.5. CvT = Caricion viridulo-trinervis, Cd = Caricion davallianae, CaS = Caricion atrofusco-saxatilis, SCl = Stygio-Caricion limosae, SwT = Sphagno warnstorfii-Tomentypnion nitentis, SaT = Saxifrago-Tomentypnion, Ns = Narthecion scardici, Cs = Caricion stantis, AJ = Anagallido tenellae-Juncion bulbosi, De = Drepanocladion exannulati, Cf = Caricion fuscae, SCc = Sphagno- Caricion canescentis, Sp = Scheuchzerion palustris.

Alliance CvT Cd CaS SCl SwT SaT Ns Cs AJ De Cf SCc Sp Group No. 1 2 3 4 5 6 7 8 9 10 11 12 13 Number of plots 138 2456 213 604 1059 101 37 17 97 449 1594 2001 542 Plot size range [m2] 1–100 1–100 1–16 1–100 1–100 1–100 2–80 1–25 1–100 1–100 1–100 1–100 1–100 Mean plot size [m2] 15 18.4 2.5 2.8 18.8 21.4 18.3 4.0 13.2 4.4 20.2 18.8 23.8 Caricion viridulo-trinervis Salix repens 99 4 . 1 5 4 . . 1 . 5 1 . Calamagrostis epigejos 77 1 . . 1 . . . . . 1 1 . Carex trinervis 64 ...... 1 . . Juncus anceps 60 ...... Mentha aquatica 87 10 . 1 2 . . . 14 . 10 1 . Potentilla anserina 60 3 . . 1 . . . 1 1 1 . . Hydrocotyle vulgaris 87 2 . 1 2 . . . 24 . 17 3 1 Schoenus nigricans 54 3 . 1 1 . . . 13 . 1 . . Juncus gerardi 35 1 . . 1 ...... Carex flacca 58 34 3 1 4 . . . 1 . 1 1 . Euphrasia stricta agg. 29 1 . . 1 1 . . . . 1 . . Juncus articulatus 72 32 1 1 22 . 19 . 15 1 18 1 . Campylium polygamum 2 8 1 . 1 1 1 . . . . 1 . 1 Ranunculus flammula 59 2 . 2 4 . . . 40 . 24 1 1 Agrostis stolonifera agg. 71 13 15 1 6 46 8 . 11 1 12 4 . Trifolium repens 33 4 1 . 2 . . . 2 . 4 1 . Glaux maritima 22 1 ...... Hippophae rhamnoides 22 1 ...... Calliergonella cuspidata 74 48 . 3 27 11 . . 13 1 34 1 2 Vaccinium macrocarpon 20 ...... 1 . Pellia endiviifolia 24 2 1 . 1 . . . 3 . 1 . . Carex arenaria 18 ...... Leontodon taraxacoides 20 1 ...... 2 . 1 . . Dactylorhiza incarnata 41 9 . 3 6 28 . . . 1 2 1 . Rubus caesius 13 ...... Samolus valerandi 13 ...... Liparis loeselii 26 3 . 2 2 14 . . . . 1 1 . Holcus lanatus 36 6 . 1 9 . . . 10 . 22 3 . Sonchus arvensis 12 1 ......

150

Alliance CvT Cd CaS SCl SwT SaT Ns Cs AJ De Cf SCc Sp Group No. 1 2 3 4 5 6 7 8 9 10 11 12 13 Phragmites australis 42 19 . 9 11 9 . . 8 1 11 8 1 Eleocharis uniglumis 18 3 . . 1 . . . 4 . 1 . . Linum catharticum 30 27 1 1 12 . . . . . 2 1 . Rhamnus catharticus 10 1 ...... Eleocharis palustris agg. 20 4 . . 1 . 5 . 3 . 2 1 . Caricion davallianae Carex davalliana . 61 8 1 18 . . . . . 2 1 . Carex hostiana . 34 1 1 4 . . . 2 . 2 . . Briza media 1 43 1 . 28 . . . 1 1 11 2 . Valeriana dioica s.l. 1 50 . 1 35 1 . . 3 1 25 2 . Schoenus ferrugineus . 21 1 2 2 . . . . . 1 . . Potentilla erecta 32 78 2 2 58 1 3 . 30 2 51 21 1 Carex panicea 29 78 7 21 53 . 3 . 45 1 44 6 1 Tofieldia calyculata . 26 8 . 7 . . . . 1 1 1 . Succisa pratensis . 41 . 1 21 . . . 20 1 17 2 . Sesleria caerulea . 12 . . 1 ...... Palustriella commutata s.l. . 27 19 1 1 . 5 . . 1 1 . . Leontodon hispidus . 22 1 . 11 . . . . 1 3 1 . Caricion atrofusco-saxatilis Carex capillaris . 2 69 1 3 ...... Juncus triglumis . 2 71 1 1 . 5 . . 1 1 . . Salix reticulata . 1 63 1 1 ...... Polygonum viviparum . 4 85 1 12 . . 6 . 6 1 1 . Carex atrofusca . 1 62 1 1 ...... Saxifraga aizoides . 3 62 1 1 . . . . 1 . . . Equisetum variegatum 5 9 70 1 7 . . . . 1 1 . . Thalictrum alpinum . 1 49 2 8 . . . . 1 1 . . Catoscopium nigritum . 1 40 1 1 ...... Tofieldia pusilla . 1 48 1 10 . . . . . 1 . . Meesia uliginosa . 1 34 1 1 . . . . 1 . . . Pedicularis oederi . 1 31 . 1 ...... Carex vaginata . 1 36 1 8 . . . . 1 1 1 . Salix myrsinites agg. . 1 37 1 9 . . . . 1 1 . . Carex parallela . 1 27 . 1 ...... Carex microglochin . 1 27 1 1 ...... Saussurea alpina . 1 35 1 10 ...... Salix arbuscula agg. . 1 25 1 1 . . . . 1 . . . Carex bicolor . 1 20 ...... Pinguicula alpina . 3 25 1 1 . . . . 1 1 . . Kobresia simpliciuscula . 1 20 ...... Oncophorus virens . 1 23 1 1 . . . . 1 1 . . Selaginella selaginoides . 9 46 3 18 . 19 . 1 1 3 1 . Salix glauca . . 26 . 3 . . . . 6 . 1 . Juncus castaneus . 1 19 . 1 . . . . 1 . . . Carex saxatilis . 1 23 1 1 . . . . 6 1 . . Distichium capillaceum . . 15 ...... 1 . . . Saxifraga oppositifolia . . 15 . 1 ...... Carex bigelowii . 1 19 . 1 . . . . 4 1 . 1 Carex norvegica . . 14 . 1 ...... Leiocolea bantriensis . 1 14 1 1 ...... Drepanocladus revolvens agg. 4 52 72 33 41 . 5 65 . 18 3 1 . Silene acaulis . . 12 ...... 1 . .

151

Alliance CvT Cd CaS SCl SwT SaT Ns Cs AJ De Cf SCc Sp Group No. 1 2 3 4 5 6 7 8 9 10 11 12 13 Pinguicula vulgaris . 34 37 3 17 . . . 1 1 3 1 1 Betula nana . 1 35 3 20 26 . . . 4 1 3 2 Brachythecium turgidum . . 10 ...... Stygio-Caricion limosae Scorpidium scorpioides 4 4 2 92 8 . . 6 5 11 1 1 1 Calliergon trifarium . 1 8 36 3 . . . . 1 1 . 1 Carex livida . 1 . 19 1 . . . . 1 1 1 1 Cinclidium stygium . 1 12 32 11 . . 6 . 8 1 1 . Utricularia minor agg. . 3 . 27 4 3 8 . 4 1 1 1 2 Sphagno warnstorfii-Tomentypnion nitentis Sphagnum warnstorfii . . . 1 85 2 11 . . 3 3 2 1 Tomentypnum nitens . 19 19 1 66 41 . 29 . 1 4 1 . Aulacomnium palustre 1 10 1 2 73 35 41 6 7 2 32 20 2 Sphagnum contortum . . . 6 29 . 3 . 5 2 3 1 . Paludella squarrosa . 1 5 3 34 6 . 6 . 13 1 1 1 Sphagnum teres . . . 2 36 2 8 . . 7 17 5 1 Angelica sylvestris . 10 . 1 22 . . . 1 . 9 1 . Saxifrago-Tomentypnion Hamatocaulis vernicosus . 4 1 6 10 86 . . . . 4 1 1 Brachythecium mildeanum 1 2 1 . 2 69 . . . . 1 1 . Stellaria crassifolia . . . 1 1 62 ...... Drepanocladus aduncus agg. 14 1 . 1 1 77 . . . . 2 1 1 Saxifraga hirculus . 1 . . 4 69 . 6 . . . 1 . Carex diandra . 3 . 12 10 85 . . . 1 11 3 1 Rumex aquaticus . . . . 1 53 . . . . 1 . . Cicuta virosa . 1 . 1 1 55 . . . 1 1 1 1 Triglochin maritima 1 2 . . 1 53 ...... Rumex acetosa s.l. 1 3 1 . 12 72 . . . . 13 3 . Epilobium palustre 4 4 1 6 17 82 . . 4 10 24 8 1 Eriophorum gracile . 1 . 3 2 38 . . 5 1 1 1 1 Lysimachia thyrsiflora . 1 . 3 4 42 . . . 2 8 6 1 Carex appropinquata . 2 . 1 7 32 . . . . 2 1 1 Cardamine pratensis s.l. 15 11 . 3 14 64 22 29 3 6 21 1 . Helodium blandowii . 1 1 1 7 24 . . . 1 1 1 . Galium palustre agg. 51 12 . 6 18 63 . . 6 1 39 8 1 Drepanocladus sendtneri 1 1 . . 1 17 . . . . 1 . . Marchantia polymorpha . 2 2 1 4 23 . . . 1 3 1 . Bryum pseudotriquetrum agg. 4 43 37 16 35 69 32 . 6 7 13 1 1 Poa palustris 9 4 1 . 5 28 . . . 1 4 1 1 Calamagrostis stricta . 1 9 1 3 30 . 6 . 10 2 1 1 Galium uliginosum 22 19 . 2 39 50 . . 2 1 25 4 1 Herminium monorchis 1 1 . . . 13 ...... Narthecion scardici Pinguicula balcanica . 1 . . 1 . 100 . . 1 3 . . Plantago gentianoides . 1 . . 1 . 68 . . . 1 . . Carex sempervirens . 1 1 . 1 . 65 . . 1 1 . . Pseudorchis frivaldii . 1 . . 1 . 62 . . 1 1 1 . Gentiana pyrenaica . 1 . . . . 51 . . . 1 . . Nardus stricta . 4 1 1 11 . 86 . 2 6 27 8 . Taraxacum sect. Alpina . 1 1 . 1 . 51 . . 1 1 1 . Sesleria comosa . 1 . . 1 . 41 . . 1 1 . . Scapania irrigua . 1 1 1 2 . 49 . . 9 2 1 .

152

Alliance CvT Cd CaS SCl SwT SaT Ns Cs AJ De Cf SCc Sp Group No. 1 2 3 4 5 6 7 8 9 10 11 12 13 Ligusticum mutellina . 1 1 1 1 . 43 . . 5 1 1 . Primula deorum ...... 32 . . . 1 . . Philonotis seriata . 1 1 . 1 . 41 . 2 5 2 1 . Sphagnum platyphyllum . . . 1 1 . 30 . . 2 2 . 1 Bruckenthalia spiculifolia . . . . 1 . 22 . . . 1 . . Luzula campestris agg. 4 6 3 . 19 . 51 . 2 2 24 4 . Saxifraga stellaris . 1 1 . 1 . 24 . . 4 2 1 . Narthecium scardicum ...... 16 ...... Soldanella pindicola agg. . 1 . . 1 . 16 . . . 1 . . Dactylorhiza cordigera . 1 . 1 1 . 19 . . . 2 1 . Deschampsia cespitosa . 13 26 1 5 . 46 . 1 12 12 4 . Cerastium cerastoides . 1 . . . . 16 . . 2 1 . . Juncus filiformis . 1 . . 1 . 27 . . 7 12 7 1 Pinus mugo . 1 . . 1 . 14 . . 1 1 1 . Sphagnum subsecundum . . . 4 10 . 35 . 10 3 28 2 1 Sphagnum compactum . . . 1 1 . 16 . . 2 2 1 2 Calliergon stramineum . 1 1 6 31 4 57 24 2 25 28 42 2 Caricion stantis Dupontia fisheri ...... 94 . 1 . . . Calliergon turgescens . . 1 . . . . 35 . 1 . . . Ranunculus hyperboreus . . . . 1 . . 29 . 3 . . . Calliergon richardsonii . . . 1 1 1 . 29 . 1 1 1 . Eriophorum scheuchzeri . 1 1 . 1 . . 35 . 13 1 . 1 Bryum cryophilum ...... 24 . 1 . . . Scapania hyperborea . . . . 1 . . 24 . 1 1 . . Aulacomnium turgidum . . 5 . . . . 24 . 1 . . . Cinclidium arcticum ...... 18 . 1 . . . Equisetum arvense 3 10 8 1 5 . . 35 . 5 3 1 . Sphagnum squarrosum . . . 1 3 1 . 24 1 1 6 3 1 Drepanocladus pseudostramineus ...... 12 . . . . . Hamatocaulis lapponicus ...... 12 . 1 . . . Dicranum angustum . . . . 1 . . 12 . . . . . Orthothecium chryseon . . 1 . . . . 12 . 1 . . . Anagallido tenellae-Juncion bulbosi Anagallis tenella 12 1 . 2 1 . . . 90 . 1 1 . Juncus bulbosus 1 1 . 1 5 . . . 89 1 7 1 4 Hypericum elodes . . . 1 . . . . 61 . 1 . 1 Carum verticillatum ...... 60 . 2 1 . Sphagnum denticulatum agg. . . . 1 2 . . . 64 1 14 1 3 Potamogeton polygonifolius 1 1 . 1 1 . . . 41 . 2 1 . Narthecium ossifragum . 1 . 1 1 . . . 37 . 1 1 1 Eleocharis multicaulis 1 1 . 1 1 . . . 36 . 1 1 1 Scutellaria minor ...... 32 . 1 1 . Wahlenbergia hederacea ...... 24 . 1 1 . Juncus acutiflorus . 2 . 1 1 . . . 31 . 7 1 . Erica tetralix 9 1 . 1 1 . . . 31 . 1 5 3 Scirpus fluitans . . . 1 . . . . 12 . 1 . . Juncus subnodulosus 2 4 . 1 1 . . . 20 . 5 2 . Drepanocladion exannulati Drepanocladus exannulatus . 1 2 7 7 . 43 29 . 89 16 2 2 Calliergon sarmentosum . 1 5 5 3 . 19 35 . 46 2 1 . Caricion fuscae

153

Alliance CvT Cd CaS SCl SwT SaT Ns Cs AJ De Cf SCc Sp Group No. 1 2 3 4 5 6 7 8 9 10 11 12 13 Viola palustris . 4 . 2 30 . . . 19 2 60 18 1 Agrostis canina 12 7 . 1 27 2 46 . 13 1 61 23 4 Sphagno-Caricion canescentis Sphagnum recurvum agg. . . . 1 28 . . . 2 1 22 100 10 Vaccinium oxycoccos agg. . 2 . 24 40 35 . . . 2 8 67 38 Polytrichum commune . 1 . . 2 . 16 12 3 5 14 34 1 Scheuchzerion palustris Sphagnum cuspidatum . . . . 1 . . . 8 1 4 4 64 Scheuchzeria palustris . 1 . 5 1 1 . . . 2 1 14 67 Sphagnum majus . . . . 1 . . . . 3 1 1 41 Drepanocladus fluitans 4 1 . 1 1 . . 18 2 1 3 5 42 Sphagnum balticum . . . . 1 . . . . . 1 1 12 Rhynchospora alba . 1 . 9 6 . . . 11 1 4 10 31 Diagnostic species for two or more alliances Epipactis palustris 63 34 . 2 17 48 . . . 1 2 1 . Eriophorum latifolium . 59 2 4 34 6 51 . 1 1 3 1 . Equisetum palustre 4 51 4 5 43 . . . 1 5 19 4 1 Primula farinosa s.l. . 42 14 1 8 . 68 . . 1 1 1 . Carex limosa . 2 . 90 13 69 . . 9 15 6 24 52 Carex chordorrhiza . 1 . 57 14 76 . . . 13 2 7 1 Carex lasiocarpa . 5 . 58 25 50 . . 3 6 13 13 7 Menyanthes trifoliata 1 18 1 77 38 83 . . 25 6 26 27 9 Carex dioica . 11 22 7 47 50 . . . 1 3 1 1 Carex nigra 38 39 15 12 47 2 95 . 9 23 65 29 2 Carex echinata . 12 . 1 31 1 70 . 66 5 67 15 1 Species reaching frequency 25% in at least one cluster Eriophorum angustifolium 12 34 29 41 51 46 11 6 33 64 60 46 22 Carex rostrata . 18 1 45 50 51 . . 3 32 35 52 22 Molinia caerulea agg. 5 57 8 10 34 . . . 57 2 33 24 9 Campylium stellatum s.l. 27 68 69 31 57 14 16 24 14 2 5 1 . Carex flava agg. 64 59 5 20 45 1 22 . 48 1 20 2 1 Parnassia palustris 55 59 27 4 39 29 22 . 12 2 13 1 . Drosera rotundifolia 1 4 . 13 42 12 . . 39 1 19 39 31 Potentilla palustris 12 2 . 21 25 36 . . 8 23 37 23 5 Equisetum fluviatile . 13 . 42 35 19 . . 2 6 26 13 2 Cirsium palustre 22 26 . 1 31 2 . . 10 . 29 6 . Festuca rubra agg. 19 21 20 1 29 7 24 . 9 1 25 8 1 Ranunculus acris 1 28 1 . 19 . . . . 1 17 2 . Anthoxanthum odoratum agg. 3 11 4 1 21 . 22 . 13 1 31 8 . Crepis paludosa . 24 . 1 27 . . . . 1 12 5 . Andromeda polifolia . 1 8 26 25 8 . . . 7 3 20 32 Betula pubescens 6 7 3 11 21 31 . . 1 1 8 19 3 Carex curta . 1 . 1 5 . 3 . 2 26 25 21 4 Prunella vulgaris 31 26 . . 11 . . . 10 . 9 1 . Eriophorum vaginatum . 1 6 1 7 5 14 . . 6 5 29 27 Fissidens adianthoides 12 28 6 1 15 . . . 3 1 3 1 . Aneura pinguis 4 13 27 24 23 4 8 . 7 2 5 1 1 Eleocharis quinqueflora 33 22 16 6 7 . 3 . 3 1 1 1 . Peucedanum palustre . 4 . 9 7 30 . . . 1 14 11 1 Lythrum salicaria 28 8 2 1 4 . 3 . 5 3 13 2 1 Utricularia intermedia agg. . 1 . 29 3 24 . . 9 1 2 1 1

154

Supporting information to the paper Peterka et al. Formalized classification of European fen vegetation at the alliance level. Applied Vegetaion Science

Appendix S8. Resuls of the ISOPAM analyses. The diagnostic species of particular cluster were calculated using the phi coefficient of association for equalized size of all groups. The species with fidelity to a particular cluster of phi > 0.3 were considered as diagnostic and the significance of fidelity was tested using Fisher’s exact test (P < 0.01). Species are sorted by decreasing fidelity within clusters.

Region plot size n Clusters and their (m2) interpretation

arctic Europe (northern 1 350 Caricion atrofusco- Caricion atrofusco-saxatilis Stygio-Caricion limosae Sphagno warnstorfii- Drepanocladion exannulati Drepanocladion mainland Norway, saxatilis/Caricion Tomentypnion nitentis exannulati/Caricion stantis Svalbard, Iceland), davallianae Campylium stellatum s.l. Juncus triglumis Scorpidium scorpioides Sphagnum teres Drepanocladus exannulatus Calliergon sarmentosum south-eastern Greenland Cinclidium stygium Triglochin palustris Menyanthes trifoliata Empetrum nigrum s.l. Carex curta Dupontia fisheri Drepanocladus revolvens Juncus alpinus Carex limosa Sphagnum warnstorfii Sphagnum lindbergii agg. Oncophorus virens Agrostis stolonifera agg. Carex rostrata Tomentypnum nitens Hamatocaulis lapponicus

Pinguicula vulgaris Sedum villosum Tritomania quiquedentata Scapania hyperborea

Carex microglochin Pohlia filum Plagiomnium affine agg. Aulacomnium turgidum Meesia uliginosa Eriophorum scheuchzeri Paludella squarrosa southern and central 1 541 Caricion davallianae Caricion atrofusco-saxatilis Stygio-Caricion limosae Sphagno warnstorfii- Drepanocladion exannulati Sphagno-Caricion Scheuchzerion palustris Scandinavia Tomentypnion nitentis canescentis

Potentilla erecta Carex vaginata Scorpidium scorpioides Drepanocladus revolvens Drepanocladus Sphagnum recurvum agg. Sphagnum cuspidatum agg. exannulatus Carex hostiana Thalictrum alpinum Carex limosa Scirpus hudsonianus Sphagnum teres Sphagnum papillosum Sphagnum tenellum Carex panicea Carex capillaris Carex livida Sphagnum warnstorfii Calliergon sarmentosum Vaccinium oxycoccos agg. Eriophorum vaginatum Prunella vulgaris Pedicularis oederi Cinclidium stygium Eriophorum scheuchzeri Carex pauciflora Sphagnum lindbergii Succisa pratensis Salix reticulata Pedicularis Carex curta Sphagnum pulchrum Andromeda polifolia sceptrumcarolinum Ctenidium molluscum Polygonum viviparum Carex dioica Rhynchospora alba Carex flacca Carex atrofusca

(15-)16 113 Caricion davallianae Caricion atrofusco-saxatilis Stygio-Caricion limosae Sphagno warnstorfii- ~ Caricion fuscae Drepanocladion exannulati Scheuchzerion palustris Tomentypnion nitentis Eleocharis quinqueflora Polygonum viviparum Scorpidium scorpioides Aulacomnium palustre Sphagnum subnitens Salix lapponum Scheuchzeria palustris Palustriella commutata s.l. Tritomaria polita Carex livida Sphagnum warnstorfii Sphagnum subsecundum Carex curta Gymnocolea inflata

Carex flava agg. Carex saxatilis Pedicularis palustris s.l. Hylocomium splendens Sphagnum compactum Eriophorum scheuchzeri Drepanocladus fluitans southern and central Scandinavia

Eriophorum latifolium Carex atrofusca Utricularia minor agg. Equisetum palustre Cephalozia bicuspidata Carex rariflora Drosera rotundifolia Triglochin palustris Thalictrum alpinum Carex limosa Tritomania quiquedentata Calliergon sarmentosum Sphagnum cuspidatum

Schoenus ferrugineus Oncophorus virens Calliergon trifarium Rhizomnium punctatum Potentilla palustris Sphagnum papillosum Carex panicea Juncus triglumis Paludella squarrosa north-eastern Europe 1 233 Stygio-Caricion limosae Sphagno warnstorfii- Saxifrago-Tomentypnion Saxifrago-Tomentypnion Sphagno-Caricion Scheuchzerion palustris (north-eastern part of Tomentypnion nitentis (cluster 1) (cluster 2) canescentis European Russia and Scorpidium scorpioides Potentilla erecta Tomentypnum nitens Hamatocaulis vernicosus Vaccinium oxycoccos agg. Sphagnum majus neighbouring part of Utricularia intermedia agg. Molinia caerulea agg. Saxifraga hirculus Carex diandra Sphagnum recurvum agg. Scheuchzeria palustris Finland) Equisetum fluviatile Carex flava agg. Betula humilis Rumex acetosa Andromeda polifolia Sphagnum annuatum agg.

Eriophorum gracile Sphagnum warnstorfii Galium uliginosum Cardamine pratensis Sphagnum magellanicum Sphagnum cuspidatum Eriophorum angustifolium Filipendula ulmaria Poa palustris Carex dioica Betula nana Sphagnum lindbergii

Selaginella selaginoides Carex rostrata Carex limosa Sphagnum papillosum Sphagnum balticum Campylium stellatum s.l. Galiumm palustre agg. Stellaria crassifolia Drosera rotundifolia

100 292 Caricion davallianae Sphagno warnstorfii- Sphagno warnstorfii- Caricion fuscae Sphagno-Caricion Scheuchzerion palustris Tomentypnion Tomentypnion canescentis nitentis/Saxifrago- nitentis/Saxifrago- Tomentypnion (cluster 1) Tomentypnion (cluster 2) Carex panicea Tomentypnum nitens Rumex acetosa Rhynchospora fusca Sphagnum recurvum agg. Scheuchzeria palustris Primula farinosa s.l. Stellaria palustris Poa palustris Drosera intermedia Ledum palustre Sphagnum majus Succisa pratensis Polemonium caeruleum Epilobium palustre Rhynchospora alba Carex lasiocarpa Sphagnum cuspidatum Carex flava agg. Salix cinerea Saxifraga hirculus Sphagnum papillosum Empetrum nigrum s.l. Vaccinium oxycoccos agg. Pinguicula vulgaris Caltha palustris Sphagnum warnstorfii Scirpus cespitosus Calluna vulgaris Andromeda polifolia Drepanocladus revolvens Geum rivale Carex appropinquata Drosera anglica Vaccinium oxycoccos agg. Rhynchospora alba agg. Campylium stellatum s.l. Aulacomnium palustre Polygonum bistorta Eriophorum angustifolium Chamaedaphne calyculata central Europe (Czech 1 320 Caricion davallianae Sphagno warnstorfii- Caricion fuscae Sphagno-Caricion Scheuchzerion Scheuchzerion and Slovak republic, Tomentypnion nitentis canescentis palustris (cluster 1) palustris (cluster 2) southern Poland, western Ukraine) Drepanocladus revolvens Tomentypnum nitens Agrostis canina Sphagnum recurvum agg. Drepanocladus fluitans Drepanocladus fluitans agg. Eleocharis quinqueflora Sphagnum warnstorfii Epilobium palustre Polytrichum commune Carex limosa Sphagnum tenellum Campylium stellatum s.l. Calliergonella cuspidata Galium uliginosum Pinus sylvestris Sphagnum majus Eriophorum vaginatum Carex flava agg. Climacium dendroides Brachythecium rivulare Scirpus cespitosus Palustriella commutata s.l. Valeriana dioica s.l. Potentilla palustris Gymnocolea inflata

Triglochin palustris Carex echinata Viola palustris Sphagnum cuspidatum Bryum pseudotriquetrum Potentilla erecta Festuca rubra agg. Vaccinium oxycoccos agg. agg. central Europe (Czech and Slovak republic, southern Poland, western Ukraine)

16 450 Caricion davallianae (cluster Caricion davallianae (cluster Sphagno warnstorfii- Caricion fuscae Sphagno-Caricion Sphagno-Caricion 1) 2) Tomentypnion nitentis canescentis (cluster 1) canescentis (cluster 2) Drepanocladus revolvens Juncus inlexus Sphagnum warnstorfii Ranunculus flammula Nardus stricta Vaccinium oxycoccos agg. agg. Eleocharis quinqueflora Palustriella commutata s.l. Tomentypnum nitens Rhytidiadelphus squarrosus Sphagnum recurvum agg. Sphagnum recurvum agg.

Pinguicula vulgaris Eupatorium cannabinum Aulacomnium palustre Galium uliginosum Sphagnum palustre s.l. Melampyrum pratense Primula farinosa s.l. Tussilago farfara Sphagnum contortum Sphagnum teres Agrostis canina Eriophorum vaginatum Triglochin palustris Carex flacca Sphagnum teres Viola palustris Polytrichum commune Polytrichum commune Parnassia palustris Cratoneuron filicinum Carex dioica Juncus bulbosus Drosera rotundifolia Carex davalliana Equisetum arvense Galium uliginosum Ranunculus auricomus agg.

50–100 109 Caricion davallianae Sphagno warnstorfii- Caricion fuscae (cluster 1) Caricion fuscae (cluster 2) Sphagno-Caricion Sphagno-Caricion Tomentypnion nitentis canescentis canescentis/Scheuchzerion palustris Carex davalliana Carex flava agg. Potentilla palustris Rhynchospora alba Sphagnum recurvum agg. Sphagnum cuspidatum Cirsium canum Calliergon stramineum Carex rostrata Phragmites australis Vaccinium myrtillus Lysimachia thyrsiflora Sanguisorba officinalis Dactylorhiza maculata agg. Carex curta Hydrocotyle vulgaris Trientalis europaea Polytrichum commune

Dactylorhiza majalis Campylium stellatum Galium palustre agg. Drosera rotundifolia Carex limosa Eriophorum vaginatum Leucanthemum vulgare agg. Carex pallescens Scutellaria galericulata Sphagnum denticulatum Calamagrostis villosa Carex lasiocarpa agg. Carex distans Sphagnum warnstorfii Carex diandra Sphagnum papillosum Picea abies Ranunculus acris Philonotis fonata agg. Sphagnum teres Pinus sylvestris Vaccinium oxycoccos agg.

the Alps (Switzerland, 10–20 429 Caricion davallianae (cluster Caricion davallianae (cluster Caricion davallianae (cluster Sphagno warnstorfii- Caricion fuscae (incl. Sphagno-Caricion Austria, northern Italy) 1) 2) 3) Tomentypnion nitentis Drepanocladion exannulati ) canescentis Agrostis stolonifera agg. Carex davalliana Drepanocladus revolvens Sphagnum warnstorfii Viola palustris Sphagnum compactum agg. Palustriella commutata s.l. Soldanella alpina Carex hostiana Carex rostrata Epilobium palustre Eriophorum vaginatum

Palustriella decipiens Plantago alpina Eriophorum latifolium Juncus filiformis Sphagnum recurvum agg. Salix myrsinites agg. Allium schoenoprasum Dactylorhiza incarnata Eriophorum scheuchzeri Sphagnum russowii Carex frigida Aster bellidiastrum Polygala amarella Carex curta Carex pauciflora Philonotis calcarea Selaginella selaginoides Utricularia minor agg. Carex nigra Odontoschisma denudatum

Juncus triglumis Carex flava agg. Carex elata Drepanocladus exannulatus Drosera rotundifolia southern and central 15-16 202 Caricion davallianae (cluster Caricion davallianae (cluster Sphagno warnstorfii- Narthecion scardici Caricion fuscae Sphagno-Caricion part of Balkan 1) 2) Tomentypnion nitentis canescentis Blysmus compressus Juncus inflexus Rumex acetosa Taraxacum sect. Alpina Sphagnum subsecundum Sphagnum recurvum agg. Penninsula (Bulgaria, Philonotis calcarea Campylium stellatum s.l. Holcus lanatus Primula farinosa s.l. Carex nigra Menyanthes trifoliata Greece, Kosovo, Eleocharis quinqueflora Molinia caerulea agg. Sphagnum contortum Scapania irrigua Eriophorum angustifolium Carex rostrata Macedonia, Montenegro, Serbia) Palustriella commutata s.l. Carex panicea Hamatocaulis vernicosus Sphagnum platyphyllum Carex curta

Cratoneuron filicinum Eleocharis uniglumis Myosotis scorpioides agg. Pinguicula balcanica Potentilla palustris Leontodon autumnalis Juncus articulatus Juncus effusus Nardus stricta Equisetum fluviatile Plantago media Carex flacca Carex pallescens Saxifraga stellaris Sphagnum palustre s.l. Appendix S9. Description of the unclassified group of plots. According to the first three DCA axes (first axis length 7.3 SDU, second axis 7.0 SDU, third axis 3.9 SDU) and several runs of unsupervised numerical analyses with different parameters (results not shown), omitted plots represent various vegetation types which can be mainly interpreted as follows: (1) species-poor communities of the Atlantic and sub-Atlantic regions characterized by Juncus bulbosus and Sphagnum denticulatum agg., (2) mixtures of fens and wet heathlands or oceanic bogs of the Atlantic region characterized by Calluna vulgaris, Erica tetralix, Juncus squarrosus and Narthecium ossifragum, (3) species-poor communities of acidic flushes, pools and ditches transitional between poor fens and dystrophic hollows characterized by Calliergon stramineum, Carex magellanica and Sphagnum riparium, (4) transitional stands towards tall-sedge and reed beds characterized by Calla palustris, Carex elata, C. lasiocarpa, Cicuta virosa, Cladium mariscus, Equisetum fluviatile and Potentilla palustris, (5) species-poor communities of the (sub-)arctic zone and mountain regions of central Europe, (6) transitions to wet meadows characterized by either Galium uliginosum, Holcus lanatus, Prunella vulgaris, Briza media, Trifolium pratense and T. repens or by Mentha aquatica, Eupatorium cannabinum and Lythrum salicaria, (7) transitions to pastures characterized by Leontodon autumnalis, Plantago lanceolata, Potentilla anserina and Trifolium repens, (8) transitions to high-mountain grassland characterized by Polygonum viviparum, Saussurea alpina and Thalictrum alpinum.

Appendix S10. Reasons for the abandonment of the Rhynchosporion albae and Caricion lasiocarpae alliances. Our analyses support the view that some alliances traditionally recognised on the basis of dominant species and hydrological parameters should be abandoned. The most controversial case is the Rhynchosporion albae alliance which is included in the Habitats Directive (habitat type 7150 Depressions on peat substrates of the Rhynchosporion) and the Caricion lasiocarpae alliance. The attempt to delimit these alliances formally was not successful. It produced small groups that overlapped with other alliances and mostly captured different habitats than the original material, and we therefore did not finally accept these alliances. The original diagnosis of Rhynchosporion albae comprises two vegetation plots from north-eastern Switzerland (Koch 1926) dominated by Rhynchospora alba and sphagna of the Subsecunda section (Sphagnum contortum, S. platyphyllum and S. subsecundum). It also includes some meadow species (Cirsium palustre, Festuca rubra, Prunella vulgaris, Succisa pratensis). The alliance was subsequently ambiguously interpreted as either (1) all stands dominated by Rhynchospora alba or Carex limosa (e.g. Dierssen 1982; Steiner 1992), (2) communities of extremely waterlogged oligotrophic acidic to slightly acidic sites, mostly initial stages of peat formation characterized by Rhynchospora alba, R. fusca, Drosera spp., Lycopodiella inundata and Sphagnum of the Subsecunda section without species of wet meadows (e.g. Rybníček et al. 1984), or (3) vegetation of dystrophic bog hollows (Schaminée et al. 1995; Coldea et al. 1997), contradicting the original description from moderately rich fens (Koch 1926). This discrepancy could lead to the different perception and delimitation of habitat type 7150 Depressions on peat substrates of the Rhynchosporion. Our attempt to define the alliance formally as the vegetation of waterlogged or early successional moderately rich fens failed due to a substantial overlap with poor fens of the Sphagno-Caricion canescentis alliance in terms of diagnostic species and DCA centroids. Moreover, this definition did not encompass plots from Koch´s original diagnosis which DCA positioned close to the centroid of Caricion fuscae (Fig. 2). Therefore the concept of Rhynchosporion albae has to be abandoned. The case of the Caricion lasiocarpae alliance is similar. Besides broad interpretation of this syntaxon including all fens dominated by Carex lasiocarpa (or other tall-sedges such as C. diandra or C. rostrata), the alliance was also interpreted as a wetter vicariant of Caricion davallianae, i.e. base-rich community on organic soils with a stable water table (cf. Wheeler 1980; Rybníček et al. 1984). The Caricetum lasiocarpae Koch 1926 association must be considered as the nomenclature type of the Caricion lasiocarpae alliance, since it is the only validly described association assigned to this alliance in its original diagnosis (Vanden Berghen in Lebrun et al. 1949). The Caricetum lasiocarpae association as described by Koch (1926), however, contains species such as Carex davalliana, C. hostiana, Dactylorhiza incarnata, Epipactis palustris, Eriophorum latifolium, Pinguicula vulgaris, Schoenus ferrugineus; therefore Caricion lasiocarpae cannot be separated unequivocally from Caricion davallianae in temperate Europe. There are extremely rich tall-sedge fens with Carex lasiocarpa and Carex chordorrhiza in the boreal-subarctic zone lacking the temperate Caricion davallianae species and sphagna. They mostly correspond to the definition of Stygio-Caricion limosae, in some cases also Saxifrago-Tomentypnion or small brown-moss patches (≤1 m2) within Sphagno warnstorfii-Tomentypnion nitentis.

159

References Coldea, G., Sanda, V., Popescu, A. & Ştefan, N. 1997. Les associations végétales de Roumanie. Tome 1. Les associations herbacées naturelles. Presses Universitaires, Cluj, RO. Dierssen, K. 1982. Die wichtigsten Pflanzengesellschaften der Moore NW-Europas. Conservatoire et Jardin botaniques, Genève, CH. Koch, W. 1926. Die Vegetationseinheiten der Linthebene unter Berücksichtigung der Verhältnisse in der Nordostschweiz. Systematisch-kritische Studie. Jahrbuch der St. Gallischen Naturwissenschaftlichen Gesellschaft 61/2: 1–144. Lebrun, J., Noirfalise, A., Heinemann, P. & Vanden Berghen, C. 1949. Les Associations végétales de Belgique. Bulletin de la Société Royale de Botanique de Belgique 82: 105–199. Rybníček, K., Balátová-Tuláčková, E. & Neuhäusl, R. 1984. Přehled rostlinných společenstev rašelinišť a mokřadních luk Československa. Studie ČSAV 1984/8: 1–124. Schaminée, J.H.J., Weeda, E.J. & Westhoff, V. (eds.) 1995. De vegetatie van Nederland 2. Plantengemeenschappen van wateren, moerassen en natte heiden. Opulus Press, Uppsala, SE. Steiner, G.M. 1992. Österreichischer Moorschutzkatalog. Ed. 4. Verlag Ulrich Moser, Graz/Wien, AT. Wheeler, B.D. 1980. Plant communities of rich-fen systems in England and Wales: I. Introduction. Tall sedge and reed communities. Journal of Ecology 68: 365–395.

160

Supporting information to the paper Peterka et al. Formalized classification of European fen vegetation at the alliance level. Applied Vegetation Science.

Appendix S11. Generalized classification scheme with distinguishing features of fen alliances. The scheme should be seen as a simple guide (dichotomous key) to identify particular alliances with respect to site conditions, biogeographical context and the presence of selected species or functional groups. It is important to mention that the key can serve as a supplementary tool for classification of fen communities only, but does not alllow classification of all stands (such as those corresponding to non-core or unclassified plots). E0 = moss layer

high-arctic mires others (circumpolar species, Balkan endemics others shallow peat) (high mountain ± moderately others dystrophic Caricion rich fens) stantis hollows Narthecion mineral soil organic soil scardici (tufa, gravel, wet rocks, sand)

dune others high pH, Ca others slacks ± without Sphagnum spp. ± non-quaking Caricion gravel, rock, tufa medium level of pH, Ca low level of pH, Ca viridulo- shallow peat; P-limited N-limited low nutrient availability low or higher nutrient availability trinervis arcto-alpine temperate boreo- no or only Ca-tolerant sphagna without calcicole species species continental (boreal species) (longer vegetation period, more (short vegetation period) oceanic regions, Fe toxicity) Caricion Caricion Saxifrago- atrofusco- davallianae Tomentypnion saxatilis

Sphagnum spp. dominate brown mosses dominate widespread species oceanic climate Scheuchzerion Ca-tolerant sphagna water table over/equal E0 atlantic species important palustris water table below E0 others lower level of pH, Ca Anagallido Fe toxicity tenellae- acidophilous sphagna dominate deep peat shallow peat Juncion lower species richness Sphagno topogenic soligenous bulbosi warnstorfii- Tomentypnion Stygio-Caricion Drepanocladion Caricion Sphagno- nitentis limosae exannulati fuscae Caricion canescetis Corresponding ecological types of fens according to Hájek et al. 2006, Perspectives in Plant Ecology, Evolution and Systematics 8: 97–114: geographically restricted types of rich calcareous fens + extremely rich fens rich fens moderately rich fens poor fens dystrophic hollows fens – moderately rich fens

Orders: Sphagno Caricetalia fuscae Caricetalia davallianae Br.-Bl. 1950 nom. conserv. propos. Sphagno warnstorfii-Tomentypnetalia Lapshina 2010 Caricetalia fuscae Koch 1926 Scheuchzerietalia palustris warnstorfii- Koch 1926 Nordhagen ex Tx. 1937 Tomentypnetalia Lapshina 2010

Paper 5 Peterka T., Hájek M., Dítě D., Hájková P., Palpurina S., Goia I., Grulich V., Kalníková V., Plesková Z., Šímová A. & Štechová T. (2018): Relict occurrences of boreal brown-moss quaking rich fens in the Carpathians and adjacent territories. – Folia Geobotanica 53: 265–276.

162

Relict occurrences of boreal brown-moss quaking rich fens in the Carpathians and adjacent territories

Tomáš Peterka1, Michal Hájek1, Daniel Dítě2, Petra Hájková1,3, Salza Palpurina1, Irina Goia4, Vít Grulich1, Veronika Kalníková1, Zuzana Plesková1, Anna Šímová1, Tápa Štechová5

1 Department of Botany and Zoology, Masaryk University, Kotlářská 2, CZ-61137 Brno, Czech Republic 2 Plant Science and Biodiversity Center, Slovak Academy of Sciences, Dúbravská cesta 9, SK-84523 Bratislava, Slovak Republic 3 Paleoecological Laboratory, Institute of Botany, The Czech Academy of Sciences, Lidická 25/27, CZ-60200 Brno, Czech Republic 4 Department of Taxonomy and Ecology, Babeş-Bolyai University, Republicii Street 42, RO-400015 Cluj Napoca, Romania 5 Department of Botany, University of South Bohemia, Branišovská 31, CZ-37005 České Budějovice, Czech Republic

Abstract Quaking rich fens dominated by boreal semi-aquatic brown-mosses such as Scorpidium scorpioides and Calliergon trifarium are extremely rare in the Carpathians. These fens harbour endangered species persisting in few localities in the region. However, their phytosociological classification has not been sufficiently solved yet, because they lack Sphagnum species as well as calcicole species characteristic for the Caricion davallianae alliance. A recent pan-European synthesis on fen vegetation suggests that these fens belong to the Stygio-Caricion limosae alliance, the boreal rich fen vegetation. The isolated occurrence of this alliance southward of the boreal zone and outside the Alps is rather exceptional and might represent a relict from an early post-glacial period. In this study, we compared phytosociological data for the Stygio-Caricion limosae alliance between northern Europe and the Carpathians plus adjacent regions (the Bohemian Massif, the Dinaric Alps) using NMDS and cluster analysis. We found that the species composition of brown-moss quaking rich fens in Central and South-Eastern Europe corresponds well with that in northern Europe, confirming their assignment to Stygio-Caricion limosae. We further reconstructed potential past distribution of the alliance in Czech Republic and Slovakia using available floristic and macrofossil data. The macrofossil data suggest that this vegetation type has been much more common in Central Europe and nowadays it persists only in ancient fens showing the long-term stability of environmental conditions. The main causes of its present-day rarity are Middle-Holocene woodland phases in fens and recent water table decreases caused by anthropogenic deterioration of water regime in the landscape.

Keywords: classification, macrofossil, mires, plant communities, vegetation, wetlands

Introduction

The Carpathian mountains are an important biogeographical interface in Europe where northern (boreal, arcto-alpine) and southern (temperate, Mediterranean) biogeographical elements meet as a consequence of specific distribution of glacial and postglacial refugia (Novák 1954; Mráz and Ronikier 2016; Jamrichová et al. 2017). The region hence harbours a wide array of vegetation types, which reflect both climatic and edaphic variability, including different types of fens (Dítě et al. 2017a). Nowadays, only few fens act as refugia of rare boreal and boreo-continental species such as Calliergon trifarium, Carex chordorrhiza, C.

163 limosa, Juncus triglumis, Meesia triquetra, Scheuchzeria palustris and Scorpidium scorpioides (Rybníček and Rybníčková 1965, 1972; Coldea 1990; Migra and Mičeta 1997; Janovicová 1998; Migra and Šoltés 1998; Dítě and Pukajová 2002a,b; Dítě and Kubandová 2005; Dítě and Šoltés 2010; Dítě et al. 2017b). Such fens occupy sites with high water level and moderately high concentration of dissolved carbonates. Due to these conditions, they can be characterized as “quaking rich fens”. However, phytosociological classification of these floristically-unique habitats remains unresolved. The vegetation of these fens is mostly composed of brown-mosses and, unlike other types of rich fens, frequently lacks Sphagnum species. Furthermore, calcium-demanding species typical of the extremely-rich fens of the Caricion davallianae alliance (e.g. Blysmus compressus, Carex davalliana, Philonotis calcarea, Primula farinosa) are generally absent. In the Carpathians, such vegetation type has been previously assigned either to the Caricion lasiocarpae alliance (Rybníček et al. 1984; Coldea 1990; Coldea et al. 1997; Hájek and Háberová 2001), which is however ambiguous name (Mucina et al. 2016), or to Caricion davallianae and Sphagno warnstorfii-Tomentypnion nitentis alliances (Dítě et al. 2007). In Scandinavia, similar vegetation has been classified as the Caricion lasiocarpae (Dierssen 1996) or Stygio-Caricion limosae alliances (Nordhagen 1943; Dahl 1956; Moen et al. 2012). A recent pan-European synthesis and classification of fen vegetation based on large set of vegetation-plot data and formalised classification approach (Peterka et al. 2017) has classified brown-moss quaking rich fens in the Carpathians (Slovakia, southern Poland, Romania) into the Stygio-Caricion limosae alliance as originally described by Nordhagen (1943). This alliance is widespread in the boreal zone of Europe, but it has never been distinguished in temperate Europe before that synthesis. Peterka et al. (2017) detected relatively frequent occurrence of the alliance in the Alps and few isolated occurrences in the Carpathians, the Dinaric Alps and, with certain compositional dissimilarity, in the Bohemian Massif (Czech Republic). All these isolated fens of the Stygio-Caricion limosae alliance harboured regionally rare species – boreal and arcto-boreal fen elements (Rybníček 1966; Rybníček and Rybníčková 1972; Birks and Walters 1973; Erzberger and Papp 2007; Štechová et al. 2010; Šoltés et al. 2015; Dítě et al. 2017a). In temperate Europe, these species are considered to be glacial or postglacial relicts, i.e. organisms that were common and widely distributed in glacial and postglacial times, but which retreated during interglacials to restricted areas with habitat or climatic conditions analogous to those of their original glacial habitats (Pearsson 1965; Frahm 2012; Dítě et al. 2017a). Since the vegetation of the Stygio- Caricion limosae alliance contains a high number of species considered as glacial relicts, we hypothesize its wider distribution during the Late Glacial and Early Holocene. This hypothesis could be tested by means of mapping macrofossil records, especially those of bryophytes, which are usually well preserved in the sediments and may be identified at the species level (Janssens 1983). A certain picture of the potential historical distribution of the alliance can be provided also using the floristic data. A focus on vegetation classification uncertainties may seem a purely academic exercise, but vegetation classification, which has a long tradition in Europe, has recently gain practical importance by underlying habitat typology that acts as a basic tool for nature conservation at both national and pan-European scales (De Cáceres et al. 2015; Chytrý et al. 2016). Thus, disharmony in the classification concepts and approaches may complicate effective habitat protection, especially if the target vegetation might represent a relict type that developed under presently non-existing climatic or edaphic conditions or both. In this study, we therefore aimed to (i) evaluate whether the vegetation of brown-moss

164 quaking rich fens in the Carpathians and adjacent areas differ from the vegetation of Stygio- Caricion limosae from northern Europe whence the alliance had been described, (ii) mapping the potential historical distribution of the alliance in Central Europe (Czech Republic, Slovakia) using both the macrofossil records and floristic data confronting recent and historical distributions and suggesting the reasons of decline of these exceptional fens in temperate Europe.

Material and Methods

What is the Stygio-Caricion limosae alliance? The Stygio-Caricion limosae alliance includes the vegetation of rich fens (sensu Sjörs 1952; Malmer 1986; Hájek et al. 2006) in topogenic, strongly waterlogged wetlands with peat accumulation. Stands are characterized by well-developed moss layer consisting mainly of “brown-mosses“, i.e. non-sphagnaceous weft-forming mosses (see also Udd et al. 2015). Most typical and often dominating bryophytes are Calliergon trifarium and Scorpidium scorpioides. They can be accompanied by other brown-mosses such as Calliergon giganteum, Campylium stellatum s.l., Cinclidium stygium, Drepanocladus exannulatus and D. revolvens agg. Sphagnum species occur rather sporadically. The herb layer is composed of sedges widespread in boreal and sub-arctic areas such as Carex chordorrhiza, C. diandra, C. lasiocarpa, C. limosa and C. livida, other species of the Cyperaceae family (Eriophorum angustifolium, Scirpus cespitosus, S. hudsonianus) and herbs of oligotrophic to mesotrophic aquatic and semi-aquatic habitats (Menyanthes trifoliata, Pedicularis palustris, Potentilla palustris). The occurrence of bladderworts (Utricularia spp.) is also typical for the alliance because of the high water level. The Stygio-Caricion limosae alliance was described by Nordhagen (1943) from the central part of southern Norway. The locus classicus is located in Sikilsdalen Valley, in the east of the Jotunheimen Mts. Nordhagen (1943) distinguished four associations within the alliance: Stygio-Caricetum chordorrhizae [Amblystegio scorpioidis-Caricetum chordorrhizae Osvald 1925], Stygio-Caricetum limosae [Amblystegio scorpioidis-Caricetum limosae Osvald 1923], Stygio-Caricetum lasiocarpae and Stygio-Eriophoretum polystachyi. The alliance has been recognized also by other Scandinavian phytosociologists (Dahl 1956, 1987; Sjörs et al. 1965; Singsaas 1989; Moen et al. 2012). The Stygio-Caricion limosae alliance occurs throughout the boreal and subarctic zones of Europe (Figure 1). The recent attempt to create fen classification on pan-European scale (Peterka et al. 2017) detected its scattered occurrence also in Iceland, Baltic states and north-eastern Poland, the Alps, the Jura Mts, Scotland and Ireland. Individual vegetation plots come also from the Massif Central Mts (central France), the Western Carpathians (Slovakia, southern Poland), the Eastern Carpathians (Romania) and the Durmitor Mts (the Dinaric Alps, Montenegro). Similar vegetation (further referred as non-core plots, see the Dataset section) had been rarely recorded also in the Bohemian Massif (Czech Republic) and along the Atlantic coast of north-western Europe.

Nomenclature The nomenclature was harmonized following Tutin et al. (1968–1993) for vascular plants and Frey et al. (2006) for bryophytes. Closely related taxa were merged (for details see Table S1 in Electronic supplementary material). Nomenclature of vegetation units

165 generally follows Mucina et al. (2016). In other cases, the author citation is given with the first reference.

Study area Our study is focused on the vegetation of the Carpathians and adjacent territories (the Dinaric Alps and the Bohemian Massif), with detailed focus on the former Czechoslovakia from where macrofossil and floristic records are available. For comparison between the study area (Figure 1) and the northern Europe, we used the dataset of vegetation-plot records (phytosociological relevés) compiled for the pan-European synthesis and classification of fen vegetation (Peterka et al. 2017; see the next section).

Fig. 1 Distribution of the Stygio-Caricion limosae alliance in the Carpathians and adjacent territories. Distribution in Europe adopted and adjusted from Peterka et al. (2017)

Dataset The dataset contains plots from the European Vegetation Archive (EVA; Chytrý et al. 2016), from private database of the authors of this study and several unpublished plots gathered by the late Kamil Rybníček. For this study we selected plots that may belong to the Stygio-Caricion limosae alliance as defined in Peterka et al. (2017). We selected not only the core plots (i.e. the relevés fully matching the formal definition; terminology adopted from Peterka et al. 2017), but also the non-core plots (i.e. relevés not matching the formal definition, but showing a high degree of compositional similarity to the core plots). For detailed description of the methodology see Peterka et al. (2017). Within our study area, core plots cover three localities in the Western Carpathians, (two near Bepadovo village and the Puchmajerovej jazierko Lake in Slovakia and one in Polana Biały Potok in Poland), one locality in the Eastern Carpathians (fen near Știol Lake, Rodna Mts, Romania) and one locality in the Dinaric Alps (Barno jezero Lake, Durmitor Mts, Montenegro; Table S2, Figures A–D). Within these areas, the non-core plots cover ≈30 localities (Table S3).

166

Vegetation data processing Vegetation plots from the Carpathians, the Durmitor Mts and the Bohemian Massif were subjected to ordination analysis together with plots (both core and non-core) from boreal and subarctic zone of Europe, namely from Norway (the loccus classicus), Sweden, Finland, Iceland and north-western part of the Russian Federation. We presupposed that comparison with the standard (type-like) material of Stygio-Caricion limosae from northern Europe can reveal potential dissimilarity of plots from the Carpathians and adjacent regions. In other words, if plots from central and south-eastern Europe would differ from those from northern Europe, their assignment to the Stygio-Caricion limosae alliance should be re- examined. To check the similarity among plots, we applied Non-Metric Multidimensional Scaling (NMDS) of Bray-Curtis distances based on square-root transformed percentage covers of individual species. Ties in the ordinal data were treated by a primary (i.e. indeterminate) approach. The NMDS solution was based on three principal axes. For visualisation, individual vegetation plots have been classified according to the geographical origin (Figure 2). The species covers had been transformed into seven-grade Braun- Blanquet´s scale before the analyses. To assess the similarity of plots from another point of view and the internal variability of Stygio-Caricion limosae, the non-hierarchical K-means cluster analysis was applied. The number of clusters was subjectively set to six, since several preliminary classifications with different parameters proved that more clusters would lack clear ecological or geographical interpretation. The species cover values were square-root transformed and the classification algorithm was repeated 20 times. The diagnostic species of a particular cluster (henceforth, group) were determined using the phi-coefficient (Chytrý et al. 2002) with all clusters standardized to an equal size. Species with fidelity to a particular group with a phi > 0.3 were regarded as diagnostic. The significance of fidelity was verified using Fisher’s exact tests (P < 0.001). The resulting groups from the K-means clustering have been further visualized in the NMDS ordination diagram (Table S4, Figure E). NMDS was performed using the CANOCO 5 package (Šmilauer and Lepš 2014). All other analyses were performed in the JUICE 7.0 software (Tichý 2002).

Palaeoecological and floristic data In order to reveal the potential historical distribution of Stygio-Caricion limosae we used macrofossil and floristic data and restricted the analysis only to the Czech and Slovak republics due to the lack of available data from the other regions. The macrofossil data were obtained from the Macrofossil Database of the Czech and Slovak republics (http://www.sci.muni.cz/botany/mirecol/paleo/). Most of them have been published in Rybníček and Rybníčková (1968), Jankovská (1970, 1980), Rybníčková and Rybníček (1972), Hájková et al. (2015a) and Gálová et al. (2016). From these sources, we extracted data on Calliergon trifarium, Carex chordorhiza, C. lasiocarpa, C. limosa, Cinclidium stygium and Scorpidium scorpioides, i.e. the diagnostic species of the Stygio-Caricion limosae alliance (cf. Peterka et al. 2017) with the exception of very common species Menyanthes trifoliata. Individual localities were considered as potential fossil record of the Stygio- Caricion limosae alliance if (i) any of the above-listed diagnostic moss species have strongly dominated the peat sample (i.e. the peat was almost completely built of remains of the species) or (ii) at least two diagnostic species have co-occurred in the sample. Note that each macrofossil record in our dataset comes from the peat sediment (i.e. fossil is deposited exactly at the place where the recorded species grew). In addition, macrofossil data captures

167 very small plots, mostly cores of ≤ 5cm diameter. Hence, the recorded species co-occured at very small spatial scale. The age of fossil samples is reported in calibrated years before present (i.e. before the year 1950) and it is based either directly on the C14 date of the layer in which the target species was found, or it is extrapolated from the depth-age model. Floristic data for vascular plants were adopted from Kaplan et al. (2016) for C. chordorrhiza in the Czech Republic, Řepka and Grulich (2014) for C. lasiocarpa and C. limosa in the Czech Republic, and Dítě and Pukajová (2002a,b) and D. Dítě (unpublished data) for the same sedges in Slovakia. Floristic data for bryophytes were taken from the review of all published records, herbarium records and own unpublished records (Hájková et al., in prep.) on which P.H., T.Š., D.D., T.P., Z.P. and M.H. participated. Widely and ambiguously localized records were not considered neither for vascular plants nor for bryophytes in any country. Individual localities were assigned as potential historical record of the Stygio-Caricion limosae alliance if at least one diagnostic, and usually dominant, moss species (either Calliergon trifarium or Scorpidium scorpioides) co-occured with at least one diagnostic species of sedge (Carex chordorrhiza, C. lasiocarpa, C. limosa).

Results

Comparison of species composition of brown-moss quaking rich fens between regions The position of plots in the NMDS scatter (Figure 2) shows that species composition of core plots of Stygio-Caricion limosae from the Western Carpathians (Slovakia, southern Poland) is similar to the species composition of core plots from the northern Europe. Regarding the locality at Barno jezero Lake (Durmitor Mts), one core plot gathered in the locality appeared to be similar to the plots from northern Europe, whereas another plot displayed slightly marginal position within the cluster of core plots together with the core plots from the Eastern Carpathians (Romania). The non-core plots from the entire Carpathians, the Bohemian Massif and the Durmitor Mts overlap generally with the non-core plots from northern Europe in the NMDS scatterplot. Hence, NMDS analysis revealed that plots identified as vegetation of Stygio-Caricion limosae sampled in the Carpathians and adjacent territories generally do not differ in species composition from those sampled in northern Europe. Analogous results were obtained by the K-means clustering (Table S4), in which the plots from central and south-eastern Europe were grouped together with those from northern Europe. The resulting groups of vegetation plots corresponded to vegetation types, roughly at the level of phytosociological associations, rather than to geographical regions. For syntaxonomical synopsis of the Stygio-Caricion limosae alliance in the Carpathians and adjacent territories see Table S5.

Potential historical distribution of the boreal brown-moss quaking rich fens in Czech and Slovak Republics Macrofossil records suggest that brown-moss quaking rich fens of the Stygio-Caricion limosae alliance might have been widely distributed in central Europe throughout the glacial and the Early-Holocene, since available palaeoecological data come mostly from the sites beyond the present distribution range of the alliance as identified by the vegetation data (Borská nížina Lowland, Upper-Hron Basin, Malé Karpaty Mts, Křivoklátsko region, Novohradské hory Mts, Ostrava Basin, Lower- and Upper-Morava river Valleys; Figure 3).

168

Fig. 2 Results of NMDS analysis. Plots are classified according to geographical origin. Eigenvalues: 0.4711 (Axis 1), 0.2739 (Axis 2), 0.2550 (Axis 3)

Potential historical localities of brown-moss quaking rich fens as indicated by floristic data largely overlap with localities of vegetation plots of the Stygio-Caricion limosae alliance (see Figure 3 and Table S6). Only few potential historical sites where the target vegetation has not been recorded are somewhat isolated geographically (e.g. Vidnavské mokřiny in Opava region and several sites in Labe river floodplain). In most cases, they have not been investigated palaeoecologically. Other potential historical sites detected exclusively on the

169 basis of floristic data are located in the Třebop Basin and the Žďárské vrchy Hills, though nearby the phytosociologically approved localities.

Fig. 3 Potential historical distribution of brown-moss rich quaking fens in Central Europe (Czech and Slovak Republics) based on macrofossil and floristic records

Discussion

Phytosociological classification Our study revealed that the Carpathian brown-moss quaking rich fens with boreal species do not differ from the vegetation of the Stygio-Caricion limosae alliance in northern Europe. Hence, the name Stygio-Caricion limosae should be newly used even for the Carpathian communities. On the one hand, previous vegetation surveys on the territory of the Carpathians (Rybníček et al. 1984; Coldea 1990; Coldea et al. 1997; Hájek and Háberová 2001; Dítě et al. 2007) did not recognize the Stygio-Caricion limosae alliance. On the other hand, at the association level, Czech and Slovak phytosociologists distinguished Amblystegio scorpioidis-Caricetum chordorrhizae and Amblystegio scorpioidis-Caricetum limosae, which were recognized by Nordhagen (1943) within the description of the Stygio-Caricion limosae alliance. Therefore, a compositional similarity of several Carpathian brown-moss quaking rich fens to the analogous boreal vegetation had been previously recognised, though remained obscured by applying the widely defined, and ambiguously interpreted, phytosociological unit – the Caricion lasiocarpae alliance (Rybníček et al. 1984; Hájek and Háberová 2001; for details see Table S4). The ambiguous status of Caricion lasiocarpae stems partly from broad interpretation of this phytosociological unit and partly from its various interpretations in different vegetation surveys (see Mucina et al. 2016 for overview). Fen vegetation from the Rodna Mts in the Eastern Carpathians was described as a separate association Swertio perennis-Caricetum chordorrhizae and assigned, again, to the Caricion lasiocarpae alliance (Coldea 1990; Coldea et al. 1997). Classifications of brown-moss quaking

170 rich fens in the Carpathians, were obviously complicated due to the low number of existing localities and, consequently, due to the low amount of phytosociological data. We believe that direct comparisons of vegetation plots with material from northern Europe brought a sufficient evidence for their classification within the Stygio-Caricion limosae alliance.

Historical distribution and reasons for the decline of brown-moss quaking rich fens in Central Europe Although being increasingly endangered across Europe (Janssen et al. 2016), brown- moss quaking rich fens are still quite common in the boreal and subarctic zone of Europe (e.g. Moen et al. 2012; Joosten et al. 2017), whereas they occur very sporadically in the temperate Europe. However, these fens might have been more common here in the past and their distribution might have resembled the present-day distribution in northern Europe. Evidence from macrofossil and partially the floristic data presented in this study support the hypothesis for their wider potential historical occurrence. We are aware that both macrofossil and floristic data may overestimate historical distribution of brown-moss quaking rich fens, because they indicate only the presences but not the absences of indicator species. The floristic data further do not provide evidence that the recorded species indeed co-occurred within the local plant community. Using floristic data, we can conclude that the brown-moss quaking rich fens might have been present in the locality, but we cannot certainly determine the time period of community co-occurrence. On the first look, the number of the potential Late Glacial and Early Holocene localities identified by palaeoecological and floristic data is equal to the number of existing recent localities identified by phytosociological data. On the second look, but considering (i) that the area explored by phytosociologists is intrinsically much larger than the area investigated by palaeoecological cores and (ii) that palaeoecologically-proved sites are mostly located in lowlands where fens have been much destroyed as compared to highlands and mountains, the difference in number of historical and recent sites illustrates clearly a decline of brown- moss rich quaking fens in temperate Europe during the Holocene. Diagnostic species of Stygio-Caricion limosae have been found in the peat profiles in the Western Carpathians even at localities where they have not been recorded recently (Magyari et al. 1999; Hájková et al. 2015a; Hájková et al. 2012, 2017; Gálová et al. 2016). The layers composed of Calliergon trifarium and Scorpidium scorpioides have been found, for example, in the Liptovská kotlina (recently calcareous fens of the Caricion davallianae alliance; Hájková et al. 2015a), in the Upper-Hron Basin (recently again Caricion davallianae) and in the Malé Karpaty Mts (recently Sphagnum-dominated birch carr; Gálová et al. 2016). The same layers have been found in the neighbouring lowland river floodplains such as Borská nížina Lowland (Hájková et al. 2015b), Lower-Morava river Valley (Rybníčková and Rybníček 1972), Upper-Morava river Valley (Hájková et al. 2017) and Ostrava Basin at the Czech-Polish border (Fejfar et al. 1955). Several localities were detected also in the Bohemian Massif (for details see Table S7). Most macrofossils data come from the period 14000–9000 calibrated years BP, i.e. before the middle-Holocene expansion of closed forest (Pokorný et al. 2015; Hájek et al. 2016). Location of macrofossil records, indicating the potential occurrence of the Stygio-Caricion limosae alliance, suggests that this vegetation type had been occurred in the lowlands and mountain basins. These landscapes seem to be geomorphologically suitable for development of topogenic quaking fens, especially when climate moistened and permafrost melted at the Pleistocene-Holocene transition. The brown-moss quaking rich fens might hence have been widely distributed in the central-

171

European landscape in that period, representing an initial successional stage of peatland development. In Middle Holocene, the climate changes and consequent woodland expansion caused reduction of these fens (Hájková et al. 2015a). As mentioned above, several palaeoecological studies proved also the replacement by calcareous fens of the Caricion davallianae alliance or Sphagnum-dominated fens. This process might be driven not only by raising the fen surface above the water level (Hájková et al. 2015a), but partially also by other factors. One of them could be general increase of phosphorus limitation during the Holocene prior to the Industrial Era (Kuneš et al. 2011), that gradually favoured the Caricion davallianae species that are adapted to phosphorus-limited wetlands (Pawlikowski et al. 2013). Another potential cause of decline is decreasing continentality in the course of the Holocene, because higher summer precipitation-to-evaporation ratio promotes spread of Sphagnum species over brown-moss fens (Vicherová et al. 2017). We believe that the persistence of the Stygio-Caricion limosae vegetation at scattered sites during the Holocene was conditioned by locally stable ecological conditions such as high and stable water level which blocks the succession towards Sphagnum and/or wooded fens. On such sites, brown- moss quaking rich fens could have survived up to the second half of 20th century. Thereafter they probably disappeared at many localities due to modern human activities, i.e. drainage for agricultural purposes, eutrophication or abandonment (Růžička 1987; Navrátilová et al. 2017). Both the middle-Holocene forest optimum and fen deterioration in modern agricultural landscapes contributed to decline of all fen types. Brown-moss rich fens however seem to be more sensitive to successional changes as compared to other fen habitats (Hájek et al. 2015; Janssen et al. 2016; Navrátilová et al. 2017). Even small water level decline or slight nutrient addition may cause transformation of a brown-moss rich fen to a depauperated fen vegetation. Hájek et al. (2015) have recently demonstrated changes in the moss layer in Czech fens manifested by the replacement of brown-mosses of pristine fens by Sphagnum species or competitively strong brown-mosses such as Calliergonella cuspidata. Analogous changes have been documented in the Netherlands (Kooijman et al. 1994; Paulissen et al. 2014) and even in Fennoscandia (Juutinen 2011; Rehell and Virtanen 2016). Kooijman (2012) partially explained these successional changes by macronutrient input into fen ecosystems that result in acidification. Vicherová et al. (2015) experimentally demonstrated the triggering effect of increased potassium concentration, which alleviates the detrimental effect of calcium on calcifuge sphagna and thus facilitates their expansion. The effect of increasing macronutrient input acts synergistically with water table decline that can occur with ongoing succession, but also as a consequence of the hydrological deterioration of entire landscapes. Udd et al. (2016) demonstrated the higher competition capability of Sphagnum species in brown-moss fens under drier conditions. Mälson and Rydin (2007) further showed that the combined effect of surface desiccation and the increased cover of vascular plants were a possible cause of the reduction of Scorpidium scorpioides populations. The effects of higher macronutrient availability (caused by atmospheric deposition, local eutrophication or abandonment of formerly mown fens), higher productivity of herb layer and increasing water uptake by vascular plants plus peat desiccation cannot be therefore easily separated from each other. The important point is that, even though seemingly negligible, few centimetres decline of calcium-enriched water table may have crucial consequences for community structure of brown-moss quaking rich fens because it may allow Sphagnum species to outcompete brown mosses (Granath et al. 2010; Vicherová et al. 2015). The persistence of brown-moss quaking rich fens up to present times in the Central-

172

European landscape is therefore a rare phenomenon which is apparently determined by a complex interplay of (i) high and stable water level, (ii) limited overall macronutrient input and (iii) Holocene continuity. The latter was well documented in Puchmajerovej jazierko Lake and Poľana Biały Potok localities (Table S2).

Conclusions Species composition of brown-moss quaking rich fens of the Carpathians and surrounding territories corresponds to the communities of the Stygio-Caricion limosae alliance in boreal Europe. In Central and South-eastern Europe brown-moss quaking rich fens represent a relict phenomenon from the Late Glacial and Early Glacial periods whose disappearance accelerated because of the wide-scale eutrophication and disruption of water regime of the central-European landscape that took place in the second half of 20th century. Besides other general threats (conversion to hay meadows, forest establishment, fertilizer application in surrounding landscape), they can be easily replaced by Sphagnum-dominated fens due to autogenic succession. The persisting brown-moss quaking rich fens should be therefore treated as the sites of priority conservation interests in Central and South-Eastern Europe.

Acknowledgements The research was funded by the Czech Science Foundation (Centre of Excellence Pladias; 14-36079G) and Masaryk University (MUNI/A/1301/2016). The research of PH was partially supported by Czech Academy of Sciences (RVO 67985939). We are grateful to Kamil Rybníček (†) for providing unpublished vegetation plots. Our unpublished data from Romania were collected with the help of Veronika Horsáková and Michal Horsák. Many thanks go to Eva Hettenbergerová for management of the Macrofossil Database of the Czech republic and Slovakia and for general support during manuscript preparation. The co-ordinating editor and two anonymous reviewers provided important comments to the first draft of the paper.

References Birks HJB, Walters SM (1973) The flora and vegetation of Barno jezero, Durmitor, Montenegro. Glas Republ Zavoda Zašt Prir Prirodnjačkog Muz Titograd 5: 5–23 Chytrý M, Hennekens SM, Jiménez-Alfaro B, Knollová I, Dengler J, Jansen F, Landucci F, Schaminée JHJ, Adid S, Agrillo E, Ambarlı D, Angelini P, Apostolova I, Attorre F, Berg C, Bergmeier E, Biurrun I, Botta-Dukát Z, Brisse H, Campos JA, Carlón L, Čarni A, Casella L, Csiky J, Dušterevska R, Dajid Stevanovid Z, Danihelka J, De Bie E, de Ruffray P, De Sanctis M, Dickoré WB, Dimopoulos P, Dubyna D, Dziuba T, Ejrnæs R, Ermakov N, Ewald J, Fanelli G, Fernández-González F, FitzPatrick Ú, Font X, García-Mijangos I, Gavilán RG, Golub V, Guarino R, Haveman R, Indreica A, Işık Gürsoy D, Jandt U, Janssen JAM, Jiroušek M, Kącki Z, Kavgacı A, Kleikamp M, Kolomiychuk V, Krstivojevid Duk M, Krstonošid D, Kuzemko A, Lenoir J, Lysenko T, Marcenò C, Martynenko V, Michalcová D, Moeslund JE, Onyshchenko V, Pedashenko H, Pérez-Haase A, Peterka T, Prokhorov V, Rašomavičius V, Rodríguez-Rojo MP, Rodwell JS, Rogova T, Ruprecht E, Rūsira S, Seidler G, Šibík J, Šilc U, Škvorc Ž, Sopotlieva D, Stančid Z, Svenning J-C, Swacha G, Tsiripidis I, Turtureanu PD, Uğurlu E, Uogintas D, Valachovič M, Vashenyak Y, Vassilev K, Venanzoni R, Virtanen R, Weekes L, Willner W, Wohlgemuth T, Yamalov S (2016) European Vegetation Archive (EVA): an integrated database of European vegetation plots. Appl Veg Sci 19: 173–180 Chytrý M, Tichý L, Holt J, Botta-Dukát Z (2002) Determination of diagnostic species with statistical fidelity measures. J Veg Sci 13: 79–90 Coldea G (1990) Munții Rodnei. Studiu Geobotanic. Acad. Române, Bucureşti Coldea G, Sanda V, Popescu A, Ştefan N (1997) Les associations végétales de Roumanie. Tome 1. Les associations herbacées naturelles. Presses Universitaires, Cluj Dahl E (1956) Rondane. Mountain vegetation in south Norway and its relation to the environment. Skr Norske Vidensk-Akad Oslo, Mat-Naturvidensk Kl 3: 1–374 Dahl E (1987) Alpine-subalpine plant communities of South Scandinavia. Phytocoenologia 15: 455–484 De Cáceres M, Chytrý M, Agrillo E, Attorre F, Botta-Dukát Z, Capelo J, Czúcz B, Dengler J, Ewald J, Faber- Langendoen D, Feoli E, Franklin SB, Gavilán R, Gillet F, Jansen F, Jiménez-Alfaro B, Krestov P, Landucci F,

173

Lengyel A, Loidi J, Mucina L, Peet RK, Roberts DW, Roleček J, Schaminée JHJ, Schmidtlein S, Theurillat J-P, Tichý L, Walker DA, Wildi O, Willner W, Wiser SK (2015) A comparative framework for broad-scale plot- based vegetation classification. Appl Veg Sci 18: 543–560 Dierssen K (1996) Vegetation Nordeuropas. Verlag Eugen Ulmer, Stuttgart Dítě D, Hájek M, Hájková P (2007) Formal definitions of Slovakian mire plant associations and their application in regional research. Biologia 62: 400–408 Dítě D, Hájek M, Šoltés R, Košuthová A, Šibíková I, Kliment J (2017a) Glacial-relict symptoms in the Western Carpathian flora. Folia Geobot (this issue) Dítě D, Kubandová M (2005) Blatnica močiarna (Scheuchzeria palustris L.) na Kubínskej holi (severné Slovensko). Bull Slov Bot Spoločn 27: 45–48 Dítě D, Peterka T, Dítětová Z, Hájková P, Hájek M (2017b) Arcto-alpine species at their niche margin: The Western Carpathian refugia of Juncus castaneus and J triglumis in Slovakia. Ann Bot Fennici 54: 67–82 Dítě D, Pukajová D (2002a) Carex limosa L., kriticky ohrozený druh flóry Slovenska. Bull Slov Bot Spoločn 24: 65– 73 Dítě D, Pukajová D (2002b) Chorologické, ekologické a fytocenologické poznámky k výskytu ostrice výbežkatej (Carex chordorrhiza Ehrh.) na Slovensku. Bull Slov Bot Spoločn 24: 61–64 Dítě D, Šoltés R (2010) Rozšírenie druhu Scorpidium scorpioides na Slovensku: minulosť a súčasnosť. Bryonora 46: 66–69 Erzberger P, Papp B (2007) New and noteworthy bryophyte records from Montenegro and Serbia. Willdenowia 37: 339–351 Fejfar O, Kneblová V, Dohnal Z, Ložek V (1955) Paleontologický a cenologický výzkum kvartérních sedimentů Ostravska v roce 1953. Anthropozoikum 4: 241–284 Frahm JP (2012) The phytogeography of European bryophytes. Botanica Serbica 36: 23–36 Frey W, Frahm JP, Fischer E, Lobin W (2006) The liverworts, mosses and ferns of Europe. Harley Books, Colchester Gálová A, Hájková P, Čierniková M, Petr L, Hájek M, Novák J, Rohovec J, Jamrichová E (2016) Origin of a boreal birch bog woodland and landscape development on a warm low mountain summit at the Carpathian- Pannonian interface. The Holocene 26: 1112–1125 Granath G, Strengbom J, Rydin H (2010) Rapid ecosystem shifts in peatlands: linking plant physiology and succession. Ecology 91: 3047–3056 Hájek M, Dudová L, Hájková P, Roleček J, Moutelíková J, Jamrichová E, Horsák M (2016) Contrasting Holocene environmental histories may explain patterns of species richness and rarity in a Central European landscape. Quat Sci Rev 133: 48–61 Hájek M, Háberová I (2001) Scheuchzerio-Caricetea fuscae R.Tx. 1937. In Valachovič M (ed) Rastlinné spoločenstvá Slovenska. 3 Vegetácia mokradí. Veda, Bratislava, pp 185–274 Hájek M, Horsák M, Hájková P, Dítě D (2006) Habitat diversity of central European fens in relation to environmental gradients and an effort to standardise fen terminology in ecological studies. Persp Pl Ecol, Evol Syst 8: 97–114 Hájek M, Jiroušek M, Navrátilová J, Horodyská E, Peterka T, Plesková Z, Navrátil J, Hájková P, Hájek T (2015) Changes in the moss layer in Czech fens indicate early succession triggered by nutrient enrichment. Preslia 87: 279–301 Hájková P, Horsák M, Hájek M, Jankovská V, Jamrichová E, Moutelíková J (2015a) Using multi-proxy palaeoecology to test a relict status of refugial populations of calcareous-fen species in the Western Carpathians. The Holocene 25: 702–715 Hájková P, Petr L, Horsák M, Rohovec J, Hájek M (2015b) Interstadial inland dune slacks in south-west Slovakia: a multi-proxy vegetation and landscape reconstruction. Quat Internat 357: 314–328 Hájková P, Potůčková A, Hájek M (2017) Zajímavé nálezy vzácných mechorostů ve starých sedimentech 2. Bryonora 59: 67–71 Hájková P, Žáčková P, Dudová L, Hájek M (2012) Zajímavé nálezy vzácných mechorostů ve starých sedimentech. Bryonora 50: 14–16 Jamrichová E, Petr L, Jiménez- Alfaro B, Jankovská V, Dudová L, Pokorný P, Kołaczek P, Zernitskaya V, Čierniková M, Břízová E, Syrovátka V, Hájková P, Hájek M (2017) Pollen-inferred millennial changes in landscape patterns at a major biogeographical interface within Europe. J Biogeogr 44: 2386–2397 Jankovská V (1970) Ergebnisse der Pollen- and Grossrestanalyse des Moors „Velanská cesta“ in Südböhmen. Folia Geobot Phytotax 5: 43–60 Jankovská V (1980) Paläogeobotanische Rekonstruktion der Vegetationsentwicklung im Becken Třeboňská pánev während des Spätglazials und Holozäns. Academia, Praha

174

Janovicová K (1998) Zaujímavé bryofloristické nálezy zo Slovenska. Bryonora 21: 2–3 Janssen JA, Rodwell JS, García Criado M, Gubbay S, Haynes T, Nieto A, Sanders N, Landucci F, Loidi J, Ssymank A, Tahvanainen T, Valderrabano M, Acosta A, Aronsson M, Arts G, Attorre F, Bergmeier E, Bijlsma RJ, Bioret F, Biţæ-Nicolae C, Biurrun I, Calix M, Capelo J, Čarni A, Chytrý M, Dengler J, Dimopoulos P, Essl F, Gardfjell H, Gigante D, Giusso del Galdo G, Hájek M, Jansen F, Jansen J, Kapfer J, Mickolajczak A, Molina JA, Molnár Z, Paternoster D, Piernik A, Poulin B, Renaux B, Schaminée JHJ, Šumberová K, Toivonen H, Tonteri T, Tsiripidis I, Tzonev R, Valachovič M (2016) European Red List of Habitats. Part 2. Terrestrial and freshwater habitats. Publications Office of the European Union, Luxembourg Janssens JA (1983) A quantitative method for stratigraphic analysis of bryophytes in Holocene peat. J Ecol 71: 189–196 Joosten H, Tanneberger F, Moen A (eds) (2017) Mires and peatlands of Europe. Status, distribution and conservation. Schweizerbart Science Publishers, Stuttgart Juutinen S (2011) The decrease of rich fen bryophytes in springs as a consequence of large-scale environmental loss A 50-year re-sampling study Lindbergia 34: 2–8 Kaplan Z, Danihelka J, Štěpánková J, Ekrt L, Chrtek J Jr, Zázvorka J, Grulich V, Řepka R, Prančl J, Ducháček M, Kúr P, Šumberová K, Brůna J (2016) Distributions of vascular plants in the Czech Republic. Part 2. Preslia 88: 229–322 Kooijman AM (2012) ‘Poor rich fen mosses’: atmospheric N-deposition and P-eutrophication in base-rich fens. Lindbergia 35: 42–52 Kooijman AM, Beltman B, Westhoff V (1994) Extinction and reintroduction of the bryophyte Scorpidium scorpioides in a rich-fen site in the Netherlands. Biological Conservation 69: 87–96 Kuneš P, Odgaard Vad B, Gaillard MJ (2011) Soil phosphorus as a control of productivity and openness in temperate interglacial forest ecosystems. J Biogeogr 38: 2150–2164 Magyari E, Jakab G, Rudner E, Sümegi P (1999) Palynological and plant macrofossil data on Late Pleistocen short term climatic osscilations in North-east Hungary. Acta Palaeobot 2: 491–502 Malmer N (1986) Vegetation gradients in relation to environmental conditions in northwestern European mires. Canad J Bot 64: 375–383 Mälson K, Rydin H (2007) The regeneration capabilities of bryophytes for rich fen restoration. Biological Conservation 135: 435–442 Migra V, Mičeta K (1997) K výskytu Vignea chordorrhiza (Ehrh. ex L. fil.) Rchb. na Slovensku . Bull Slov Bot Spoločn 19: 29–31 Migra V, Šoltés R (1998) Glaciálny relikt Meesia triquetra (Bryophyta) na Kubínskej holi (Oravská Magura). Bull Slov Bot Spoločn 20: 48–50 Moen A, Lyngstad A, Øien D (2012) Boreal rich fen vegetation formerly used for haymaking. Nord J Bot 30: 226– 240 Mráz P, Ronikier M (2016). Biogeography of the Carpathians: evolutionary and spatial facets of biodiversity. Biol J Linn Soc 119: 528–559 Mucina L, Bültmann H, Dierssen K, Theurillat JP, Raus T, Čarni A, Šumberová K, Willner W, Dengler J, García RG, Chytrý M, Hájek M, Di Pietro R, Iakushenko D, Pallas J, Daniëls FJA, Bergmeier E, Guerra AS, Ermakov N, Valachovič M, Schaminée JHJ, Lysenko T, Didukh YP, Pignatti S, Rodwell JS, Capelo j, Weber HE, Solomeshch A, Dimopoulos P, Aguiar C, Hennekens SM, Tichý L (2016) Vegetation of Europe: Hierarchical floristic classification of vascular plant, bryophyte, lichen, and algal communities. Appl Veg Sci 19 (suppl. 1): 3–264 Navrátilová J, Hájek M, Navrátil J, Hájková P, Frazier RJ (2017) Convergence and impoverishment of fen communities in a eutrophicated agricultural landscape of the Czech Republic. Appl Veg Sci 20: 225–235 Nordhagen R (1943) Sikilsdalen og Norges fjellbeiter. En plantesosiologisk monografi. Bergens Mus. Skr. 22: 1– 607 Novák FA (1954) Přehled Československé květeny s hlediska ochrany přírody a krajiny. In Veselý J (ed) Ochrana Československé přírody a krajiny. Díl II. Nakladatelství ČSAV, Praha, pp 193–409 Paulissen MPCP, Schaminée JHJ, During HJ, Wieger Wamelink GW, Verhoeven JTA (2014) Expansion of acidophytic late successional bryophytes in Dutch fens between 1940 and 2000. J Veg Sci 25: 525–533 Pawlikowski P, Abramczyk K, Szczepaniuk A, Kozub Ł (2013) Nitrogen:phosphorus ratio as the main ecological determinant of the differences in the species composition of brown-moss rich fens in north-eastern Poland. Preslia 85: 349–367 Pearson RG (1965) Problems of post-Glacial refugia. Proc R Soc Lond 161: 324–330 Peterka T, Hájek M, Jiroušek M, Jiménez-Alfaro B, Aunina L, Bergamini A, Dítě D, Felbaba-Klushyna L, Graf U, Hájková P, Hettenbergerová E, Ivchenko TG, Jansen F, Koroleva NE, Lapshina ED, Lazarevid PM, Moen A,

175

Napreenko MG, Pawlikowski P, Plesková Z, Sekulová L, Smagin VA, Tahvanainen T, Thiele A, Biţa-Nicolae C, Biurrun I, Brisse H, Dušterevska R, De Bie E, Ewald J, FitzPatrick Ú, Font X, Jandt U, Kącki Z, Kuzemko A, Landucci F, Moeslund JE, Pérez-Haase A, Rašomavičius V, Rodwell JS, Schaminée JHJ, Šilc U, Stančid Z, Chytrý M (2017) Formalized classification of European fen vegetation at the alliance level. Appl Veg Sci 20: 124–142 Pokorný P, Chytrý M, Juřičková L, Sádlo J, Novák J, Ložek V (2015) Mid-Holocene bottleneck for central European dry grasslands: Did steppe survive the forest optimum in northern Bohemia, Czech Republic? The Holocene 25: 716–726 Rehell S, Virtanen R (2016) Rich-fen bryophytes in past and recent mire vegetation in a successional land uplift area. The Holocene 26: 136–146 Řepka R, Grulich V (2014) Ostřice České republiky. Lesnická Práce, Praha Růžička I (1987) Výsledky záchranného výzkumu ohrožené květeny mizejících rašelinišť a rašelinných luk v okolí Telče. Vlastiv Sborn Vysoč, Odd Věd Přír 8: 153–192 Rybníček K (1966) Glacial relics in the bryoflora of the highlands Českomoravská vrchovina (Bohemian- Moravian Highlands); their habitat and cenotaxonomic value. Folia Geobot Phytotax 1: 101–119 Rybníček K, Balátová-Tuláčková E, Neuhäusl R (1984) Přehled rostlinných společenstev rašelinišť a mokřadních luk Československa. Stud ČSAV 1984/8: 1–124 Rybníček K, Rybníčková E (1965) Přechodové rašeliniště u Bepadova na Oravě. Biológia 20: 373–375 Rybníček K, Rybníčková E (1968) The history of flora and vegetation on the Bláto mire in southeastern Bohemia, Czechoslovakia (palaeoecological study). Folia Geobot Phytotax 3: 117–142 Rybníček K, Rybníčková E (1972). Nálezy vzácných rašelinných mechorostů na Oravě. Biológia, 27: 795–798 Rybníčková E, Rybníček K (1972) Erste Ergebnisse paläogeobotanischer Untersuchungen des Moores bei Vracov, Südmähren. Folia Geobot Phytotax 7: 285–308 Singsaas S (1989) Classification and ordination of the mire vegetation of Stormyra near Tynset, S Norway. Nord J Bot 9: 413–423 Sjörs H (1952) On the relation between vegetation and electrolytes in north Swedish mire waters. Oikos 2: 241– 258 Sjörs H, Björkbäck F, Nordqvist Y (1965) Northern mires. Acta Phytogeogr Suec 50: 180–197 Šmilauer P, Lepš JŠ (2014) Multivariate Analysis of Ecological Data using CANOCO 5. Cambridge University Press, Cambridge Šoltés R, Dítě D, Mihálik D, Ondreičková K, Hrehová Z, Maximová N, Sedláková B (2015) Seasonal variation in bryophytes cover in the calcareous mire Belianské lúky, Slovakia. Pak J Bot 47: 255–262 Štechová T, Manukjanová A, Holá E, Kubešová S, Novotný I, Zmrhalová M (2010) Současný stav populací druhů Helodium blandowii (Thuidiaceae) a Scorpidium scorpioides (Calliergonaceae) v České republice. Bryonora 46: 24–33 Tichý L (2002) JUICE, software for vegetation classification. J Veg Sci 13: 451–453 Tutin TG, Heywood VH, Burges NA, Moore DM, Valentine DH, Walters SM, Webb DA (eds) (1968–1993) Flora Europaea, vols 2–5, 2nd edn. of Vol. 1. Cambridge University Press, Cambridge Udd D, Mälson K, Sundberg S, Rydin H (2015) Explaining species distributions by traits of bryophytes and vascular plants in a patchy landscape. Folia Geobot 50: 161–174 Udd D, Sundberg S, Rydin H (2016) Multi-species competition experiments with peatland bryophytes. J Veg Sci 27: 165–175 Vicherová E, Hájek M, Hájek T (2015) Calcium intolerance of fen mosses: physiological evidence, effects of nutrient availability and successional drivers. Persp Pl Ecol Evol Syst 17: 347–359 Vicherová E, Hájek M, Šmilauer P, Hájek T (2017) Sphagnum establishment in alkaline fens: Importance of weather and water chemistry. Sci Tot Environ 580: 1429–1438

176

Supplementary material

Table S1. List of closely related species merged to aggregates and species complexes. Name Included taxa Anthoxanthum odoratum agg. A. alpinum, A. odoratum Bryum pseudotriquetrum agg. B. bimum, B. pseudotriquetrum, B. subneodamense Campylium stellatum s.l. C. stellatum var. protensum, C. stellatum var. stellatum Carex buxbaumii agg. C. buxbaumii, C. hartmanii Carex flava agg. C. demissa, C. flava, C. jemtlandica, C. lepidocarpa, C. serotina Drepanocladus revolvens agg. D. cossonii, D. revolvens Galium palustre agg. G. elongatum, G. palustre Luzula campestris agg. L. campestris, L. multiflora, L. pallescens, L. sudetica Molinia caerulea agg. M. arundinacea, M. caerulea Philonotis fontana agg. P. fontana, P. tomentella Plagiomnium affine agg. P. affine, P. elatum, P. ellipticum, P. medium Utricularia intermedia agg. U. intermedia, U. ochroleuca Utricularia minor agg. U. bremii, U. minor Vaccinium oxycoccos agg. V. microcarpum, V. oxycoccos Valeriana dioica s.l. V. dioica subsp. simplicifolia, V. dioica subsp. dioica

Table S2. The phytosociological table of vegetation of Stygio-Caricion limosae (core plots) in the Carpathians and adjacent territories. The species covers were estimated using the seven-grade Braun-Blanquet´s scale (plots 1–6) and nine-grade Braun-Blanquet´s scale (plots 7–12). For table presentation, the Domin scale (used by Birks and Walters 1973, plot 13) was transformed into seven-grade Braun-Blanquet scale according to van der Maarel (1979). Diagnostic species of Stygio-Caricion limosae are adopted from Peterka et al. (2017). Species are sorted according to decreasing frequency. Water pH and conductivity, both standardized at 20 °C, were measured directly in situ mostly in a shallow depression with water within a plot using portable instruments. Plot 13: pH measured in lake littoral zone (see Birks and Walters 1973).

Plot number 1 2 3 4 5 6 7 8 9 10 11 12 13 Plot area (m2) 3 6 1.4 8 4 8 16 10 1 16 16 16 4 Total cover (%) 60 60 70 – – – – – – 90 – 75 80 Cover herb layer (%) 40 40 40 65 30 70 60 65 40 50 60 65 – Cover moss layer (%) 40 60 60 90 60 30 80 90 98 90 80 60 – Cover open water (%) 50 50 60 6 95 – 90 95 – – – – – pH – – – 6.3 6.9 6.3 5.7 5.9 7.3 6.5 6.1 6.8 5.5– 5.8 Conductivity (µS/cm) – – – 214 269 198 124 117 133 143 142 43 – Diagnostic species Carex limosa + 2 1 1 2 1 2a 1 2b 2a 2a 3 2 Menyanthes trifoliata 1 2 2 1 2 1 3 3 2a 1 2a 2b 2 Carex chordorrhiza 3 2 2 2 1 3 2b 2b . + 2a . . Scorpidium scorpioides 1 3 2 4 3 2 4 5 4 . . . 2 Utricularia minor agg. 2 1 2 . . . . . + . . 1 . Calliergon trifarium . r ...... 2b 2a + . Carex lasiocarpa . r ...... 1 4

Other species Drepanocladus revolvens agg. + 1 2 1 . 2 2a 2a 2b 4 3 2a . Carex rostrata . r . r + 1 1 1 + 2a 1 + . Eriophorum angustifolium 1 1 + 2 1 1 1 1 + . . . . Carex flava agg. + + + 2 + 1 2m 2a . 2a . . .

177

Plot number 1 2 3 4 5 6 7 8 9 10 11 12 13 Carex panicea + . + 1 1 + 2a 1 1 . . + . Bryum pseudotriquetrum . + + 1 1 . + 1 . + + + . Carex nigra + . . 1 1 . 1 1 . 1 1 . 2 Calliergon giganteum 3 2 3 2 + . . . + . . . 2 Eleocharis quinqueflora . . + 1 1 1 1 1 1 . . . . Campylium stellatum s.l...... 1 + 2a 1 2b 1 . Potentilla palustris 1 1 + . . 1 . + . . . . . Drosera anglica . . . 1 1 . 1 1 + . . . . Vaccinium oxycoccos agg. . . . 1 . 1 + + 1 . . . . Equisetum fluviatile . . . + 1 + . . + . . . 2 Sphagnum contortum . . . . 1 1 + + . . . . 1 Aneura pinguis ...... + + + + + . . Equisetum palustre + + 1 . . . . . + . . . . Parnassia palustris . . . + . . . . . + + + . Valeriana dioica s.l. + + ...... + . . . . (V. d. subsp. simplicifolia) Galium palustre agg. + r 1 ...... Agrostis canina + r ...... + . . . . Pinguicula vulgaris . . . r . . 1 1 . . . . . Plagiomnium affine agg...... 2 r + . . . . . Molinia caerulea agg...... + . 1 . . . + . Potentilla erecta ...... + + . + . Selaginella selaginoides ...... + + + . Veronica scutellata 1 . r ...... Galium uliginosum + + ...... Meesia triquetra r + ...... Mentha aquatica . + 1 ...... Viola palustris . r . . . . . + . . . . . Drepanocladus fluitans . . . 2 2 ...... Scheuchzeria palustris ...... + + . . . . . Eriophorum latifolium ...... + . . . + . Swertia perennis ...... 2a 2a . . Palustriella commutata ...... 1 1 . . Ligusticum mutellina ...... + + . . Dactylorhiza cordigera ...... + + . . Allium schoenoprasum ...... + + . . Pinguicula alpina ...... + + . . Oncophorus virens ...... + + . . Carex echinata ...... + . + . Equisetum variegatum 1 ...... Juncus articulatus + ...... Triglochin palustris + ...... Succisa pratensis . r ...... Caltha palustris . . r ...... Sphagnum teres ...... + . . . . . Marchantia polymorpha ...... + . . . . . Sphagnum warnstorfii ...... + . . . . Philonotis calcarea ...... + . . . . Picea abies ...... r . . . . Nardus stricta ...... 1 . . . Deschampsia cespitosa ...... + . . . Carex capillaris ...... 1 . . Juncus triglumis ...... + . .

178

Plot number 1 2 3 4 5 6 7 8 9 10 11 12 13 Hypnum pratense ...... + . . Leiocolea bantriensis ...... + . . Barbilophozia quadriloba ...... + . . Bartsia alpina ...... + . . Pedicularis verticillata ...... + . . Trisetum alpestre ...... + . . Soldanella hungarica ...... + . . Gymnadenia conopsea ...... r . . Sphagnum subsecundum ...... 3 . Drepanocladus exannulatus ...... 2a . Phragmites australis ...... 1 . Carex buxbaumii agg...... + . Scapania irrigua ...... + . Carex hostiana ...... + . Abies alba ...... r . Carex diandra ...... 2 Carex curta ...... 2 Hamatocaulis vernicosus ...... 2

Localities of the plots:

1–3. Slovakia, Bepadovo, 1–1,5 km SW of the village; approx. 49.42083, 19.32972 (Rybníček K., ined.) The locality was located southwest of the Bepadovo village in the alluvium of Mútpanka river (see also Rybníček and Rybníčková 1965) and was disturbed in the past, thus vegetation of the alliance does not already occur.

4–6: Slovakia, Kubínska hoľa Mt., W of Puchmajerovej jazierko Lake; approx. 49.288056, 19.318889 (Dítě and Pukajová 2002: table 1, plots 34–36) 7–8: dtto (Dítě and Kubandová 2004, page 46, plots 1, 2) The small lake called Puchmajerovej jazierko lies north of the peak of Minčol (part of Oravská Magura Mts, Kubínska hoľa Mt). Fen is situated to the west from the lake, surrounded by coniferous forests. The fen has originated in the Middle Holocene (ca 7300 calibrated years BP; Čierniková 2016), had continual linear sedimentation rate and almost from the beginning it has hosted relict bryophytes like Meesia triquetra. For photos of the locality see Fig. A.

Fig. A. The locality of Puchmajerovej jazierko Lake. Photo: P. Hájková (2013).

179

9: Poland, Poľana Biały potok; 49.28417, 19.84694 (Hájek M., Hájková P., Šímová A., ined.) Poľana Bialy Potok is one of the best preserved boreal fens in the Western Carpathians with small pools harbouring relict bryophytes. The fen has the whole-Holocene history as well, as indicated by its initiation between 9261–9187 BC (11.211–11.137 BP) and continual, yet slow sedimentation, with continuous curve of Cyperaceae pollen (L. Petr, unpublished data). For photos of the locality see Fig. B.

Fig. B. The locality of Poľana Biały potok. Photo: E. Hettenbergová (2017).

10–11: Romania, Știol; 47.57361, 24.80167; 47.57333, 24.8025 (Dítě D., Goia I., Hájek M., Hájková P., ined.) Fen at north-western foothills of Gærgælæu Peak (Rodna Mts, Știol). According to Coldea (1990) the vegetation is developed on 1.5–3 m deep peat. The water regime is undisturbed and the fen is not managed, except for occasional grazing. It is floristically unique site, only one compositionally similar yet slightly desiccating fen is located nearby. For photos of the locality see Fig. C.

Fig. C. The locality of Știol. Photo: P. Hájková (2014; left), T. Peterka (2017; right).

180

12: Montenegro, Durmitor, Barno jezero Lake; 43.15667, 19.09167 (Kalníková V., Palpurina S., Peterka T., Plesková Z., ined.) 13: dtto; (Birks and Walters 1973: page 14, plot 1) The small lake called Barno jezero lies west of Žabljak on a high limestone plateau at the massif of the Durmitor Mts, a part of the Dinarids Mts. The remarkable feature of this region is presence of a number of lakes of presumed glacial origin (Birks and Walters 1973). The lake lies in a shallow depression among moraines (Adamovid et al. 1996). The littoral zone of the lake is occupied by fens and tall sedges. For photos of the locality see Fig. D.

Fig. D. The locality of Barno jezero Lake. Photo: V. Kalníková (2014).

181

Table S3. Localities of vegetation-plot records of the Stygio-Caricion limosae alliance in the Carpathians and adjacent territories. c/n. c: core plots, n: non-core plots Reg. (Region). EC: Eastern Carpathians, WC: Western Carpathians, BM: Bohemian Massif, D: Durmitor Country. CZ: Czech Republic, Mon: Montenegro, PL: Poland, RO: Romania, SK: Slovak Republic Year. Year of sampling. Original classification. AmbSco-CarCho: Amblystegio scorpioidis-Caricetum chordorrhizae, AmbSte-CarPan: Amblystegio stellati-Caricetum paniceae, CarCan-Fus: Caricion canescenti-fuscae, CarFusDre: Caricetum fuscae drepanocladetosum, CarCho: Caricetum chordorrhizae, CarCho-SphObt: Carici chordorrhizae-Sphagnetum obtusi, CarDem: Caricion demissae, CarLas: Caricion lasiocarpae, Chr-TriAlp: Chrysohypno-Trichophoretum alpini, DreRev-CarLas: Drepanoclado revolventis-Caricetum lasiocarpae, EriGra: Eriophorion gracilis, Mee-CarCho: Meesio-Caricetum chordorrhizae, Peu-CarLas: Peucedano-Caricetum lasiocarpae, Sco-Utr: Scorpidio-Utricularietum, SphTom: Sphagno-Tomentypnion, SphUtr: Sphagno-Utricularion, SwePer-CarCho: Swertio perennis-Caricetum chordorrhizae CL. Cluster number in Table S4 in Electronic supplementary material FR. Number of full vegetation plot (relevé) in Table S2 in Electronic supplementary material c/n Reg. Country Locality Source Year Original classification CL FR c WC SK Podbeskydská brázda Furrow, Bepadovo, 1–1.5 SW of the village Rybníček K. (ined.) 1964 CarLas, 4 1 AmbSco-CarCho c WC SK Podbeskydská brázda Furrow, Bepadovo, 1–1.5 SW of the village Rybníček K. (ined.) 1964 CarLas, 4 3 AmbSco-CarCho c WC SK Podbeskydská brázda Furrow, Bepadovo, 1–1.5 SW of the village Rybníček K. (ined.) 1977 CarLas, 1 2 AmbSco-CarCho c WC SK Oravská Magura Mts, Kubínska hoľa, W of Puchmajerovej jazierko Lake Dítě and Pukajová (2002); tab. 1, 2001 CarLas, 2 4 plot 34 AmbSco-CarCho c WC SK Oravská Magura Mts, Kubínska hoľa, W of Puchmajerovej jazierko Lake Dítě and Pukajová (2002); tab. 1, 2001 CarLas, 1 5 plot 35 AmbSco-CarCho c WC SK Oravská Magura Mts, Kubínska hoľa, W of Puchmajerovej jazierko Lake Dítě and Pukajová (2002); tab. 1, 2001 CarLas, 2 6 plot 36 AmbSco-CarCho c WC SK Oravská Magura Mts, Kubínska hoľa, W of Puchmajerovej jazierko Lake Dítě and Kubandová (2005); p. 46, 2004 CarLas, 2 7 plot 1 AmbSco-CarCho c WC SK Oravská Magura Mts, Kubínska hoľa, W of Puchmajerovej jazierko Lake Dítě and Kubandová (2005); p. 46, 2004 ? CarCho-SphObt 4 8 plot 2 c WC PL Kościelisko-Biały Potok, 0.5 km W of the village, Polana Bialy Potok fen Hájek M., Hájková P., Šímová A. 2012 - 4 9 (ined.) c EC RO Rodna Mts, Știol, Gærgælæu Peak, NW foothills Dítě D., Hájek M., Hájková P., Goia 2014 - 4 10 I. (ined.) c EC RO Rodna Mts, Știol, Gærgælæu Peak, NW foothills Dítě D., Hájek M., Hájková P., Goia 2014 - 4 11 I. (ined.) c D Mon Durmitor Mts, Žabljak, Barno jezero Lake Birks and Walters (1972); p. 14 1971 - 3 13 c D Mon Durmitor Mts, Žabljak, Barno jezero Lake Peterka T., Kalníková V., Palpurina 2014 - 4 12 S., Plesková Z. (ined.)

c/n Reg. Country Locality Source Year Original classification CL FR n WC SK Liptovská kotlina Basin, Dúbrava-Chraste, on the rigth bank of the brook Dítě D. (ined.) 2003 - 4 - of Čemník n WC SK Oravská kotlina Basin, between villages of Ťapešovo and Lokca Rybníček K. (ined.), 2 plots 1971 - 4 - n WC SK Oravská kotlina Basin, Bobrov Rybníček K. (ined.) 1968 - 2 - n WC SK Oravská kotlina Basin, Oravská Polhora, 2 km E of the village, at the Biela Dítě D. (ined.) 2003 - 4 - farma hotel n WC SK Oravská kotlina Basin, Oravská Polhora, Slaná voda Rybníček K. (ined.) 1970 - 4 - n WC SK Oravská kotlina Basin, Trstená Rybníček K. (ined.) 1977 - 4 - n WC SK Oravská kotlina Basin, Trstená, 5 km NE of the town Hájek M. (ined.) 2004 - 4 - n WC SK Oravská Magura Mts, Kubínska hoľa, W od Puchmajerovej jazierko Lake Hájek M. (ined.) 1999 - 4 - n WC SK Oravská Magura Mts, Kubínska hoľa, W od Puchmajerovej jazierko Lake Dítě and Pukajová (2002); tab. 1, 2001 CarLas, AmbSco-CarCho 4 - plot 33 n WC SK Podbeskydská brázda Furrow, Bepadovo, 1–1.5 SW of the village Rybníček K. (ined.) 1968 - 2 - n WC SK Podbeskydská brázda Furrow, Bepadovo, 1–1.5 SW of the village Rybníček K. (ined.) 1968 - 4 - n WC SK Podbeskydská brázda Furrow, Bepadovo, 1–1.5 SW of the village, Dítě D. (ined.) 2002 - 4 - Bepadovské rašelinisko reserve n WC SK Podtatranská brázda Furrow, Oravice, Peciska Dítě D. (ined.) 2003 - 4 - n WC SK Spišská kotlina Basin, Spišská Belá-Šarpanec *Belianske lúky+ Rybníček K. (ined.) 1975 - 4 - n WC SK Spišská kotlina Basin, Spišská Belá-Šarpanec *Belianske lúky+ Rybníček K. (ined.), 2 plots 1979 - 4 - n WC SK Spišská kotlina Basin, Spišská Belá-Šarpanec, *Belianske lúky+ Dítě D. (ined.) 2003 - 4 - n EC RO Rodna Mts, Știol Coldea (1990); tab. 42, plot 3 1982 CarLas, SwePer-CarCho 4 - n EC RO RodnaMts, Știol Coldea (1990); tab. 42, plot 5 1982 CarLas, SwePer-CarCho 4 - n EC RO RodnaMts, Știol Coldea (1990); tab. 42, plot 10 1982 CarLas, SwePer-CarCho 4 - n D Mon Durmitor Mts, Žabljak, Barno jezero Lake Birks and Walters (1973); tab. 1, 1971 - 4 - plot 31 n D Mon Durmitor Mts, Žabljak, Barno jezero Lake Peterka T., Kalníková V., Palpurina 2014 - 4 - S., Plesková Z. (ined.) n BM CZ Bohemian-Moravian highlands, Dačice-Lipolec, 1.5 km NW of the village, Rybníček (1974); tab. 8, plot 12 1966 CarDem, Chr-TriAlp 4 - close to the Nové Dvory settlement n BM CZ Bohemian-Moravian highlands, Dačice-Lipolec, 1.5 km NW of the village, Rybníček (1974); tab. 8, plot 15 1966 CarDem, Chr-TriAlp 4 - close to the Nové Dvory settlement n BM CZ Bohemian-Moravian highlands, Kunžak-Suchdol, SE of the Jalovčí Rybníček (1974); tab. 5, plot 3 1961 CarDem, Sco-Utr 4 - settlement

c/n Reg. Country Locality Source Year Original classification CL FR n BM CZ Bohemian-Moravian highlands, Kunžak-Suchdol, SE of the Jalovčí Rybníček (1974); tab. 8, plot 36 1961 CarDem, Chr-TriAlp 4 - settlement n BM CZ Bohemian-Moravian highlands, Kunžak-Suchdol, SE of the Jalovčí Rybníček (1974); tab. 8, plot 31 1962 CarDem, Chr-TriAlp 4 - settlement n BM CZ Bohemian-Moravian highlands, Doupě, close to Panský pond Rybníček (1974); tab. 8, plot20 1961 CarDem, Chr-TriAlp 4 - n BM CZ Bohemian-Moravian highlands, Dušejov, 1 km W of the village, close to Rybníček (1974); tab. 5, plot 4 1961 CarDem, Sco-Utr 4 - Jedlovský pond n BM CZ Bohemian-Moravian highlands, Dušejov, 1 km W of the village, close to Rybníček (1974); tab. 8, plot 22 1961 CarDem, Chr-TriAlp 4 - Jedlovský pond n BM CZ Bohemian-Moravian highlands, Dušejov, 1 km W of the village, close to Rybníček (1974); tab. 8, plot 30 1961 CarDem, Chr-TriAlp 4 - Jedlovský pond n BM CZ Bohemian-Moravian highlands, Dušejov, 1 km W of the village, close to Rybníček (1974); tab. 5, plot 10 1964 CarDem, Sco-Utr 4 - Jedlovský pond n BM CZ Bohemian-Moravian highlands, Heřmaneč, between villages of Heřmaneč Rybníček (1974); tab. 8, plot 13 1966 CarDem, Chr-TriAlp 4 - and Léskovec n BM CZ Bohemian-Moravian highlands, Jezdovice, 1.5 km W of the village Rybníček (1974); tab. 5, plot 9 1962 CarDem, Sco-Utr 4 - n BM CZ Bohemian-Moravian highlands, Jezdovice, 1.5 km W of the village Rybníček (1974); tab. 5, plot 8 1964 CarDem, Sco-Utr 4 - n BM CZ Bohemian-Moravian highlands, Jihlávka, 1.5 km SSW of the village, Rybníček (1974); tab. 8, plot 24 1960 CarDem, Chr-TriAlp 4 - V Lísovech n BM CZ Bohemian-Moravian highlands, Jihlávka, 1.5 km SSW of the village, Rybníček (1974); tab. 5, plot 6 1962 CarDem, Sco-Utr 4 - V Lísovech n BM CZ Bohemian-Moravian highlands, Jihlávka, close to the railway station, Rybníček (1974); tab. 5, plot 7 1960 CarDem, Sco-Utr 4 - locality Kaliště n BM CZ Bohemian-Moravian highlands, Jihlávka, close to the railway station, Rybníček (1974); tab. 8, plot 26 1960 CarDem, Chr-TriAlp 4 - locality Kaliště n BM CZ Bohemian-Moravian highlands, Jihlávka, close to the railway station, Rybníček (1974); tab. 40, plot 5 1960 EriGra, Mee-CarLim 4 - locality Kaliště n BM CZ Bohemian-Moravian highlands, Jihlávka, close to the railway station, Rybníček (1974); tab. 40, plot 6 1960 EriGra, Mee-CarLim 4 - locality Kaliště n BM CZ Bohemian-Moravian highlands, Jihlávka, close to the railway station, Rybníček (1974); tab. 40, plot 8 1960 EriGra, Mee-CarLim 4 - locality Kaliště n BM CZ Bohemian-Moravian highlands, Jihlávka, close to the railway station, Rybníček (1974); tab. 5, plot 5 1962 CarDem, Sco-Utr 4 - locality Kaliště n BM CZ Bohemian-Moravian highlands, Jihlávka, close to the railway station, Rybníček (1974); tab. 45, plot 2 1962 EriGra, CarCho 2 - locality Kaliště n BM CZ Bohemian-Moravian highlands, Jihlávka, close to the railway station, Rybníček (1974); tab. 45, plot 3 1962 EriGra, CarCho 2 - locality Kaliště

c/n Reg. Country Locality Source Year Original classification CL FR n BM CZ Bohemian-Moravian highlands, Jihlávka, close to the railway station, Rybníček and Rybníčková (1961), CarCan-Fus, CarFusDre 4 - northern margin of pond p.79 n BM CZ Bohemian-Moravian highlands, Popice, 0.3 km SW of Popice, close to Rybníček (1974); tab. 8, plot 21 1962 CarDem, Chr-TriAlp 4 - Hrazený pond n BM CZ Bohemian-Moravian highlands, Radostín, margin of the Padrtiny mire, Rybníček (1974); tab. 5, plot 2 1957 CarDem, Sco-Utr 4 - below the hill Radostínský kopec n BM CZ Bohemian-Moravian highlands, Řásná, 1 km SE of the village, close to Rybníček (1974); tab. 40, plot 10 1965 EriGra, Mee-CarLim 4 - Smrkovský pond n BM CZ Bohemian-Moravian highlands, Řásná, 1 km SE of the village, close to Rybníček (1970); tab. 3, plot 535 1965 CarDem, Chr-TriAlp 4 - Smrkovský pond n BM CZ Bohemian-Moravian highlands, Ústí, close to the Klábovka settlement Rybníček (1974); tab. 5, plot 1 1964 CarDem, Sco-Utr 2 - n BM CZ Bohemian-Moravian highlands, Vilánec, 1.7 km NWW of the village Rybníček (1974); tab. 8, plot 17 1960 CarDem, 4 - Chr-TriAlp n BM CZ Bohemian-Moravian highlands, Vilánec, 1.7 km NWW of the village Rybníček (1974); tab. 8, plot 16 1960 CarDem, 4 - Chr-TriAlp n BM CZ Bohemian-Moravian highlands, Vilánec, 1.5 km N of the village Rybníček (1974); tab. 8, plot 14 1960 Chr-TriAlp 4 - n BM CZ Bohemian-Moravian highlands, Žďárské vrchy Hills, Žďár nad Sázavou- Rybníček (1974); tab. 40, plot 13 1961 EriGra, 4 - Zámek, close to the northern shore of Pilský pond Mee-CarLim n BM CZ Ralsko-bezdězská tabule Plateau, Doksy region, Doksy-Břehyně, a swamp Rybníček (1970); tab. 1, plot 446 Caricetum lasiocarpae 3 - NE of the fish-pond Břehypský rybník, close to the town of Doksy n BM CZ Ralsko-bezdězská tabule Plateau,Doksy region, Doksy-Břehyně, NE shore Rybníček (1970); tab. 1, plot 445 1963 Sco-Utr 3 - of Břehypský rybník pond n BM CZ Ralsko-bezdězská tabule Plateau,Doksy region, Doksy-Břehyně, the Štechová et al. (2010); tab. 1., plot 2009 - 3 - Břehyně-Pecopala Nature Reserve 5 n BM CZ Ralsko-bezdězská tabule Plateau,Doksy region, Doksy-Staré splavy, 1 km N Hájková P. (Czech National Phytos. 2005 CarLas, 4 - of the village, former pond called Baronský rybník Database, plot number: 478937) DreRev-CarLas n BM CZ Ralsko-bezdězská tabule Plateau,Doksy region, Jestřebí, meadow in wood Rybníček (1970); tab. 1, plot 526 1963 SphUtr, Sco-Utr 4 - at railway guard´s house E of line Jestřebí-Staré Splavy n BM CZ Northern Bohemia, Manušice close to the Česká Lípa, mire between the S Rybníček (1970); tab. 1, plot 444 1963 SphUtr, Sco-Utr 2 - end of the village and the first fishpond below the village n BM CZ Třebop Basin, Dolní Miletín, northern margin of Nový pond Březina et al. (1963); p. 239, plot 3 1954 CarDem, 4 - AmbSte-CarPan n BM CZ Třebop Basin, Ponědrážka, W edge of the mire complex between Rybníček (1970); tab. 3, plot 479 - SphTom, 4 - fishponds Švarcenberský a Horusický rybník Chr-TriAlp n BM CZ Třebop Basin, Ponědrážka, W edge of the mire complex between Rybníček (1970); tab. 1, plot 512 - Caricetum lasiocarpae 3 - fishponds Švarcenberský a Horusický rybník

Table S4. Results of K-means clustering. Synoptic table of clusters of vegetation plots. The frequency values are shown. The frame and background shading indicate diagnostic species of cluster in the cases when phi > 0.3. As dominant species marked with double asterisk (**) are regarded those reaching the cover higher than 25% in more than 25% plots forming the cluster. Diagnostic species are sorted according to decreasing fidelity, other species are sorted according to frequency within the dataset. Only species with at least 25 occurrences within the analysed dataset are shown.

Cluster no. 1 2 3 4 5 6 No. of plots 184 172 177 157 215 109

Ratio of core / non-core plots 146/38 78/94 111/66 24/133 36/179 29/80 the Western Carpathians 2/– 3/2 –/– 4/16 –/– –/– the Eastern Carpathians –/– –/– –/– 2/3 –/– –/– the Durmitor Mts –/– –/– 1/– 1/2 –/– –/– the Bohemian Massif –/– –/4 –/4 –/34 –/– –/–

Carex limosa 98** 40 65 37 34 77 Carex livida 45 11 19 3 12 7 Eriophorum angustifolium 40 85 33 65 43 32 Carex lasiocarpa 16 29 99** 37 52 30 Juncus articulatus 1 . . 18 . . Carex flava agg. 9 28 18 59 27 2 Potentilla erecta 1 1 . 29 15 . Triglochin palustris 2 5 1 27 7 . Parnassia palustris . 1 1 22 5 3 Valeriana dioica s.l. 2 1 . 15 . . Carex panicea 6 48 16 59 31 1 Agrostis canina 1 . 3 16 1 1 Scirpus cespitosus 10 48 2 3 91** 7 Selaginella selaginoides . 3 . 12 44 . Andromeda polifolia 18 28 15 25 83 21 Molinia caerulea agg. 1 8 1 19 49 2 Tofieldia pusilla 1 3 . 8 33 2 Drepanocladus badius 3 6 2 2 29 4 Betula nana 2 9 2 8 32 7 Scirpus hudsonianus 16 33 14 23 58 13 Epilobium palustre . 1 5 4 2 30 Carex diandra 3 4 5 5 2 34 Potentilla palustris 9 14 11 24 12 54 Galium trifidum . . . . . 14 Drepanocladus tundrae 9 1 6 3 5 30 Calliergon richardsoni . . 2 1 1 16 Stellaria crassifolia . 1 . . 1 12 Drepanocladus revolvens agg. 21 37 25 96** 96** 22 Campylium stellatum s.l. 16 23 19 82 90** 7

Other species Scorpidium scorpioides 97** 98** 100** 50 48 48 Menyanthes trifoliata 75 54 75 73 63 85 Carex rostrata 52 34 60 67 59 70 Carex chordorrhiza 64 78** 28 38 41 67 Equisetum fluviatile 57 25 58 37 34 63 Cinclidium stygium 18 15 23 38 54 22 Calliergon trifarium 43 14 40 31 28 6

186

Cluster no. 1 2 3 4 5 6 Aneura pinguis 24 13 23 42 30 7 Drosera anglica 43 8 31 10 18 14 Utricularia minor agg. 34 7 31 27 7 17 Utricularia intermedia agg. 39 5 38 3 6 22 Carex nigra 12 15 10 42 16 20 Pedicularis palustris 19 5 27 25 7 25 Vaccinium oxycoccos agg. 10 10 25 22 15 22 Carex dioica 5 17 3 24 35 12 Calliergon giganteum 9 6 8 32 2 34 Myrica gale 18 1 12 3 22 19 Betula pubescens 5 3 9 7 26 19 Equisetum palustre 4 10 6 29 16 4 Bryum pseudotriquetrum 6 2 4 31 10 22 Drepanocladus exannulatus 16 10 12 9 2 23 Drosera rotundifolia 9 1 10 21 11 11 Eriophorum latifolium 2 5 2 20 25 1 Pinguicula vulgaris 1 12 . 20 20 1 Phragmites australis 9 6 6 12 13 10 Salix phylicifolia 12 6 9 3 9 17 Calliergon stramineum 2 16 5 7 7 13 Calliergon sarmentosum 3 10 4 7 15 6 Sphagnum contortum 7 6 7 21 3 3 Eleocharis quinqueflora 7 7 3 20 2 1 Carex buxbaumii agg. 2 14 1 7 12 . Rhynchospora alba 1 1 20 8 1 3 Sphagnum subsecundum 3 3 5 6 8 9 Meesia triquetra 2 3 1 20 2 6 Paludella squarrosa 1 1 1 10 10 10 Carex magellanica 1 17 1 6 5 2 Viola palustris 3 4 1 17 1 4 Thalictrum alpinum . 3 . 4 . 17 Sphagnum warnstorfii 2 2 1 7 5 12 Galium palustre agg. 2 3 3 9 16 1 Polygonum viviparum . 3 1 16 . 7 Pinus sylvestris 2 . 2 3 4 12 Salix lapponum 2 10 1 5 1 3 Salix myrsinites . 2 1 6 3 9 Tomenthypnum nitens . 1 . 14 2 5 Fissidens adianthoides . . . 16 . 6 Sphagnum teres 3 1 3 5 9 3 Schoenus ferrugineus . 4 1 3 . 10 Scheuchzeria palustris 4 1 13 1 . 1 Riccardia chamaedryfolia 7 5 2 . 3 2 Hamatocaulis vernicosus 1 1 1 8 12 . Peucedanum palustre . . 1 6 16 . Viola epipsila . . . 3 . 11 Hammarbya paludosa 8 1 5 1 1 1 Aulacomnium palustre 1 1 1 6 6 3 Eriophorum gracile 1 3 3 . 9 1 Juncus alpinus . 1 2 12 1 1 Juncus stygius 5 2 2 . 3 2 Carex echinata 1 1 . 12 1 1

187

The non-hierarchical K-means classification algorithm distinguished six clusters roughly corresponding with the association level. Plots from northern Europe are evenly distributed across all clusters. Plots from the Carpathians and adjacent territories (the Bohemian Massif and the Durmitor Mts) are distributed across clusters 1–4. Cluster 5–6 comprise exclusively plots from northern Europe. Cluster 1 includes vegetation of extremely waterlogged fens dominated by Carex limosa and Scorpidium scorpioides, which are traditionally classified within the Amblystegio scorpioidis-Caricetum limosae association. The group contains one plot from the Bepadovo locality and one from the fen close to the Puchmajerovej jazierko Lake (see Table S3). Cluster 2 comprises stands dominated by Carex chordorrhiza and Scorpidium scorpioides, which can be interpreted as the Amblystegio scorpioidis-Caricetum chordorrhizae association. Eriophorum angustifolium was identified as diagnostic species of this group. This fact suggests that the Stygio-Eriophoretum polystachyi association distinguished by Nordhagen (1943) might not be clearly separated from Amblystegio scorpioidis- Caricetum chordorrhizae. Within the study area, the cluster contains 3 core plots from Puchmajerovej jazierko Lake and several non-core plots from the Western Carpathians and the Bohemian Massif 4. Cluster 3 represents stands dominated by Scorpidium scorpioides and tall-sedge Carex lasiocarpa. These communities correspond to the Stygio-Caricetum lasiocarpae association. Besides data from northern Europe, the cluster contains one core plot from Barno jezero Lake and several non-core plots from the Bohemian Massif (mostly from the Doksy region; Table S3). As compared to previous groups, bryophyte layer of plots in clusters 4–5 is largely formed by semiaquatic but not submergely growing mosses Campylium stellatum and Drepanocladus revolvens agg. Scorpidium scorpioides and Calliergon trifarium still occur frequently, though they do not belong among dominant mosses. Cluster 4 is differentiated by the species indicating increased calcium content (Parnassia palustris, Triglochin palustris and Valeriana dioica s.l.), several short sedges (Carex flava agg., C. panicea) and other graminoids (Agrostis canina, Juncus articulatus). The group contains remaining core and non-core plots from the Western Carpathians, the Durmitor Mts and the Bohemian Massif. Rybníček (1974) mostly classified these communities as Chrysohypno-Trichophoretum alpini Březina et al. 1963 and Meesio-Caricetum limosae Rybníček 1974 (see Table S3). In that concept, the Chrysohypno-Trichophoretum alpini association was however delimited by dominance of brown-mosses and presence of Scirpus hudsonianus, which contradicts the type material of the association. Original description of this association (Březina et al. 1963) comprises 5 plots with Sphagnum species (S. contortum, S. warnstorfii, late-successional S. palustre, S. recurvum agg.). Hence, Chrysohypno-Trichophoretum alpini is not adequate name for mostly non-sphagnaceous fens. We suggest that Carici limosae-Sphagnetum contorti Warén 1926 (syn. Meesio-Caricetum limosae) might be suitable denomination of relict brown-moss fens with boreal elements occurring at sub-neutral pH and high (but not extremely high) water level. Contrary to extremely rich fens of Caricion davallianae, they lack calcicole species

4 Rybníček et al. (1984) and Hájek & Háberová (2001) recognized the associations Amblystegio scorpioidis-Caricetum chordorrhizae and Amblystegio scorpioidis-Caricetum limosae, both recognized by Nordhagen (1943) within the description of the Stygio-Caricion limosae alliance, though placed them into the Caricion lasiocarpae alliance, i.e. widely defined, and ambiguously interpreted, phytosociological unit. The ambiguous status of Caricion lasiocarpae stems partly from broad interpretation of this phytosociological unit and partly from its various interpretations in different vegetation surveys (see Mucina et al. 2016 for overview). The original description of Caricion lasiocarpae (Vanden Berghen in Lebrun et al. 1949) includes different types of wetland vegetation with tall sedges, including also herb vegetation on mesotrophic organic muddy sediments dominated by Calla palustris. In the concept introduced by Rybníček et al. (1984), the alliance comprised both quaking rich fens (Amblystegio scorpioidis- Caricetum chordorrhizae, Amblystegio scorpioidis-Caricetum limosae) and calcareous spring fens (Valeriano-Caricetum flavae Pawłowski et al. 1960, Carici flavae-Eriophoretum latifolii Soó 1944). The latter associations were characterized by presence of purely calcicole species (Eriophorum latifolium, Pinguicula vulgaris) and species of wet meadows that hardly ever enter pristine quaking fens (Briza media, Caltha palustris, Lythrum salicaria, Mentha aquatica, Prunella vulgaris, Ranunculus acris). Therefore, the Caricion lasiocarpae alliance was indistinctly separated from the temperate calcareous spring fens of the Caricion davallianae alliance. Later, this classification approach was adopted and, to a certain extent, re- evaluated by Hájek and Háberová (2001), who understood Caricion lasiocarpae as sub-alkaline fens developed under stable water table on organic substrates. Finally, Dítě et al. (2007) placed Amblystegio scorpioidis-Caricetum limosae into Caricion davallianae and Amblystegio scorpioidis-Caricetum chordorrhizae into Sphagno warnstorfii-Tomentypnion nitentis. This solution reflected the rare phytosociological material of both associations available from Slovakia. The differences between Caricion davallianae and Stygio-Caricion limosae were discussed above. The Sphagno warnstorfii-Tomentypnion nitens alliance as defined within its original description (Dahl 1956) however comprises vegetation with increased cover of Sphagnum species and with species of drier (i.e. oxic) conditions, e.g. Anthoxanthum odoratum agg., Luzula campestris agg. and Rumex acetosa.

188

(Blysmus compressus, Carex davalliana, Philonotis calcarea, Primula farinosa). Contrary to the core of Stygio- Caricion limosae, their bryophyte layer is dominated by other brown mosses (Calliergon giganteum, Campylium stellatum, Drepanocladus revolvens agg.) and Calliergon trifarium plus Scorpidium scorpioides are absent or reach lower cover. Contrary to rich fens with calcium-tolerant sphagna (Sphagno warnstorfii-Tomentypnion nitentis), Sphagnum species barely occur, except for the submergely growing Sphagnum contortum. The association represents a transition between quaking brown-moss fens of the Stygio-Caricion limosae alliance and initial type of the Sphagno warnstorfii-Tomentypnion nitentis alliance. This cluster further contains 5 plots sampled in the Eastern Carpathians (the Rodna Mts, Romania). This vegetation comprises sedge-moss communities with both Carex chordorrhiza and C. limosa on deep peat fen with sub-neutral reaction. The stands are further enriched by species of mountain alkaline springs (Allium sibiricum, Pinguicula alpina, Swertia perennis) and grasslands (Deschampsia cespitosa, Ligusticum mutellina). Coldea (1990) distinguished the Swertio perennis-Caricetum chordorrhizae association and included it into the Caricion lasiocarpae alliance. It is a question whether to recognize Swertio perennis-Caricetum chordorrhizae as a separate local association within Stygio-Caricion limosae. This vegetation and its relationship to other fen communities deserve further research and more phytosociological data is obviously needed for its critical syntaxonomical evaluation. Physiognomy of the communities in Cluster 5 is determined by Scirpus cespitosus and Scirpus hudsonianus. Other diagnostic species are typical elements of boreal rich fens, e.g. Andromeda polifolia, Drepanocladus badius, Selaginella selaginoides and Tofieldia pusilla. The vegetation can be interpreted as Drepanoclado revolventis-Trichophoretum cespitosi Nordhagen 1928. Cluster 6 is delimited partly by species spanning from fens to other oligotrophic wetlands (Carex diandra, Potentilla palustris) and partly by species with boreal, boreo-continental or arcto-boreal distributional ranges (Calliergon richardsonii, Drepanocladus tundrae, Galium trifidum, Stellaria crassifolia). This cluster may represent transition towards marshes on oligotrophic organic sediments (Magno-Caricion elatae) in the boreal zone of Europe. The clusters are clearly separated in the NMDS analysis (Fig. E). The first NMDS axis can be interpreted as the gradient of water table depth stretching from extremely waterlogged fens (Amblystegio scorpioidis- Caricetum limosae; left part of the diagram) across less waterlogged fens (Amblystegio scorpioidis-Caricetum chordorrhizae, Stygio-Caricetum lasiocarpae) to fens with water table depth equal to the level of bryophyte layer or reaching slightly below the surface of bryophyte layer (Clusters 4–5; right part of the diagram). The second axis separated group of plots transient to communities of oligotrophic marshes (Cluster 6). The clusters related to the associations Amblystegio scorpioidis-Caricetum chordorrhizae and Stygio-Caricetum lasiocarpae are distinguished along the third axis.

Fig. E. NMDS analysis. Plots are classified according to results of K-means clustering

189

Table S5. Syntaxonomical synopsis of the Stygio-Caricion limosae alliance in the Carpathians and adjacent territories based on a literature review and our results. Nomenclature of higher syntaxa (class, order) adopted from Mucina et al. (2016). Associations that need further syntaxonomical evaluation and further research (see also Table S4) are marked by asterisk (*). Scheuchzerio palustris-Caricetea fuscae Tx. 1937 Sphagno warnstorfii-Tomentypnetalia Lapshina 2010 Stygio-Caricion limosae Nordhagen 1943 Amblystegio scorpioidis-Caricetum limosae Osvald 1923 Amblystegio scorpioidis-Caricetum chordorrhizae Osvald 1925 * Carici limosae-Sphagnetum contorti Warén 1926 Stygio-Caricetum lasiocarpae Nordhagen 1943 * Swertio perennis-Caricetum chordorrhizae Coldea 1990

Table S6. Potential historical localities of brown-moss rich quaking fens identified according to floristic data. C. = Carex, Calliergon = Calliergon trifarium, Scorpidium = Scorpidium scorpioides Locality Diagnostic Diagostic species Vegetation data of species – moss - sedge Stygio-Caricion limosae from the locality SK, Kubinská hol'a Mt, fen slose to Puchmajerovej Scorpidium C. chordorrhiza, core plot jazierko Lake C. limosa SK, Podbeskydská brázda Furrow, Bepadovo Calliergon, C. chordorrhiza, core plot Scorpidium C. lasiocarpa, C. limosa SK, Podtatranská brázda Furrow, Oravice, Peciská Calliergon C. lasiocarpa non-core plot SK, Spišská kotlina Basin, Šarpanec, Belianske lúky Calliergon, C. limosa non-core plot Scorpidium CZ, Bohemian-Moravian Higlands, Dušejov, Chvojnov Calliergon, C. lasiocarpa non-core plot mire Scorpidium CZ, Bohemian-Moravian Higlands, Jihlavské vrchy Calliergon, C. chordorrhiza, non-core plot Hills, Jihlávka, Kaliště mire Scorpidium C. lasiocarpa, C. limosa CZ, Bohemian-Moravian Higlands, Jihlavské vrchy Calliergon, C. lasiocarpa, non-core plot Hills, Jihlávka, V Lísovech mire Scorpidium C. limosa CZ, Bohemian-Moravian Higlands, Jihlavské vrchy Scorpidium C. limosa non-core plot Hills, Kunžak-Suchdol CZ, Bohemian-Moravian Higlands, Jihlavské vrchy Scorpidium C. limosa non-core plot Hills, Řásná CZ, Bohemian-Moravian Higlands, Žďárské vrchy Scorpidium C. lasiocarpa - Hills, Radostín, Malé Dářko CZ, Bohemian-Moravian Higlands, Žďárské vrchy Scorpidium C. chordorrhiza, non-core plot Hills, Radostín, Velké Dářko (Padrtiny mire) C. lasiocarpa CZ, eastern basin of the Labe river, Petrovice Scorpidium C. lasiocarpa - CZ, middle basin of the Labe river, Hrabanov, Scorpidium C. lasiocarpa - Hrabanovská černava CZ, middle basin of the Labe river, Všetaty, ? Scorpidium C. lasiocarpa - Všetatská černava fen CZ, Ralsko-bezdězská tabule Plateau, Doksy, Calliergon, C. chordorrhiza, non-core plot surroundings of Máchovo jezero pond and Scorpidium C. lasiocarpa, Břehypský rybník pond C. limosa CZ, Třebop Basin, Lužnice, Velký Tisý pond Scorpidium C. lasiocarpa, - C. limosa CZ, Třebop Basin, Pístinná ? Scorpidium C. lasiocarpa, - C. limosa CZ, Třebop Basin, Ponědrážka, W edge of the mire Scorpidium C. lasiocarpa, non-core plot complex between the fish-ponds Švarcenberský C. limosa a Horusický rybník CZ, Třebop Basin, Putim, Řežabinec pond Scorpidium C. lasiocarpa - CZ, Vidnavsko-osoblažská pahorkatina Hills, Vidnava, Calliergon, C. lasiocarpa - Vidnavské mokřiny (Vidnavské loučky) Scorpidium

190

Table S7. Potential glacial and Early Holocene localities of the brown-moss rich quaking fens identified according to macrofossil records from the Czech Republic and Slovakia. C. = Carex; BP = before present; PB = Preboreal; AT = Atlantic; AL = Allerød; BO = Boreal; DR3 = Younger Dryas; SA = Subatlantic

Locality Diagnostic species of Stygio-Caricion Period / time (cal Source / author limosae BP) SK, Borská nížina Lowland, Calliergon trifarium, Scorpidium scorpioides 14020–13360 cal BP Hájková et al. Plavecký Peter, Hanšpile 1 and 2 (2015) SK, Upper-Hron Basin, Telgárt, Scorpidium scorpioides (dominant species) 13160– ? cal BP Hájková P. Meandre Hrona SK, Malé Karpaty Mts, Limbach, Scorpidium scorpioides, Menyanthes PB, 11250 cal BP Hájková P. Nad Šenkárkou trifoliata SK, Liptovská kotlina Basin, Štrba, Scorpidium scorpioides (dominant species, AT, ca 8000–5500 Hájková P. Brezové cca 40 %) cal BP CZ, Nová Bystřice, Blato Calliergon trifarium, Scorpidium scorpioides, AL, DR3, PB Rybníček K. C. limosa CZ, Třebop Basin, Červené Blato Calliergon trifarium, Scorpidium scorpioides, AL, DR3, PB Jankovská V. C. chordorrhiza, Menyanthes trifoliata CZ, Třebop Basin, Nová Ves u Calliergon trifarium, Scorpidium scorpioides PB Rudolph (1917) Třeboně CZ, Ostrava Basin, Ostrava- Scorpidium scorpioides, Cinclidium stygium AT Fejfar et al. (1955) Radvanice CZ, Novohradské hory Mt., Scorpidium scorpioides, Menyanthes DR3, BO Jankovská V. Velanská cesta trifoliata CZ, Křivoklátsko region, Rynholec Calliergon trifarium, Scorpidium scorpioides, PB, BO, ca 10500– Hájková P. Menyanthes trifoliata 9500 cal BP CZ, Upper-Morava river Valley, Calliergon trifarium (dominant species), C. 14340–12710 cal BP Hájková et al. Hnojice lasiocarpa, Carex limosa, Menyanthes (2017) trifoliata CZ, Lower-Morava river Valley, Calliergon trifarium, C. lasiocarpa, C. limosa, SA Rybníčková and Vracov Menyanthes trifoliata Rybníček (1972)

References Adamovid ŽR, Andjus L, Mihajlovid L (1996) Habitat distribution and biogeographical features of the Odonata in the Durmitor range, Montenegro. Notul Odonatol 4: 109–114 Birks HJB, Walters SM (1973) The flora and vegetation of Barno jezero, Durmitor, Montenegro. Glas Republ Zavoda Zašt Prir Prirodnjačkog Muz Titograd 5: 5–23 Březina P, Hadač E, Ježek V, Kubička J (1963) Poznámky o vegetaci třebopských blat. Sborn Pedag Inst v Plzni, Zeměp Přír 4: 207–272 Coldea G (1990) Munții Rodnei. Studiu Geobotanic. Acad. Române, Bucureşti Dahl E (1956) Rondane. Mountain vegetation in south Norway and its relation to the environment. Skr Norske Vidensk-Akad Oslo, Mat-Naturvidensk Kl 3: 1–374 Dítě D, Hájek M, Hájková P (2007) Formal definitions of Slovakian mire plant associations and their application in regional research. Biologia 62: 400–408 Dítě D, Kubandová M (2005) Blatnica močiarna (Scheuchzeria palustris L.) na Kubínskej holi (severné Slovensko). Bull Slov Bot Spoločn 27: 45–48 Dítě D, Pukajová D (2002) Carex limosa L., kriticky ohrozený druh flóry Slovenska. Bull Slov Bot Spoločn 24: 65–73 Fejfar O, Kneblová V, Dohnal Z, Ložek V (1955) Paleontologický a cenologický výzkum kvartérních sedimentů Ostravska v roce 1953. Anthropozoikum 4: 241–284 Hájek M, Háberová I (2001) Scheuchzerio-Caricetea fuscae R.Tx. 1937. In Valachovič M (ed) Rastlinné spoločenstvá Slovenska. 3 Vegetácia mokradí. Veda, Bratislava, pp 185–274 Hájková P, Petr L, Horsák M, Rohovec J, Hájek M (2015) Interstadial inland dune slacks in south-west Slovakia: a multi-proxy vegetation and landscape reconstruction. Quat Internat 357: 314–328 Hájková P, Potůčková A, Hájek M (2017) Zajímavé nálezy vzácných mechorostů ve starých sedimentech 2. Bryonora 59: 67– 71 Lebrun J, Noirfalise A, Heinemann P, Vanden Berghen C (1949) Les Associations végétales de Belgique. Bull Soc R Bot Belg 82: 105–199 Mucina L, Bültmann H, Dierssen K, Theurillat JP, Raus T, Čarni A, Šumberová K, Willner W, Dengler J, García RG, Chytrý M, Hájek M, Di Pietro R, Iakushenko D, Pallas J, Daniëls FJA, Bergmeier E, Guerra AS, Ermakov N, Valachovič M, Schaminée JHJ, Lysenko T, Didukh YP, Pignatti S, Rodwell JS, Capelo j, Weber HE, Solomeshch A, Dimopoulos P,

191

Aguiar C, Hennekens SM, Tichý L (2016) Vegetation of Europe: Hierarchical floristic classification of vascular plant, bryophyte, lichen, and algal communities. Appl Veg Sci 19 (suppl. 1): 3–264 Nordhagen R (1943) Sikilsdalen og Norges fjellbeiter. En plantesosiologisk monografi. Bergens Mus. Skr. 22: 1–607 Peterka T, Hájek M, Jiroušek M, Jiménez-Alfaro B, Aunina L, Bergamini A, Dítě D, Felbaba-Klushyna L, Graf U, Hájková P, Hettenbergerová E, Ivchenko TG, Jansen F, Koroleva NE, Lapshina ED, Lazarevid PM, Moen A, Napreenko MG, Pawlikowski P, Plesková Z, Sekulová L, Smagin VA, Tahvanainen T, Thiele A, Biţa-Nicolae C, Biurrun I, Brisse H, Dušterevska R, De Bie E, Ewald J, FitzPatrick Ú, Font X, Jandt U, Kącki Z, Kuzemko A, Landucci F, Moeslund JE, Pérez- Haase A, Rašomavičius V, Rodwell JS, Schaminée JHJ, Šilc U, Stančid Z, Chytrý M (2017) Formalized classification of European fen vegetation at the alliance level. Appl Veg Sci 20: 124–142 Rudolph K (1917) Untersuchungen über den Aufbau böhmischer Moore I. Aufbau und Entwicklungsgeschichte südböhmischer Hochmoore. Abh Königl Zool Bot Ges Wien 9: 1–116 Rybníček K (1970) Rhynchospora alba (L.) Vahl, its distribution, communities and habitat conditions in Czechoslovakia. Part 2. Folia Geobot 5: 221–263 Rybníček K (1974) Die Vegetation der Moore im südlichen Teil der Böhmisch-Mährischen Höhe. Academia, Praha Rybníček K, Balátová-Tuláčková E, Neuhäusl R (1984) Přehled rostlinných společenstev rašelinišť a mokřadních luk Československa. Stud ČSAV 1984/8: 1–124 Rybníček K, Rybníčková E (1961) Rašeliniště Jihlavských vrchů. Ochr Přír 16: 78–84 Rybníček K, Rybníčková E (1965) Přechodové rašeliniště u Bepadova na Oravě. Biológia 20: 373–375 Rybníčková E, Rybníček K (1972) Erste Ergebnisse paläogeobotanischer Untersuchungen des Moores bei Vracov, Südmähren. Folia Geobot Phytotax 7: 285–308 Štechová T, Manukjanová A, Holá E, Kubešová S, Novotný I, Zmrhalová M (2010) Současný stav populací druhů Helodium blandowii (Thuidiaceae) a Scorpidium scorpioides (Calliergonaceae) v České republice. Bryonora 46: 24–33 van der Maarel E (1979) Transformation of cover-abundance values in phytosociology and its effects on community similarity. Vegetatio 39: 97–114

192

Paper 6 Peterka T., Plesková Z., Palpurina S., Kalníková V., Lazarevid P. & Hájek M. (2016): Meesia triquetra, new relict moss for the Republic of Macedonia. – Herzogia 29: 66–71.

193

Meesia triquetra, a new relict moss for the Republic of Macedonia

Tomáš Peterka1, Zuzana Plesková1, Salza Palpurina1, Veronika Kalníková1, Predrag M. Lazarevid2, Michal Hájek1

1 Department of Botany and Zoology, Faculty of Science, Masaryk Univerzity, Kotlářská 2, 61137 Brno, Czech Republic. 2 Institute for Nature Conservation of Serbia, Dr. Ivana Ribara 91, 11070 Belgrade, Serbia.

Abstract: Meesia triquetra is a circumpolar boreo-arctic moss species typical for well preserved fens, being very rare in southern Europe. During field research on mire vegetation in the Balkan Peninsula, a population of Meesia triquetra was found at the Begovo pole wetland in the Jakupica Mts. It is the first record of this species for the Republic of Macedonia. A description of the locality, ecological conditions and vegetation is presented. The moss grows here in the sedge-moss vegetation similar to temperate fens of the Caricion davallianae alliance. The vegetation with Meesia triquetra in Begovo pole was evaluated in the context of entire fen vegetation variability in the Balkans using the detrended correspondence analysis. The study site appeared to be similar to other Balkan calcium-rich brown-moss fens without calcium carbonate precipitation. This habitat resembles the optimum habitat of the species in central and northern Europe.

Key words: Balkans, bryophytes, fens, plant communities

Introduction Meesia triquetra (family Meesiaceae) is a characteristic species of well-preserved minerotrophic mires (fens) with stable water regime (Berg & Dengler 2005, Pawlikowski 1996, Štechová et al. 2010). The distributional range of the species spans Europe, Caucasus, Siberia, Greenland, northern part of North America and south-eastern Australia (Ochyra et al. 1988). In Europe, the species is mostly confined to the arctic and boreal zone and becomes rare to the south where it is usually restricted to mountain or highland areas, e.g the Alps, the Bohemian Massif, and the Carpathians (Rybníček 1966, Krisai 1985, Nebel & Philippi 2000, Dítě et al. 2010, Šoltés et al. 2015). Although isolated occurrences of the moss can be a result of recent long-distance dispersal by spores, those of Meesia triquetra in temperate and submediterranean areas are usually considered to be of relict origin due to its ecological relation to rich fens (sensu Hájek et al. 2006) which were more common in the landscape during the Early Holocene (Rybníček 1966, Odgaard 1988, Frahm 2012). The moss is absent from north-western (Atlantic) Europe except for an isolated locality in NW Ireland (Odgaard 1988). However, this population as well as a large number of populations in central Europe went extinct due to the deterioration of suitable habitats by human activities (i.e. drainage, fertilization, abandonment) and consequent successional changes in fen communities (Nebel & Philippi 2000, Šoltés 2000, Hájek et al. 2015). A few isolated localities of Meesia triquetra have been reported from the Balkan Peninsula. A reported occurrence at the Vlasina Lake in south-eastern Serbia has not been confirmed for almost 100 years (Randjelovid & Zlatkovid 2010, Papp et al. 2012). Further, a record of Meesia triquetra in the north-central part of Greece was published by Ganiatsas in the 1930s (Preston 1984). The species was reported also from Mt Bolovan in the Stara planina Mts in Bulgaria. However, upon a check, Hájková et al. (2007) found that the herbarium voucher contains another Meesia species, Meesia uliginosa, and the occurrence

194 of Meesia triquetra in Bulgaria is therefore uncertain. According to Sabovljevid et al. (2008), the species has never been reported for the Republic of Macedonia. The aims of this paper are (1) to describe the newly discovered locality of Meesia triquetra in Macedonia with an emphasis on vegetation, (2) to evaluate plant community with Meesia from the investigated locality in the context of entire fen vegetation variability in the Balkans.

Methods During an extensive field research on mire vegetation in the Balkan Peninsula, a population of Meesia triquetra was found in the Begovo pole locality in the central part of Macedonia. A phytosociological relevé was recorded in the patch where Meesia triquetra was present. All vascular plants and bryophytes were recorded within the plot. Their cover was estimated using the nine-grade Braun-Blanquet´s scale (van der Maarel 1979). Coordinates of the relevé were taken using a portable GPS device (WGS-84 system). Water pH and conductivity, both standardized at 20 °C, were measured directly in situ in a shallow depression with water using portable instruments (GMH Greisinger). Nomenclature follows Tutin et al. (1993) for vascular plants and Frey et al. (2006) for bryophytes. In order to evaluate vegetation with Meesia triquetra in a wider context, we subjected all available relevés from fens in the central and southern Balkans (Bulgaria, Greece, Kosovo, Macedonia, Montenegro, Serbia) to detrended correspondence analysis (DCA) using the CANOCO 5 package (Šmilauer & Lepš 2014) and showed the position of the relevé with Meesia triquetra in ordination space. The ordination was based only on vegetation data with cover values of individual species. The analysed relevés (406 plots) were either sampled by the authors of this paper, or taken from the European Mire Vegetation Database (Peterka et al. 2015). Cover values of individual species were square- root transformed and rare species were downweighted according to the protocol of the software.

Locality description The Begovo pole wetland (Fig. 1) is situated on the Mt. Jakupica massif in the central part of the Republic of Macedonia, 5–6 km to the west of the Gorno Jabolčište village, and 2–3 km to the north–northeast of the peak of Solunska glava (2538 m). The altitude is 1950–1970 m asl. The locality is situated in the high-mountain karst plateau surrounded by several peaks reaching more than 2000 m (Micevski et al. 2008). The flat plateau is covered by fine-scale mosaic of shallow water pools, small streams, springs, and stands of fen vegetation. The predominant bedrock of Mt. Jakupica is Precambrian marble (Melovski et al. 2013). Begovo pole serves as an extensive cattle and horse pasture. This kind of management seems to be appropriate to prevent expansion of competitively strong graminoids or dicots which can potentially suppress the populations of fen species, including bryophytes that could be supressed by shading of vascular plant canopy (Bergamini & Pauli 2001).

Results and discussion During sampling, we observed about hundreds to thousands of stems of Meesia triquetra in a few nearby clumps, within ca. 6–8 sq.m. Nevertheless, the population size of the moss can be estimated only with difficultly.

195

The following phytosociological relevé was recorded in the locality. Mt. Jakupica, 2.7 km of the peak of Solunska Glava, coordinates: 41°43'34"N, 21°25'08"E, 16 m2, plane, pH 6.7, conductivity 61 µS/cm, July 25, 2014, T. Peterka & Z. Plesková. – Etotal (80 %), E1 (60 %): Blysmus compressus 4, Carex echinata 2m, Eriophorum latifolium 2m, Carex flava agg. 1, C. nigra 1, Deschampsia cespitosa +, Equisetum palustre +, Epilobium palustre +, Juncus alpinus +, Leontodon cf. hispidus +, Nardus stricta +, Parnassia palustris +, Pinguicula balcanica +, Utricularia minor +, Cardamine sp. r, Trifolium sp. r. – E0 (80 %): Calliergon giganteum 3, Drepanocladus cossonii 3, Climacium dendroides 2a, Meesia triquetra 2a, Drepanocladus exannulatus 1, Philonotis calcarea 1, Bryum pseudotriquetrum +, Hypnum lindbergii +.

Further, Scapania irrigua, Palustriella commutata var. falcata and Campylium stellatum were found outside the relevé. The vegetation can be classified to the Caricion davallianae Klika 1934 alliance due to the dominance of calcium-demanding species (e.g. Blysmus compressus, Drepanocladus cossonii, Eriophorum latifolium, Parnassia palustris) and the total absence of Sphagnum species. However, some elements of high-mountain moderately rich fen communities of the Drepanocladion exannulati Krajina 1933 alliance are also present (Carex echinata, Drepanocladus exannulatus). Detrended correspondence analysis of Balkan fen relevés displayed two principal gradients along the first and second ordination axes (Fig. 2). The first gradient spans from temperate calcareous fens of the Caricion davallianae alliance (with Blysmus compressus, Campylium stellatum, Palustriella commutata) to the alpine fens rich in Balkan endemics (the Narthecion scardici Lakušid 1968 alliance). The second gradient sorted the relevés according to species richness, pH and representation of grassland species, from poor fens (with Sphagnum recurvum s.lat.) to fen grasslands. The relevé with Meesia triquetra did not exhibit marginal position within the ordination space, but was shifted towards the calcareous end of the main gradient (1st axis). The non-marginal position of the relevé with Meesia triquetra in the ordination space suggests that the species´ occurrence is not restricted to extreme environmental conditions at the end of main gradients and could therefore potentially tolerate conditions at some other Balkan fens. What factor can be then responsible for its isolated occurrence in Begovo pole? We can conclude that local history and chance might play a role in its survival, like in survival of many other fen relicts (Hájková et al. 2015). Even in boreal regions, Meesia triquetra is more common in peat profiles than in the present mire vegetation (Rehell & Virtanen 2015). Meesia triquetra has its optimum in rather calcium-rich, brown-moss fens with a stable water regime (Albertson 1949, Rybníček 1966, Pawlikowski 1996, Rehell & Virtanen 2015), but avoids extremely calcareous, strongly tufa-forming fens (Hájek et al. 2005). This ecological preference is reflected also in the position of the analysed relevé in the DCA diagram, close to the Caricion davallianae fens and far from the alpine fens with Balkan endemics, fen meadows with grassland species and the poor fens dominated by peatmosses. In such conditions this species might have been quite common during glacial and early-Holocene times, providing greater opportunities to survive as a relict locally up to present times.

Acknowledgement We are grateful to Petra Hájková who verified the identification of Meesia triquetra and Philonotis calcarea. Jiří Vápa identified Scapania irrigua, Vladimír Řehořek identified Juncus alpinus. The research of European mire vegetation was funded by the Czech Science Foundation (grant number: GB14-36079G) and Masaryk University (MUNI/A/1456/2014).

196

References Albertson, N. 1949. Calliergon sarmentosum och Meesia triquetra i sodra Sverige. Nçgra ord om Mellomsjomyren i Dala. – Svensk Bot. Tidskr. 43: 163–194 (in Swedish). Berg, C. & Dengler, J. 2005. Moose und Flechten als diagnostische Arten von Pflanzengesellschaften–eine Übersicht aus Mecklenburg-Vorpommern. – Herzogia 18: 145–161. Bergamini, A. & Pauli, D. 2001. Effects of increased nutrient supply on bryophytes in montane calcareous fens. – Journal of Bryology 23: 331–339. Dítě, D., Šoltés, R. & Jasík, M. 2010. Opätovne potvrdený výskyt Meesia triquetra v Národnom parku Nízke Tatry v kontexte výskytu druhu na Slovensku. – Bryonora 46: 1–5 (in Slovak). Frahm, J.-P. 2012. The phytogeography of European bryophytes. – Botanica Serbica 36: 23–36. Frey, W, Frahm J.-P., Fischer, E., Lobin, W. 2006. The liverworts, mosses and ferns of Europe. English edition. – Colchester: Harley Books. Hájek, M., Hájková, P., Rybníček, K. & Hekera, P. 2005. Present vegetation of spring fens and its relation to water chemistry. – In: Poulíčková, A., Hájek, M. & Rybníček, K. (eds). Ecology and palaeoecology of spring fens of the West Carpathians. Pp. 69–103. – Olomouc: Palacký University Press. Hájek, M., Horsák, M., Hájková, P. & Dítě, D. 2006. Habitat diversity of central European fens in relation to environmental gradients and an effort to standardise fen terminology in ecological studies. – Perspect. Pl. Ecol. Evol. Syst. 8: 97–114. Hájek, M., Jiroušek, M., Navrátilová, J., Horodyská, E., Peterka, T., Plesková, Z., Navrátil, J., Hájková, P. & Hájek, T. 2015. Changes in the moss layer of Czech fens indicate early succession triggered by nutrient enrichment. – Preslia 87: 279–301. Hájková, P., Horsák, M., Hájek, M., Jankovská, V., Jamrichová, E. & Moutelíková, J. 2015. Using multi-proxy palaeoecology to test a relict status of refugial populations of calcareous-fen species in the Western Carpathians. – Holocene 25: 702–715. Hájková, P., Plášek, V. & Hájek, M. 2007. A contribution to the Bulgarian bryoflora. – Phytologia Balcanica 13: 307–310. Krisai, R. 1985. Zum rezenten und subfossilen Vorkommen subarktischer Moose im salzburgisch/oberösterreichischen Alpenvorland. – Verh. Zool.-Bot. Ges. Österreich 123: 143–150. Melovski, L., Markovski, B., Hristovski, S., Jovanovska, D., Anastasovski, V., Klincharov, S., Velevski, M., Velkovski, N., Trendafilov, A., Matevski, V., Kostadinovski, M., Karadelev, M., Levkov, Z. & Kolchakovski, D. 2013. Regional division of the Republic of Macedonia for the needs of biological databases. – Macedonian Journal of Ecology and Environment 15: 81–111. Micevski, N., Micevski, B. & Bedjanič, M. 2008. Aeshna cyanea and A. juncea, new for the fauna of Macedonia (Odonata: Aeshnidae). – Libellula 27: 267–274. Nebel, M. & Philippi, G. 2000. Die Moose Baden-Württembergs, vol. 2. – Stuttgart: Ulmer. Ochyra, R., Szmajda, F., Bednarek, H. & Bocheoski, W. 1988. Meesia triquetra (Richt.) Aongstr. – In: Tobolewski, Z. & Wojterski, T.(ed.). Atlas rozmieszczenia roślin zarodnikowych w Polsce. Seria V. Mchy (Musci), part 3. Pp. 21–26. – Warsaw (In Polish). Odgaard, B. V. 1988. Glacial relicts and the moss Meesia triquetra in Central and Western Europe. – Lindbergia 14: 73–78. Papp, B., Szurdoki, E. & Sabovljevid, M. 2012. Bryophyte flora of Lake Vlasina and its surroundings (SE Serbia). – Studia botanica hungarica 43: 27–45. Pawlikowski, P. 1996. Habitat preferences and indicator value of eight threatened brown moss species in rich fens of the Lithuanian Lake District (NE Poland). – Polish J. Environm. Stud. 15 (5d): 232–237. Peterka, T., Jiroušek, M., Hájek, M. & Jiménez-Alfaro, B. 2015. European Mire Vegetation Database: a gap- oriented database for European fens and bogs. – Phytocoenologia 45: 291–298. Preston, C.D. 1984. A check-list of Greek mosses. – Journal of Bryology 13: 43–95. Randjelovid, V.N. & Zlatkovid, B.K. 2010. Flora i vegetacija Vlasinske visoravni. – Niš: Univerzitet u Nišu (In Serbian). Rehell, S. & Virtanen, R. 2015. Rich-fen bryophytes in past and recent mire vegetation in a successional land uplift area. – Holocene (doi: 10.1177/0959683615596831). Rybníček, K. 1966. Glacial relics in the bryoflora of the highlands Českomoravská vrchovina (Bohemian- Moravian Highlands); their habitat and cenotaxonomic value. – Folia Geobot. Phytotax. 1: 101–119. Sabovljevid, M., Natcheva, R., Dihoru, G., Tsakiri, E., Dragidevid, S., Erdağ, A. & Papp, B. 2008. Check-list of the mosses of SE Europe. – Phytologia Balcanica 14: 207–244. Šmilauer, P. & Lepš, J.Š. 2014. Multivariate analysis of ecological data using CANOCO 5. – Cambridge University

197

Press: Cambridge. Šoltés, R. 2000. Habitats of rare bryophytes in Slovakia. – Lindbergia 25: 124–127. Šoltés, R., Dítě, D., Mihálik, D., Ondreičková, K., Hrehová, Z., Maximová, N. & Sedláková, B. 2015. Seasonal variation in bryophytes cover in the calcareous mire Belianské lúky, Slovakia. – Pak. J. Bot. 47: 255–262. Štechová, T., Holá, E., Gutzerová, N., Hradílek, Z., Kubešová, S., Lysák, F., Novotný, I. & Peterka, T. 2010. Současný stav lokalit druhů Meesia triquetra a Paludella squarrosa (Meesiaceae) v České republice. – Bryonora 45: 1–11 (In Czech). Tutin, T.G., Burges, N.A., Chater, A.O., Edmondson, J.R., Heywood, V.H., Moore, D.M., Valentine, D.H., Walters, S.M. & Webb, D.A. 1993. Flora Europaea, vol. 1–5. – Cambridge: Cambridge University Press. van der Maarel, E. 1979: Transformation of cover-abundance values in phytosociology and its effects on community similarity. – Vegetatio 39: 97–114.

Fig. 1: The Begovo pole wetland, locality of Meesia triquetra. Photo Z. Plesková.

198

Fig. 2: DCA ordination of available plots of fen vegetation from Balkans. Relevé with Meesia triquetra is indicated by filled black circle. Positions of the species (A) and plots (B) along the first two ordination axes are shown. The eigenvalues of the axes: 1st (DCA1) 0.462 (6.42% of total inertia), 2nd (DCA2) 0.304 (4.23%). Only species with a weight above 10% are shown. AgrCan = Agrostis canina, AlcSpe = Alchemilla sp., AntOdo = Anthoxanthum odoratum, AulPal = Aulacomnium palustre, BlyCom = Blysmus compressus, BriMed = Briza media, BryPse = Bryum pseudotriquetrum, CalCus = Calliergonella cuspidata, CalPal = Caltha palustris, CalStr = Calliergon stramineum, CamSte = Campylium stellatum, CarCur = Carex curta, CarEch = Carex echinata, CarFla = Carex flava agg., CarNig = Carex nigra, CarPan = Carex panicea, CarPra = Cardamine pratensis agg., CarRos = Carex rostrata, CliDen = Climacium dendroides, DacCor = Dactylorhiza cordigera, DesCes = Deschampsia cespitosa, DreExa = Drepanocladus exannulatus, DroRot = Drosera rotundifolia, EpiPal = Epilobium palustre, EquPal = Equisetum palustre, EriAng = Eriophorum angustifolium, EriLat = Eriophorum latifolium, FesRub = Festuca rubra agg., GalPal = Galium palustre, GeuCoc = Geum coccineum, JunArt = Juncus articulatus, JunEff = Juncus effusus, LuzCam = Luzula campestris agg., MolCae = Molinia caerulea/arundinacea, NarStr = Nardus stricta, PalCom = Palustriella commutata, ParPal = Parnassia palustris, PhiFon = Philonotis fontana, PhiSer = Philontis seriata, PinBal = Pinguicula balcanica, PolCom = Polytrichum commune, PotEre = Potentilla erecta, PriFar = Primula farinosa, PruVul = Prunella vulgaris, PseFri = Pseudorchis frivaldii, RanAcr = Ranunculus acris, ScaIrr = Scapania irrigua, SphCon = Sphagnum contortum, SphRec = Sphagnum recurvum s.lat., SphSub = Sphagnum subsecundum, SphTer = Sphagnum teres, SucPra = Succisa pratensis, TarAlp = Taraxacum sect. Alpina.

199

Paper 7 Peterka T., Kalníková V. & Plesková Z. (2017): Pseudocalliergon lycopodioides, a new bryophyte species for Montenegro. – Herzogia 30: 496–500.

200

Pseudocalliergon lycopodioides, a new bryophyte species for Montenegro

Tomáš Peterka, Veronika Kalníková, Zuzana Plesková

Department of Botany and Zoology, Faculty of Science, Masaryk University, Kotlářská 2, 61137 Brno, Czech Republic

Abstract: Pseudocalliergon lycopodioides is a European boreo-temperate species confined to extremely rich fens and other calcareous wetlands. The species belongs to endangered wetland biota; many localities across Europe have recently disappeared. Pseudocalliergon lycopodioides has been exceptionally documented in the Balkan Peninsula. Here we report the species as new for Montenegro. The moss was found on the plateau Jezerska Površ at the eastern edge of the Durmitor Mts. Pseudocalliergon lycopodioides grows here in moss- sedge vegetation of the Caricion davallianae alliance.

Key words: Balkans, Dinarids, endangered species, fens, moss, vegetation, wetland

Introduction Pseudocalliergon lycopodioides (Brid.) Hedenäs, syn. Drepanocladus lycopodioides (Brid.) Warnst., Scorpidium lycopodioides (Brid.) H.K.G. Paul, is a pleurocarpous moss of the Amblystegiaceae family. It grows in semi-aquatic mostly calcareous habitats, such as extremely rich fens, pools and wet depressions on limestone areas, lake shores and dune slacks (Nebel & Philippi 2000, Hedenäs 2003, Atherton et al. 2010). According to Smith (2004), its distributional range is characterized as “European Boreo- temperate”. Pseudocalliergon lycopodioides is distributed in Iceland, southern Scandinavia, Finnish Lapland, Denmark, the Baltic states, Ukraine, western and central Europe. The moss rarely occurs also in Western Siberia, where it reaches its eastern distribution limit (Hedenäs 2003). Pseudocalliergon lycopodioides has been rarely documented from the Mediterranean belt including the Balkan Peninsula (cf. Ros et al. 2013). Within the Balkan Peninsula, Sabovljevid et al. (2008) and Hodgetts (2015) report its occurrence only for Serbia and Slovenia. In Serbia, the moss was found at the Jankove Bare locality in the Kopaonik Mts (Papp et al. 2004). Pseudocalliergon lycopodioides has recently decreased in many parts of Europe (Hedenäs & Bisang 2015). The species is hence frequently included in national Red-lists of individual countries, especially of western and central Europe; for example as “regionally extinct” in Czech Republic (Kučera et al. 2012), “critically endangered” in Norway (http://www.artsdatabanken.no), “endangered” in Hungary (Papp et al. 2010) or “vulnerable” in Slovenia (Martinčič 2016). During the extensive field research of fen vegetation in the Balkan Peninsula, the population of Pseudocalliergon lycopodioides was discovered in the plateau Jezerska Površ in the massif of the Durmitor Mts. (the Dinarids) in Montenegro. The aim of this short study is to present details about this finding and characterize the new locality.

Methods A phytosociological relevé (vegetation plot) was gathered in the locality following the Braun- Blanquet approach. All vascular plants and bryophytes were recorded within the plot; the species covers were estimated using the nine-grade Braun-Blanquet´s scale (van der Maarel 1979). Coordinates of the relevé were obtained using a portable GPS device (WGS-84

201 system). Water pH and conductivity, both standardized at 20 °C, were measured directly in situ in a shallow depression saturated with water using portable instruments (GMH Greisinger). Nomenclature follows Tutin et al. (1968–1993) for vascular plants, Hill et al. (2006) for bryophytes and Mucina et al. (2016) for vegetation units.

Results and Discussion

Locality The newly discovered locality is situated on the plateau Jezerka Površ at the eastern edge of the Durmitor Mts. The bedrock is composed of limestones and dolomites covered by glacial deposits (Hughes et al. 2011). The area is characterised by broad linear moraine crests alternating with terrain depressions in which several glacial lakes had developed (e.g. Riblje jezero lake, Vražje jezero lake, Ševarita lokva). Pseudocalliergon lycopodioides was discovered in a large flat terrain depression (Fig. 1) with stagnant water in some places.

The following phytosociological relevé was obtained in the locality: Jezerska Površ, 6 km to the south-east of Žabljak, 1.4 km to the north-east of Riblje jezero lake, 1400 m a.s.l., coordinates: 43°06'17"N, 19°09'35"E; 16 m2; pH: 7.0, conductivity: 266 µS/cm; July 31, 2014; T. Peterka, V. Kalníková, S. Palpurina & Z. Plesková. Etotal (95 %), E1 (80 %): Molinia caerulea agg. 3, Sesleria caerulea 2a, Carex davalliana 2m, Carex hostiana 1, Carex panicea 1, Eriophorum latifolium 1, Potentilla erecta 1, Blysmus compressus +, Equisetum palustre +, Juncus articulatus +, Parnassia palustris +, Salix rosmarinifolia +, Sanguisorba officinalis +, Succisa pratensis +, Lotus corniculatus r. – E0 (85 %): Campylium stellatum 5, Scorpidium revolvens agg. 2a, Pseudocalliergon trifarium 1, Bryum pseudotriquetrum +, Fissidens adianthoides +, Palustriella commutata +, Pseudocalliergon lycopodioides +.

In the recent attempt to create a harmonized classification system of European fens (Peterka et al. 2017), the relevé met the assumptions of formal definition of the Caricion davallianae alliance comprising temperate calcareous and extremely-rich fens. This community is characterized by the presence of calcicole species (e.g. Campylium stellatum, Carex davalliana, C. hostiana, Eriophorum latifolium, Palustriella commutata, Parnassia palustris) and absence of Sphagnum species. The occurrence of Molinia caerulea agg., Salix rosmarinifolia, Sanguisorba officinalis and Succisa pratensis in the analysed plot nevertheless suggests a certain syntaxonomical relationship to intermittently wet meadows of the Molinion caeruleae alliance.

Conservation remarks Besides other general threats for fen ecosystems (modification of water regime, pollution, cultivation for agricultural purposes; Janssen 2016), the studied locality can be potentially negatively affected by the expansion of Molinia caerulea agg. This competitively-strong grass effectively utilizes nutrients, produces biomass and hard-to-decompose litter and can thus gradually transform fen habitats into monodominant species-poor stands almost without bryophyte layer (Hájková et al. 2009). The locality acts as occasional sheep or cattle pasture, which might be suitable management to prevent the undesirable successional changes towards Molinia-dominated stands (Stammel et al. 2003). However we have no detailed information about frequency and intensity of grazing to assess properly its positive or negative impact on plant communities.

202

During the brief excursion in the Durmitor Mts., we observed several sites resembling the investigated locality of Pseudocalliergon lycopodioides in terms of stand physiognomy and presupposed habitat conditions. Therefore, more populations of the target species might be discovered in the plateau Jezerska Površ as well as in other parts of the Durmitor Mts. In spite of this fact, the moss should be included in the national red list.

Acknowledgements We would like to thank Petra Hájková who verified the identification of Pseudocalliergon lycopodioides. Salza Palpurina helped us during the fieldwork. Michal Hájek provided useful comments to the preliminary version of the text. Two anonymous referees helped us to improve the manuscript. The research was supported by the Czech Science Foundation (PLADIAS, grant number: GB14-36079G) and Masaryk University (MUNI/A/1301/2016).

References Atherton, I., Bosanquet, S. & Lawley, M. (eds) 2010. Mosses and Liverworts of Britain and Ireland – a field guide. – British Bryological Society. Hájková, P., Hájek, M. & Kintrová, K. 2009. How can we effectively restore species richness and natural composition of a Molinia-invaded fen? – Journal of Applied Ecology 46: 417–425. Hedenäs, L. 2003. The European species of the Calliergon-Scorpidium-Drepanocladus complex, including some related or similar species. – Meylania 28: 1–116. Hedenäs, L. & Bisang, I. 2015. Infraspecific diversity in a spore-dispersed species with limited distribution range. – Systematics and Biodiversity 13: 17–27. Hill, M. O., Bell, N., Bruggeman-Nannenga, M. A., Brugués, M., Cano, M. J., Enroth, J., Flatberg, K. I., Frahm, J.- P., Gallego, M. T., Garilleti, R., Guerra, J., Hedenäs, L., Holyoak, D. T., Hyvönen, J., Ignatov, M. S., Lara, F., Mazimpaka, V., Muqoz, J. & Guerra, J. 2006. An annotated checklist of the mosses of Europe and Macaronesia. – Journal of Bryology 28: 198–267. Hodgetts, N. G. 2015. Checklist and country status of European bryophytes – towards a new Red List for Europe. – Irish Wildlife Manuals, No. 84. National Parks and Wildlife Service, Department of Arts, Heritage and the Gaeltacht, Ireland. Hughes, P. D., Woodward, J. C., van Calsteren, P. C. & Thomas, L. E. 2011. The glacial history of the Dinaric Alps, Montenegro. – Quaternary Science Reviews 30: 3393–3412. Janssen, J. A. M., Rodwell, J. S., García Criado, M., Gubbay, S., Haynes, T., Nieto, A., Sanders, N., Landucci, F., Loidi, J., Ssymank, A., Tahvanainen, T., Valderrabano, M., Acosta, A., Aronsson, M., Arts, G., Attorre, F., Bergmeier, E., Bijlsma, R.-J., Bioret, F., Biţæ-Nicolae, C., Biurrun, I., Calix, M., Capelo, J., Čarni, A., Chytrý, M., Dengler, J., Dimopoulos, P., Essl, F., Gardfjell, H., Gigante, D., Giusso del Galdo, G., Hájek, M., Jansen, F., Jansen, J., Kapfer, J., Mickolajczak, A., Molina, J. A., Molnár, Z., Paternoster, D., Piernik, A., Poulin, B., Renaux, B., Schaminée, J. H. J., Šumberová, K., Toivonen, H., Tonteri, T., Tsiripidis, I., Tzonev, R. & Valachovič, M. 2016. European Red List of Habitats - Part 2. Terrestrial and freshwater habitats. – Luxembourg: Publications Office of the European Union. Kučera, J., Vápa, J. & Hradílek, Z. 2012. Bryophyte flora of the Czech Republic: updated checklist and Red List and a brief analysis. – Preslia 84: 813–850. Martinčič, A. 2016. Updated Red list of bryophytes of Slovenia. – Hacquetia 15: 107–126. Mucina, L., Bültmann, H., Dierßen, K., Theurillat, J.-P., Raus, T., Čarni, A., Šumberová, K., Willner, W., Dengler, J., Gavilán García, R., Chytrý, M., Hájek, M., Di Pietro, R., Iakushenko, D., Pallas, J., Daniëls, F. J. A., Bergmeier, E., Santos Guerra, A., Ermakov, N., Valachovič, M., Schaminée, J. H. J., Lysenko, T., Didukh, Ya. P., Pignatti, S., Rodwell, J. S., Capelo, J., Weber, H. E., Solomeshch, A., Dimopoulos, P., Aguiar, C., Hennekens, S. M. & Tichý, L. 2016. Vegetation of Europe: Hierarchical floristic classification system of vascular plant, bryophyte, lichen, and algal communities. – Applied Vegetation Science 19, Suppl. 1: 3– 264. Nebel, M. & Philippi, G. 2000. Die Moose Baden-Württembergs, Vol. 2. – Stuttgart: Ulmer. Papp, B., Erzberger, P. & Sabovljevid, M. 2004. Contributions to the bryophyte flora of Kopaonik Mts (Serbia, Serbia-Montenegro). – Studia botanica hungarica 35: 67–79. Papp, B., Erzberger, P., Ódor, P., Hock, Zs., Szövényi, P., Szurdoki, E. & Tóth, Z. 2010. Updated checklist and red list of Hungarian bryophytes. – Studia botanica hungarica 41: 31–59. Peterka, T., Hájek, M., Jiroušek, M., Jiménez-Alfaro, B., Aunina, L., Bergamini, A., Dítě, D., Felbaba-Klushyna, L.,

203

Graf, U., Hájková, P., Hettenbergerová, E., Ivchenko, T. G., Jansen, F., Koroleva, N. E., Lapshina, E. D., Lazarevid, P. M., Moen, A., Napreenko, M. G., Pawlikowski, P., Plesková, Z., Sekulová, L., Smagin, V. A., Tahvanainen, T., Thiele, A., Biţæ-Nicolae, C., Biurrun, I., Brisse, H., Dušterevska, R., De Bie, E., Ewald, J., FitzPatrick, Ú., Font, X., Jandt, U., Kącki, Z., Kuzemko, A., Landucci, F., Moeslund, J. E., Pérez-Haase, A., Rašomavičius, V., Rodwell, J. S., Schaminée, J. H. J., Šilc, U., Stančid, Z. & Chytrý, M. 2017. Formalized classification of European fen vegetation at the alliance level. – Applied Vegetation Science 20: 124–142. Ros, R. M., Mazimpaka, V., Abou-Salama, U., Aleffi, M., Blockeel, T.L., Brugués, M., Cros, R. M., Dia, M. G., Dirkse, G. M., Draper, I., El-Saadawi, W., Erdağ, A., Ganeva, A., Gabriel, R., González-Mancebo, J. M., Granger, C., Herrnstadt, I., Hugonnot, V., Khalil, K., Kürschner, H., Losada-Lima, A., Luís, L., Mifsud, S., Privitera, M., Puglisi, M., Sabovljevid, M., Sérgio, C., Shabbara, H. M., Sim-Sim, M., Sotiaux, A., Tacchi, R., Vanderpoorten, A. & Werner, O. 2013. Mosses of the Mediterranean, an annotated checklist. – Cryptogamie, Bryologie 34: 99–283. Sabovljevid, M., Natcheva, R., Dihoru, G., Tsakiri, E., Dragidevid, S., Erdağ, A. & Papp, B. 2008. Check-list of the mosses of SE Europe. – Phytologia Balcanica 14: 207–244. Smith, A. J. E. 2004. The Moss Flora of Britain and Ireland. – Cambridge: Cambridge University Press. Stammel, B., Kiehl, K. & Pfadenhauer, J. 2003. Alternative management on fens: Response of vegetation to grazing and mowing. – Applied Vegetation Science 6: 245–254. Tutin, T. G., Burges, N. A., Chater, A. O., Edmondson, J. R., Heywood, V. H., Moore, D. M., Valentine, D. H., Walters, S. M. & Webb, D. A. 1968–1993. Flora Europaea, Vol. 2–5, 2nd ed. of Vol. 1. – Cambridge: Cambridge University Press. van der Maarel, E. 1979. Transformation of cover-abundance values in phytosociology and its effects on community similarity. – Vegetatio 39: 97–114.

Fig. 1: Habitat of Pseudocalliergon lycopodioides. Photo: V. Kalníková.

204

Paper 8 Peterka T., Dítě D., Hájková P. & Hájek M. (2016): Ověření výskytu suchopýru štíhlého (Eriophorum gracile) ve Žďárských vrších. – Východočeský sborník přírodovědný, Práce a studie 23: 47–56.

205

Vč. sb. přír. – Práce a studie, 23 (2016): 47–56 ISSN 1212-1460

OVĚŘENÍ VÝSKYTU SUCHOPÝRU ŠTÍHLÉHO (ERIOPHORUM GRACILE) VE ŽĎÁRSKÝCH VRŠÍCH

Confirmation of the occurence ofEriophorum gracile in the Žďárské vrchy Hills

Tomáš PETERKA1, Daniel DÍTĚ2, Petra HÁJKOVÁ1,3, Michal HÁJEK1

1 Ústav botaniky a zoologie, Přírodovědecká fakulta, Masarykova Univerzita, Kotlářská 2, CZ-611 37 Brno; e-mail: [email protected], [email protected], [email protected] 2 Botanický ústav SAV, Dúbravská cesta 9, SK-845 23 Bratislava; e-mail: [email protected], 3 Oddělení vegetační ekologie, Botanický ústav AV ČR, Lidická 25/27, CZ-602 00 Brno

Na rašelinné louce nedaleko rybníka Zlámanec u obce Vortová ve Žďárských vrších byl recentně ověřen výskyt kriticky ohroženého suchopýru štíhlého (Eriophorum gracile). Druh zde roste ve společenstvech asociace Agrostio caninae-Caricetum diandrae. Fytocenologické snímky ze Zlámance byly porovnány s dalšími dostup- nými vegetačními zápisy ze střední a severní Evropy.

Klíčová slova: Cyperaceae, Českomoravská vrchovina, ohrožené druhy, rašeliniště Keywords: Cyperaceae, the Bohemian-Moravian Highlands, endangered species, mires

Úvod Suchopýr štíhlý (Eriophorum gracile) má nesouvislý cirkumpolární areál, který zahrnuje Evropu (kromě nejjižnějších oblastí, Islandu a arktických ostrovů), západní a střední Sibiř, Kamčatku a severní část Severní Ameriky (Hultén & Fries 1986, Pro- cházka 1999). V Evropě je poněkud hojnější pouze v boreální zóně, jinde se vyskytuje jen velmi roztroušeně. Suchopýr štíhlý patří podle červeného seznamu mezi kriticky ohrožené taxony naší flóry (Grulich 2012). Do stejné kategorie patří i ve smyslu Vyhlášky č. 395/92 Sb. k zákonu č. 114/92 Sb. o ochraně přírody a krajiny. Druh provází málo narušená minerot- rofní rašeliniště se stabilním vodním režimem. V minulosti se častěji vyskytoval pouze na Třeboňsku, Českomoravské vrchovině a v severozápadních Čechách; další lokality byly zaznamenány na Šluknovsku, Dokesku, Blatensku, v Českém ráji, v Jesenickém podhůří, u Vidnavy a Úvalna ve Slezsku a u Popůvek nedaleko Brna (Balátová-Tuláčková 1965, Procházka 1999, Kaplan et al. 2015). Staré údaje z Hrubého Jeseníku se považují za sporné (Bureš 2013). V současnosti Eriophorum gracile přežívá s jistotou v Třeboňské pánvi na lokalitách Ruda u Horusic a Hovízna u Ponědrážky (Navrátilová & Navrátil 2005a,b, Grulich 2013). Další recentní naleziště se nachází v na Šumavských pláních

47 u říčky Křemelná (Řepka & Lustyk 1998, Ekrt in Hadinec & Lustyk 2011) a u obce Srní (Grulich in Hadinec & Lustyk 2015). Nevelká populace byla nedávno objevena také na Táborsku u obce Ratibořské Hory (Koptík in Lepší & Lepší 2010). Kratochví- lová (2010) ověřila známou lokalitu v Podtroseckém údolí u rybníka Vidlák v Českém ráji (cf. Balátová-Tuláčková 1968, Faltys & Faltysová 1997). Současné i historické nálezy suchopýru štíhlého na Českomoravské vrchovině pochází převážně z širšího okolí Jihlavy a ze Žďárských vrchů. Na Jihlavsku byl druh v minulosti zaznamenán na lokalitách Praskolesy u Telče, Nad Svitákem a Na Oklice u Milíčova a na Vílaneckém rašeliništi (Rybníček & Rybníčková 1966, Růžička 1987, Růžička et al. 1998, Čech et al. 2002, Kaplan et al. 2015). Z Vílaneckého rašeliniště je suchopýr doložen aktuálními sběry Ester Ekrtové a Jiřího Juřičky z let 2009 a 2012, které se nachází v herbáři Muzea Vysočiny v Jihlavě (MJ). Suchopýr dosud roztroušeně roste i na rašeliništi Na Oklice (E. Ekrtová in litt.) a recentní výskyt nelze vyloučit ani na lokalitě Nad Svitákem (J. Juřička in litt.). Údaje o rozšíření Eriophorum gracile ve Žďárských vrších a jejich blízkém okolí shrnuli ve svém článku Bureš & Řepka (1989). V té době byla z území doložena jen tři historická naleziště a suchopýr štíhlý zde proto koncem 80. let 20. století patřil mezi nezvěstné druhy. Lokalita u Kubovského rybníka u Nového Města na Moravě zanikla kvůli těžbě rašeliny již ve 40. letech (Balátová-Tuláčková 1965). Rašeliniště u Pilského rybníka u Žďáru nad Sázavou bylo pro změnu zničeno zřízením vodní nádrže Pilská asi o deset let později. Rašelinná louka u Rváčova u Hlinska, kde suchopýr hojně rostl ještě v 70. letech, padla za oběť melioracím (Procházka 1999). Během floristického kurzu v Hlinsku v roce 1990 však Vladimír Skalický objevil do té doby neznámou lokalitu u rybníka Zlámanec nedaleko Vortové (Bureš 1993). Z téhož floristického kurzu prav- děpodobně pochází i herbářový doklad Čestmíra Deyla datovaný 8. 7. 1990 a uložený v herbáři Vlastivědného muzea v Olomouci (OLM). O devět let později Eriophorum gracile u Zlámance ověřil Aleš Hájek (doklad v HR; Kaplan et al. 2015). Výskyt sucho- pýru štíhlého u Zlámance je uveden rovněž v kompendiu Chráněná území ČR (Čech et al. 2002) a v inventarizačním průzkumu PP Zlámanec (Bureš 2004). První z autorů tohoto příspěvku zde po suchopýru opakovaně bezúspěšně pátral při studiu rašelinné vegetace v letech 2011 a 2012. Ověřen byl až druhým z autorů při společné exkurzi v květnu 2013. Tento článek má následující cíle: (1) blíže charakteri- zovat jedinou recentně známou lokalitu Eriophorum gracile ve Ždárských vrších a (2) porovnat zdejší vegetační a stanovištní poměry s dalšími lokalitami suchopýru v ČR i jinde v Evropě.

Metodika Na lokalitě jsme zapsali fytocenologické snímky metodou curyšsko-montpellierské školy. Pokryvnost jednotlivých rostlinných druhů vyjadřuje devítičlenná Braun-Blanque- tova stupnice (van der Maarel 1979). Hodnoty pH a konduktivity vody (standardi- zované na 20 °C) byly změřeny přímo v terénu pomocí přenosných měřících přístrojů Greisinger. Konduktivita (vodivost) vyjadřuje kvantitativní zastoupení iontů ve vodě a tedy i míru trofie ve smyslu celkové koncentrace minerálů. Fytocenologické snímky ze Zlámance byly porovnány s dalšími dostupnými vegetačními zápisy se suchopýrem štíhlým. Kromě snímků z České národní fytoce- nologické databáze (ČNFD; Chytrý & Rafajová 2003) jsme pro srovnání použili (1) publikované snímky z boreální zóny Evropy, jmenovitě ze Švédska (Almquist 1929,

48 Elveland 1976), Estonska (Paasio 1939), Finska a severozápadní části Ruska (Warén 1926, Ruuhijärvi 1960, Lounamaa 1961), (2) snímek z bulharského pohoří Rodopy (Hájek et al. 2008), (3) nepublikované zápisy Kamila Rybníčka z Oravské kotliny od obce Trstená (2 sn.), (4) nepublikovaná data autorů této studie ze Slovenska (5 sn., Zvo- lenská kotlina, PR Jelšovec), severovýchodního Polska (3 sn.), Švédska (1 sn., oblast Norrland) a Švýcarska (1 sn., Mt. Camignolo). Dostupné zápisy z území ČR pochází z rašelinišť Ruda a Hovízna na Třeboňsku (16 sn. včetně nepublikovaných dat, cf. Chytrý & Rafajová 2003), z lokality Vidlák v Českém ráji (5 sn.), ze Šumavy (1 sn., Grulich in Hadinec & Lustyk 2015), Táborska (1 sn., Koptík in Lepší & Lepší 2010), Českomoravské vrchoviny (2 sn., Rybníček 1974) a Vidnavských luk (1 sn., Vicherek 1958). K hledání hlavních směrů floristické variability v datovém souboru jsme použili dentrendovanou korespondenční analýzu (DCA). Před vlastní analýzou byly problema- tické taxony sloučeny do druhových komplexů. Pokryvnosti jednotlivých druhů jsme převedli na procenta (střední hodnota daného stupně) a odmocninově transformovali. Výpočet DCA i konstrukce ordinačního diagramu probíhaly v programu CANOCO 5 (Šmilauer & Lepš 2014). Nomenklatura cévnatých rostlin je sjednocena podle Seznamu cévnatých rostlin ČR (Danihelka et al. (2012), nomenklatura mechorostů podle Seznamu a červeného seznamu mechorostů ČR (Kučera & Váňa (2005). V ostatní případech je za jménem druhu uveden autor. Názvy syntaxonů odpovídají monografii Vegetace ČR (Hájek & Hájková 2011).

Charakteristika lokality a vegetace Lokalita se nachází asi 0,8–0,9 km jižně od kapličky v obci Vortová, západně od hráze rybníka Zlámanec v nadmořské výšce 620 m. Dle regionálně fytogeografického členění ČR (Skalický 1988) území náleží do fytochorionu 91. Žďárské vrchy. Jedná se o poměrně zachovalý fragment lučního rašeliniště a navazujících vlhkých pcháčo- vých luk a smilkových trávníků o celkové rozloze asi 1 ha. Eriophorum gracile roste v jihovýchodní části rašelinné enklávy (obr. 1) na zvodnělých plochách sycených vodou z pramenů. V květnu a srpnu 2013 a v červenci 2014 jsme zde zaznamenali patnáct až dvacet plodných rostlin roztroušených na ploše cca 500 m2. Vzhledem k nenápadnosti druhu, který prakticky nelze určit ve sterilním stavu, a zachovalému stavu společenstev však lze předpokládat, že zdejší populace je ve skutečnosti početnější. Vegetaci s výsky- tem suchopýru štíhlého dokumentují následující fytocenologické snímky.

Sn. 1. WGS-84: 49°42′18′′ N, 15°55′56′′ E ± 5 m, 16 m2, rovina, pH 6,4, konduktivita

127,5 µS/cm, 24. 5. 2013 D. Dítě, M. Hájek, P. Hájková & T. Peterka. – Ecelk. (35 %), E1 (30 %): Menyanthes trifoliata 2a, Carex diandra 2m, Carex panicea 1, Carex demissa +, Carex lasiocarpa +, Carex nigra +, Carex rostrata +, Comarum palustre +, Drosera rotundifolia +, Eriophorum gracile +, Juncus bulbosus +, Valeriana dioica +, Viola palustris +, Alnus glutinosa juv. r, Cirsium palustre r, Salix cinerea juv. r. – E0 (15 %): Calliergon giganteum 1, Warnstorfia exannulata 1, Aulacomnium palustre +, Bryum pseudotriquetrum +, Calliergonella cuspidata +, Campylium stellatum +, Sphagnum flexuosum +, Sphagnum subsecundum +, Sphagnum teres +, Straminergon stramineum +.

Sn. 2. WGS-84: 49°42′17′′ N, 15°55′56′′ E ± 4 m, 16 m2, rovina, pH 5,8, konduktivita

76 µS/cm, 27. 8. 2013 T. Peterka. – Ecelk. (75 %), E1 (65 %): Carex rostrata 2b, Carex

49 lasiocarpa 2a, Carex panicea 2a, Comarum palustre 2a, Eriophorum angustifolium 2a, Menyanthes trifoliata 2a, Viola palustris 2a, Carex diandra 2m, Carex nigra 1, Juncus bulbosus 1, Agrostis canina +, Carex dioica +, Carex echinata +, Epilobium palustre +, Equisetum fluviatile +, Equisetum palustre +, Eriophorum gracile +, Galium palustre agg. +, Juncus alpinoarticulatus +, Peucedanum palustre +, Valeriana dioica +. – E0 (35 %): Sphagnum flexuosum 2a, Sphagnum subsecundum 2a, Sphagnum teres 2a, Warnstorfia exannulata 1, Calliergonella cuspidata +, Straminergon stramineum +.

Vegetace odpovídá asociaci Agrostio caninae-Caricetum diandrae (svaz Caricion canescenti-nigrae). Ve Žďárských vrších a přilehlých územích se tato asociace v sou- časnosti vyskytuje jen velmi vzácně. Kromě Zlámance se její porosty dále nacházejí u Dolního Ratajského rybníka v PP Ratajské rybníky u Hlinska, v litorálu Nového rybníka u Rohozné severně od Trhové Kamenice (Jirásek 1998, Štechová & Štech 2009, Peterka 2013) a ve fragmentech i v PP Šafranice u Veselíčka (Peterka ined.). Degradovaný porost této asociace byl zaznamenán také na jihozápadním břehu Kamen- ného rybníka u Polničky v PR Pod Kamenným vrchem (Juřička et al. 2013). Rybník Zlámanec i přilehlé rašeliniště jsou chráněny jako přírodní památka. Rašeli- niště je pravidelně sečeno od roku 2003. Současný management se zdá zcela dostačující, je však nutné v něm dále pokračovat. Z dalších ochranářsky významných druhů se zde

Obr. 1: Porosty asociace Agrostio caninae-Caricetum diandrae v přírodní památce Zlámanec. Fig. 1: Stands of the Agrostio caninae-Caricetum diandrae association in the nature monument Zlámanec.

50 vyskytuje např. Trichophorum alpinum a Utricularia minor (Čech et al. 2002), z mechů Hamatocaulis vernicosus a Paludella squarrosa (Štechová et al. 2014). Naposledy byl suchopýr na lokalitě ověřen v roce 2015 pracovníky AOKP ČR a pracovníkem Muzea Vysočiny v Jihlavě (K. Juřičková in litt).

Srovnání s jinými lokalitami Ačkoliv výsledky DCA ovlivňuje řada okolností (počet snímků, vlastnosti souboru apod.), ukázala ordinační analýza (obr. 2) některé ne zcela nezajímavé skutečnosti. Podél první osy, která vyjadřuje hlavní směr variability v datech, se oddělily snímky ze střední Evropy, Alp a Bulharska (levá část diagramu) od snímků ze Švédska, Pobaltí, Finska a severozápadního Ruska (pravá část diagramu). Pravou část diagramu charakterizují boreální a boreálně-arktické druhy, které jinde v Evropě zcela chybí, např. Carex livida (Wahlenb.) Willd. či Drepanocladus tundrae (Arnell) Loeske, nebo se zde vyskytují jen velmi roztroušeně (Carex chordorrhiza, Trichophorum cespitosum). V levé části diagramu se kromě druhů s geografickou vazbou na střední Evropu (Carex davalliana, Valeriana dioica) objevují i mechorosty a cévnaté rostliny typické pro vlhké pcháčové louky či jiné typy produktivní mokřadní vegetace (Caliergonella cuspidata, Caltha palustris, Galium uliginosum, Lythrum salicaria). Jejich přítomnost ukazuje na vyšší dostupnost hlavních živin, která může souviset s narušením vodního režimu lokalit, absencí exportu živin při kosení a dnešní celkovou eutrofizací středoevropské krajiny (viz např. Bollens et al. 1998, Koch & Jurasinski 2014, Hájek et al. 2015). První osa tedy naznačuje i určitou míru narušení rašelinišť v temperátní Evropě. Přítomnost lučních druhů ve snímcích z temperátní Evropy však můžeme interpretovat i v tom smyslu, že druh přežil v těchto oblastech jen na „náhradních“, člověkem vytvořených nebo udržovaných rašelinných loukách, zatímco jeho původní biotopy představující primární bezlesí už zanikly. Z našich lokalit se boreálním rašeliništím svým druhovým složením nejvíce podobají rašeliniště na Třeboňsku, Šumavě a na Zlámanci. Druhou ordinační osu lze interpretovat jako komplexní gradient pH a koncentrace minerálů (zejména vápníku) v prostředí. Zatímco v horní části diagramu najdeme kal- cikolní druhy (Campylium stellatum, Eriophorum latifolium) a typické prvky minerálně bohatých slatinišť s kalcitolerantními rašeliníky svazu Sphagno warnstofii-Tomentypnion nitentis (Paludella squarrosa, Sphagnum warnstorfii), v dolní části se objevují acido- fyty (Scheuchzeria palustris, Sphagnum riparium, Warnstorfia fluitans). Z ordinačního diagramu je proto nepřímo patrné, že suchopýr štíhlý v rámci svého evropského areálu toleruje relativně široký úsek gradientu pH a vápnitosti podobně jako jiné rašeliništní druhy cévnatých rostlin (např. Carex lasiocarpa, C. nigra, C. rostrata, Eriophorum angustifolium, Menyanthes trifoliata). Tuto skutečnost podporují i hodnoty pH a konduktivity vody zjištěné na našem území i v jiných oblastech Evropy (tab. 1). Výskyt Eriophorum gracile na určité lokalitě tedy zřejmě ovlivňují jiné faktory. V případě rašeliniště u Vortové může jít např. o stabilní vodní režim, pravidelný management, který omezuje konkurenčně silné cévnaté rostliny či mechorosty, a v neposlední řadě také o kontinuum rašelinných biotopů v území během holocénu (cf. Rybníčková & Rybníček 1988). Recentní výskyt suchopýru štíhlého na Zlámanci zdůrazňuje nejen regionální hod- notu této jedinečné lokality, ale také význam Žďárských vrchů pro ochranu rašelinišť a jejich diverzity v rámci celé České republiky.

- - - ======agg., Vac = GalUli EriGra AulPal CxRos CalGig = EquFlu = WarExa CamSte Comarum Sphagnum S. centrale ), = Carex flava Carex = = Straminergon VacOxy VacOxy Trichophorum Trichophorum Carex limosa , Drepanocladus = Carex davalliana , = Potentilla erecta , Scorpidium revol CxFla Lysimachia vulgaris , Lysimachia ComPal SphCon Molinia caerulea Molinia Sphagnum subsecun Sphagnum TriAlp = TriAlp StrStr = = CxLim Carex panicea , Carex = Juncus alpinoarticulatus , = CxDav = = Warnstorfia sarmentosa. Warnstorfia PotEre = ScoRev = Andromeda polifolia , Drosera rotundifolia , LysVul = LysVul Galium palustre agg., Veronica scutellata , Veronica = MolCae CxPan Calliergon richardsonii , Caliergonella cuspidata , Caliergonella JunAlp SphSub Eriophorum angustifolium , = = Carex echinata , WarSar = WarSar Utricularia minor agg., Sphagnum palustre (incl. AndPol DroRot = GalPal = VerScu = VerScu Cinclidium stygium , = CalRic = Carex lasiocarpa , EriAng CalCus = Scorpidium scorpioides , Scorpidium Tomentypnum nitens , Tomentypnum Sphagnum recurvum agg. ( S. angustifolium , CxEch = Carex nigra , Carex Carex chordorrhiza , Sphagnum warnstorfii , = = SphPal = Phragmites australis , CinSty Scheuchzeria palustris , Lysimachia thyrsiflora , Menyanthes trifoliata , Menyanthes CxCho = TomNit = TomNit Alnus glutinosa , Sphagnum riparium , Sphagnum = SphRec = Drosera anglica , = Warnstorfia fluitans , Warnstorfia = Betula nana , SphWar = SphWar Valeriana dioica , Valeriana Carex dioica , Caltha palustris , PhrAus = Hamatocaulis vernicosus , Hamatocaulis SchPal = = Epilobium palustre , Eriophorum latifolium , LysThy = LysThy = MenTri C. protensum ), AlnGlu SphRip G.Roth, DreTun Drepanocladus badius (Hartm.) ScoSco revolvens ), WarFlu = WarFlu CxDio = BetNan = CalPal = S. Salix cinerea , Sphagnum obtusum , CxNig Willd., (Wahlenb.) EriLat = EpiPal HamVer HamVer Carex canescens , Utricularia intermedia agg., UtrMin = = Sphagnum teres , DreBad = Carex heleonastes L., CxLas = flexuosum ), SxCin =

SphObt = Sphagnum papillosum , Agrostis canina , Paludella squarrosa , Rhynchospora alba , S. Juncus bulbosus , = UtrInt = CxCan (Arnell) Loeske, DroAng = (Arnell) Lythrum salicaria , Lythrum = = = Carex diandra , Carex livida Carex = = = = agg. ( S. cossonii , agg. SphTer tundrae LytSal PalSqu Eriophorum gracile , Equisetum fluviatile , Equisetum RhyAlb agg., CxHel = vens Warnstorfia exannulata , Warnstorfia palustre , alpinum , Campylum stellatum (incl. Calliergon giganteum , CxDia CxLiv Carex rostrata , S. fallax , S. SphPap dum , Galium uliginosum , Galium ValDio = cinium oxycoccos agg., ValDio contortum , stramineum , Aulacomnium palustre , AgrCan JunBul

and druhů species a first positions in : druhovém vysvětluje (A) The suchopýrem v

osa se

axes. variability snímků - the of snímků variability ordinační

ordination Eriophorum gracile Eriophorum diagramy 3,6% První two with respectively, osa osy. first relevés ordinační 3.6%, along druhá and fytocenologických

(B) ordinační 4.4% DCA

souboru, druhé species a phytosociological explain and of

první axes (A) Eriophorum gracile ): ( Eriophorum DCA variability podél relevés lenská kotlina, Slovensko), Pr: Praskolesy (Jihlavské vrchy), Rh: Ratibořské Hory (Votická pahorkatina), Š: Šumavské pláně, Trstená T: (Oravská kotlina, Slovensko), V: Slovensko), Vidlák (Český V: ráj), VL: Vidnavské louky (Slezská pahorkatina), the second složení snímků. Zobrazeny jsou pouze druhy s váhou vyšší než 5%. composition of the relevés. Only species with a weight above 5% are shown. ZPr: Pilský rybník (Žďárské vrchy). of Fig. 2: štíhlým Obr. 2: Výsledky 2: Obr. (B) 1-ZL: Zlámanec (snímek č. 1), 2-ZL: Zlámanec (sn. č. 2), J: Jelšovec (Zvo 4,4%

52 Tab. 1: Hodnoty pH a konduktivity z lokalit suchopýru štíhlého (Eriophorum gracile). Tab. 1: Values of pH and conductivity measured at localities of Eriophorum gracile.

Koduktivita Region, lokalita pH Zdroj [µS/cm] ČR, Žďárské vrchy, 5,8–6,4 76–128 Zlámanec ČR, Český Ráj, Vidlák 6,2 Kratochvílová (2010) ČR, Třeboňsko, Ruda 4,6–5,5 18–63 Navrátilová & Navrátil (2005b) ČR, Třeboňsko, Hovízna 5,4 43 Navrátilová & Navrátil (2005a) ČR, Třeboňsko 5,4 Rybníček (1970) Bulharsko, Rodopy, Batak 5,8 64 Hájek et al. (2008) Polsko, SV část 6,4–7,1 256– 497 Dítě, Hájek, Hájková (ined.) Slovensko, Zvolenská 5,8–6,1 73–159 Dítě, Hájek, Hájková (ined.) kotlina, Jelšovec Slovensko, Oravská kotlina, 6,6–6,7 Rybníček (ined.) Trstená Belgie, Antwerp 6,5–7,2 van Straaten & Lembrechts (1982) Švýcarsko, Mt. Camignolo 6,4 123 Dítě, Hájek, Hájková (ined.) Švédsko, Bergslagen 6,2 Sjörs (1948) Švédsko, Dalarna, 6,2 Gunnarsson et al. (2000) Skattlösbergs Stormosse Švédsko, Norrland, Peterka, Hettenbergerová, Jiroušek, 5,2 26 Sappetavan Plesková (ined.) Švédsko, Norrland, Störon 6,8–7,4 110–230 Elveland (1976)

Summary Eriophorum gracile is presently known from only about ten localities within the entire Czech Republic, namely from the Třeboň basin, the Šumava Mts, the southwestern part of the Bohemian- Moravian Highlands, Český ráj, and the sourroundings of the town of Tábor. The occurence of this critically endangered plant species in the Žďárské vrchy Hills (northeastern part of Bohemian- Moravian Highlands) was recently confirmed. The locality, originally documented by Č. Deyl and V. Skalický in 1990 and confirmed by A. Hájek in 1999, is situated close to the village of Vortová nearby the Zlámanec fishpond at the altitude of 620 m a.s.l. Eriophorum gracile grows there in a rather well-preserved vegetation of the Agrostio caninae-Caricetum diandrae association (alliance Caricion canescenti-nigrae) which belongs among very rare mire communities within the territory of the Žďárské vrchy Hills now. Phytosociological relevés from Zlámanec were compared with other available vegetation-plot records with the presence of Eriophorum gracile. Plots stored in Czech National Phytosociological database were accompanied by unpublished relevés sampled by authors of this paper and some published data scattered in local journals or monographs. The resulting dataset comprised relevés from Czech and Slovak Republics, Bulgaria, Switzerland, Poland, Estonia, Sweden, Finland, and northwestern part of Russia. Main gradients in the floristic composition of relevés were assessed using the detrended correspondence analysis (DCA). A simple DCA ordination diagram (Fig. 2) sorted the samples with E. gracile from those recorded in temperate zone to those recorded in boreal zone. The second DCA axis can be interpreted as gradient of pH and total mineral richness, stretching from

53 rich fens with calcicole species (the Sphagno warnstorfii-Tomentypnion nitentis alliance) to plots with prevailing acidophilous species. We conclude that Eriophorum gracile tolerates different conditions along the pH/mineral richness gradient and its occurrence is hence influenced by other factors, such as stable water level, regular management or continuum of mire habitats in a region during the Holocene.

Poděkování Děkujeme Veronice Horsákové a Michalu Horsákovi za nápad uspořádat společný botanicko-zoologický výlet na Vysočinu a Ondrovi a Štěpánovi Hájkovým za ochotu trávit se svými rodiči čas v terénu. Vítu Grulichovi vděčíme za komentář k současnému výskytu suchopýru štíhlého v ČR, Ester Ekrtové a Jiřímu Juřičkovi za poznámky k recentním nalezištím v jihozápadní části Českomoravské vrchoviny. Petru Burešovi jsme zavázáni za všeobecné informace o lokalitě a o flóře Žďárských vrchů, Daně Michalcové za výpis z České národní fytocenologické databáze, Evě Hettenbergerové, Martinu Jirouškovi, Zuzaně Pleskové a dalším kolegům za nezištnou pomoc při zápisu fytocenologických snímků v různých částech Evropy. Za zasvěcené komentáře, které pomohly k podstatnému vylepšení první verze textu, děkujeme Kamile Juřičkové a Janu Košnarovi. Vznik článku byl podpořen projektem 14-36079G (Centrum excelence PLADIAS), Masarykovou uni- verzitou (projekt specifického výzkumu č. MUNI/A/1456/2014) a Botanickým ústavem AVČR (projekt RVO 67985939).

Literatura Almquist E., 1929: Upplands vegetation och flora. Acta Phytogeographica Suecica 1: 1–732. Balátová-Tuláčková E., 1965: Eriophorum gracile Koch na úvalenských loukách a jeho rozšíření na Moravě a ve Slezsku. Acta musei Silesiae, Ser. A 14: 127–130. Balátová-Tuláčková E., 1968: Významná lokalita v Českém ráji (severovýchodní Čechy). Zpr. Čs. Bot. Společ. 3: 89–90. Bollens U., Güsewell S. & Klötzli F., 1998: Zur relativen Bedeutung von Nährstoffein- trag und Wasserstand für die Biodiversität in Streuwiesen. Bulletin of the Geobotanical Institute ETH 64: 91–101. Bureš L., 2013: Chráněné a ohrožené rostliny Chráněné krajinné oblasti Jeseníky. Rubico, Olomouc. Bureš P. & Řepka R., 1989: Rozšíření vybraných ohrožených druhů cévnatých rostlin v západní části CHKO Žďárské vrchy I. Vlastiv. Sborn. Vysočiny, Odd. Věd Přír. 9: 101–116. Bureš P., 1993: Rozšíření vybraných ohrožených druhů cévnatých rostlin v CHKO Žďárské vrchy III. Rod Scirpus L. – regionálně fytogeografická studie. Vlastiv. Sborn. Vysočiny, Odd. Věd Přír. 11: 129–169. Bureš P., 2004: Inventazirační průzkum přírodní památky Zlámanec (CHKO Žďárské vrchy). Ms. [depon. in: AOPK ČR, Regionální pracoviště Správa CHKO Žďárské vrchy, Žďár nad Sázavou]. Čech L., Šumpich J. & Zabloudil V. (eds), 2002: Chráněná území ČR, 7. Jihlavsko. Agentura ochrany přírody a krajiny ČR & EkoCentrum Brno, Praha. Danihelka J., Chrtek J. Jr. & Kaplan Z., 2012: Checklist of vascular plants of the Czech Republic. Preslia 84: 647–811. Elveland J., 1976: Myrar på Storön vid norrbottenskusten. Wahlenbergia 3: 1–274. Faltys V. & Faltysová H., 1997: Příspěvek k rozšíření některých vzácnějších rostlin v oblasti Českého ráje. Vč. Sborn. Přír. – Práce a studie 5: 63–68.

54 Grulich V., 2012: Red List of vascular plants of the Czech Republic: 3rd edition. Preslia 84: 631–645. Grulich V., 2013: Eriophorum gracile W. D. J. Koch – suchopýr štíhlý. In: Lepší P., Lepší M., Boublík K., Štech M. & Hans V. (eds), Červená kniha květeny jižní části Čech, Jihočeské museum v Českých Budějovicích, České Budějovice: 188–189. Gunnarsson U., Rydin H. & Sjörs H., 2000: Diversity and pH changes after 50 years on the boreal mire Skattlosbergs Stormosse, Central Sweden. J. Veg. Sci. 11: 277–286. Hadinec J. & Lustyk P. (eds), 2011: Additamenta ad floram Reipublicae Bohemicae IX. Zpr. Čes. Bot. Společ. 46: 51–160. Hadinec J. & Lustyk P. (eds), 2015: Additamenta ad floram Reipublicae Bohemicae XIII. Zpr. Čes. Bot. Společ. 50: 23–129. Hájek M. & Hájková P., 2011: Vegetace slatinišť, přechodových rašelinišť a vrchovištních šlenků (Scheuchzerio palustris-Caricetea nigrae). In: Chytrý M. (ed.), Vegetace České republiky 3. Vodní a mokřadní vegetace, Academia, Praha: 614–704. Hájek M., Hájková P. & Apostolova I., 2008: New plant associations from Bulgarian mires. Phytologia Balcanica 14: 377–399. Hájek M., Jiroušek M., Navrátilová J., Horodyská E., Peterka T., Plesková Z., Navrátil J., Hájková P. & Hájek T., 2015: Changes in the moss layer of Czech fens indicate early succession triggered by nutrient enrichment. Preslia 87: 279–301. Hultén E. & Fries M., 1986: Atlas of North European vascular plants north of the Tropic of Cancer. Koeltz, Königstein. Chytrý M. & Rafajová M., 2003: Czech National Phytosociological Database: basic statistics of the available vegetation-plot data. Preslia 75: 1–15. Jirásek J., 1998: Rostlinná společenstva vod a mokřadů, stepí, skal a ruderálních míst Železných hor. Železné hory, Sborn. prací 7: 1–78. Juřička J., Juřičková K., Kubešová S., Novotný I., 2013: Vegetace a flóra PR Pod Kamenným vrchem v CHKO Žďárské vrchy. Acta rerum naturalium 14: 1–20. Kaplan Z., Danihelka J., Štěpánková J., Bureš P., Zázvorka J., Hroudová Z., Duchá- ček M., Grulich V., Řepka R., Dančák M., Prančl J., Šumberová K., Wild J. & Trávníček B., 2015: Distributions of vascular plants in the Czech Republic. Part 1. Preslia 87: 417–500. Koch M. & Jurasinski G., 2014: Four decades of vegetation development in a percolation mire complex following intensive drainage and abandonment. Plant Ecology & Diversity 8: 49–60. Kratochvílová L., 2010: Vegetace rašelinišť a pH: vliv metody měření. Ms. [dipl. práce; depon. in Ústav botaniky a zoologie, Masarykova univerzita, Brno]. Kučera J. & Váňa J., 2005: Seznam a červený seznam mechorostů České republiky (2005). Příroda 23: 1–104. Lepší M. & Lepší P., 2010: Nálezy zajímavých a nových druhů v květeně jižní části Čech XVI. Sborn. Jihočes. Muz. v Čes. Budějovicích, Přír. vědy 50: 75–96. Lounamaa J., 1961: Untersuchungen über die eutrophen Moore des Tulemajärvi-Gebietes im südwestlichen Ostkarelien, KASSR. Ann. Bot. Soc. Zool. Bot. Fenn. ‚Vanamo‘ 32: 1–63. Navrátilová J. & Navrátil J., 2005a: Hlavní typy nelesní rašeliništní vegetace Třeboňské pánve. Sborn. Jihočesk. Muz. v Čes. Budějovicích Přír. Vědy 45: 47–58. Navrátilová J. & Navrátil J., 2005b: Stanovištní nároky některých ohrožených a vzácných rostlin rašelinišť Třeboňska. Zpr. Čes. Bot. Společ. 40: 279–299. Paasio I., 1939: Zur Vegetation der eigentlichen Hochmoore Estlands. Ann. Bot. Soc. Zool. Bot. Fenn. ‚Vanamo‘ 11(2): 1–114.

55 Peterka T., 2013: Vegetace rašelinišť severovýchodní části Českomoravské vrchoviny a její vztah k vlastnostem prostředí. Ms. [dipl. práce; depon. in Ústav botaniky a zoologie, Masarykova Univerzita, Brno]. Procházka F., 1999: Eriophorum gracile Koch ex Roth. In: Čeřovský J., Feráková V., Holub J., Maglocký Š. & Procházka F. (eds), Červená kniha ohrožených a vzácných druhů rostlin a živočichů ČR a SR 5. Vyšší rostliny, Príroda, Bratislava: 150. Ruuhijärvi R., 1960: Über die regionale Einteilung der Nordfinnischen Moore. Ann. Bot. Soc. Zool. Bot. Fenn. ‚Vanamo‘ 31(1): 1–360. Růžička I., 1987: Výsledky záchranného výzkumu ohrožené květeny mizejících rašelinišť a rašelinných luk v okolí Telče na Českomoravské vrchovině. Vlastiv. Sborn. Vysočiny, Odd. Věd Přír. 8: 153–192. Růžička I., Čech L. & Lacina J., 1998: Floristický materiál z centrální části Českomo- ravské vrchoviny jihozápadně od Jihlavy. Vlastiv. Sborn. Vysočiny, Odd. Věd Přír. 13: 221–281. Rybníček K. & Rybníčková E., 1966: Poznámky k nálezům vzácných rašelinných rostlin v jižní části Českomoravské vysočiny. Preslia 38: 413–416. Rybníček K., 1970: Rhynchospora alba (L.) Vahl, its distribution, communities and habitat conditions in Czechoslovakia, Part 2. Folia Geobot. Phytotax. 5: 221–263. Rybníček K., 1974: Die Vegetation der Moore im südlichen Teil der Böhmisch-Mährischen Höhe. Vegetace ČSSR, ser. A 6: 1–243. Rybníčková E. & Rybníček K., 1988: Holocene Palaeovegetation and Palaeoenvironment of the Kameničská kotlina Basin (Czechoslovakia). Folia Geobot. Phytotax. 23: 285–301. Řepka R. & Lustyk P., 1998: Flóristické údaje vybraných druhů pro Květenu Šumavy. – Zpr. Čes. Bot. Společ. 32: 161–181. Sjörs H., 1948: Myrvegetation i Bergslagen. Acta Phytogeographica Suecica 21: 1–380. Skalický V., 1988: Regionálně fytogeografické členění. In: Hejný S. & Slavík B. (eds), Květena Československé socialistické republiky 1, Academia, Praha: 103–121. Šmilauer P. & Lepš J. Š., 2014: Multivariate analysis of ecological data using CANOCO 5. Cambridge Univ. Press, Cambridge. Štechová T. & Štech M., 2009: Současné lokality Hamatocaulis vernicosus (Mitt.) Hedenäs na Českomoravské vrchovině. Acta rerum naturalium 6: 13–24. Štechová T., Peterka T., Lysák F., Bradáčová J., Holá E., Hradílek Z., Kubešová S., Novotný I., Bartošová V., Velehradská T. & Kučera, J., 2014: Významné mechorosty rašelinišť na Českomoravské vrchovině na prahu 21. století. Acta rerum naturalium 17: 7–32. van der Maarel E., 1979: Transformation of cover-abundance values in phytosociology and its effects on community similarity. Vegetatio 39: 97–114. van Straaten D. & Lembrechts J., 1982: Gradient investigation of a peat-bog (Bui- tengoor-Meergoor/Mol, Belgium). 2. Phytosociological description. Bull. Soc. Roy. Botan. Belg. 115: 337–356. Vicherek J., 1958: Rostlinná společenstva rašelinných luk u Vidnavy. Přír. Sborn. Ostrav. kraj. 19: 185–221. Warén H., 1926: Untersuchungen über Sphagnumreiche Pflanzengesellschaften der Moore Finnlands unter Berücksichtigung der soziologischen Bedeutung der einzelnen Arten. Acta Soc. Fauna Flora Fenn. 55: 1–133.

Došlo: 25. 4. 2015

56 Curriculum vitae

Personal details Tomáš Peterka, born on 31st December 1987, Svitavy5, Czech Republic

Education 2013–present Masaryk university, Doctoral degree programme: Botany 2010–2013 Masaryk university, Master's degree programme: Ecological and Evolutionary Biology, thesis: Mire vegetation of the northeastern part of the Českomoravská vrchovina highlands and its relationships to environmental characteristics (in Czech) 2007–2010 Masaryk university, Bachelor's degree programme: Systematic Biology and Ecology; thesis: Wetland plant communities in the surrounding of the town of Polička (in Czech) 1999–2007 grammar school: Gymnázium Polička

Employment history 2013–present Masaryk University, Department of Botany and Zoology; research fellow

Research interests Mire ecology, vegetation classification, boreal ecosystems, flora and vegetation of the Bohemian-Moravian Highlands, management of wetlands, bryophytes Membership Česká botanická společnost (Czech Botanical Society) Sdružení Krajina

Publications

Publications in international peer-reviewed journals (in English):

Peterka T., Plesková Z., Jiroušek M. & Hájek M. (2014): Testing floristic and environmental differentiation of rich fens in the Bohemian Massif. – Preslia 86: 337–366. Hájek M., Plesková Z., Syrovátka V., Peterka T., Laburdová J., Kintrová K., Jirousek M. & Hájek T. (2014): Patterns in moss element concentrations in fens across species, habitats, and regions. – Perspectives in Plant Ecology, Evolution and Systematics 16: 203–218. Hájek M., Jiroušek M., Navrátilová J., Horodyská E., Peterka T., Plesková Z., Navrátil J., Hájková P. & Hájek T. (2015): Changes in the moss layer in Czech fens indicate early succession triggered by nutrient enrichment. – Preslia 87: 279–301. Peterka T., Jiroušek M., Hájek M. & Jiménez-Alfaro B. (2015): European Mire Vegetation Database: a gap- oriented database for European fens and bogs. – Phytocoenologia 45: 291–297. Zhai M., Hřívová D. & Peterka T. (2015): The harpacticoid assemblages (Copepoda: Harpacticoida) in the Western Carpathian spring fens in relation to environmental variables and habitat age. – Limnologica 53: 84–94.

5 It is necessary to mention that Svitavy is only place where I was born and I have never lived there. My personal and early scientific live is associated with the town of Polička, the place with genius loci and the gate of the Bohemian-Moravian Highlands (past and partly also current fen kingdom of central Europe).

216

Chytrý M., Hennekens S.M., Jiménez-Alfaro B., Knollová I., Dengler J., Jansen F., Landucci F., Schaminée J.H.J., Adid S., Agrillo E., Ambarlı D., Angelini P., Apostolova I., Attorre F., Berg C., Bergmeier E., Biurrun I., Botta- Dukát Z., Brisse H., Campos J.A., Carlón L., Čarni A., Casella L., Csiky J., Dušterevska R., Dajid Stevanovid Z., Danihelka J., De Bie E., de Ruffray P., De Sanctis M., Dickoré W.B., Dimopoulos P., Dubyna D., Dziuba T., Ejrnæs R., Ermakov N., Ewald J., Fanelli G., Fernández-González F., FitzPatrick Ú., Font X., García-Mijangos I., Gavilán R.G., Golub V., Guarino R., Haveman R., Indreica A., Işık Gürsoy D., Jandt U., Janssen J.A.M., Jiroušek M., Kącki Z., Kavgacı A., Kleikamp M., Kolomiychuk V., Krstivojevid Duk M., Krstonošid D., Kuzemko A., Lenoir J., Lysenko T., Marcenò C., Martynenko V., Michalcová D., Moeslund J.E., Onyshchenko V., Pedashenko H., Pérez-Haase A., Peterka T., Prokhorov V., Rašomavičius V., Rodríguez-Rojo M.P., Rodwell J.S., Rogova T., Ruprecht E., Rūsira S., Seidler G., Šibík J., Šilc U., Škvorc Ž., Sopotlieva D., Stančid Z., Svenning J.-C., Swacha G., Tsiripidis I., Turtureanu P.D., Uğurlu E., Uogintas D., Valachovič M., Vashenyak Y., Vassilev K., Venanzoni R., Virtanen R., Weekes L., Willner W., Wohlgemuth T. & Yamalov S. (2016): European Vegetation Archive (EVA): an integrated database of European vegetation plots. – Applied Vegetation Science 19: 173–180. Plesková Z., Jiroušek M., Peterka T., Dítě D., Hájková P., Navrátilová J., Šímová A., Syrovátka V. & Hájek M. (2016): Testing inter-regional variation in pH niches of fen mosses. – Journal of Vegetation Science 27: 352–364. Peterka T., Plesková Z., Palpurina S., Kalníková V., Lazarevid P. & Hájek M. (2016): Meesia triquetra, new relict moss for the Republic of Macedonia. – Herzogia 29: 66–71. Dítě D., Peterka T., Dítětová Z., Hájková P. & Hájek M. (2017): Arcto-alpine species as their niche margin: The Western Carpathian refugia of Juncus castaneus and J. triglumis in the European context. – Annales Botanici Fennici 54: 67–82. Horsák M., Hájek M., Horsáková V., Hlaváč J., Hájková P., Dítě D., Peterka T., Divíšek J., Potůčková A. & Preece R. (2017): Refugial occurrence and ecology of the land snail Vertigo lilljeborgi in fen habitats in temperate mainland Europe. – Journal of Molluscan Studies 83: 451–460. Kalníková V., Palpurina S., Peterka T., Kubešová S., Plesková Z. & Sabovljevid M. (2017): Bryophytes on river gravel bars in the Balkan mountains: new records and insights into ecology. – Herzogia 30: 370–386. Peterka T., Hájek M., Jiroušek M., Jiménez-Alfaro B., Aunina L., Bergamini A., Dítě D., Felbaba-Klushyna L., Graf U., Hájková P., Hettenbergerová E., Ivchenko T.G., Jansen F., Koroleva N.E., Lapshina E.D., Lazarevid P.M., Moen A., Napreenko M.G., Pawlikowski P., Plesková Z., Sekulová L., Smagin V.A., Tahvanainen T., Thiele A., Biţæ-Nicolae C., Biurrun I., Brisse H., Dušterevska R., De Bie E., Ewald J., FitzPatrick Ú., Font X., Jandt U., Kącki Z., Kuzemko A., Landucci F., Moeslund J.E., Pérez-Haase A., Rašomavičius V., Rodwell J.S., Schaminée J.H.J., Šilc U., Stančid Z. & Chytrý M. (2017): Formalized classification of European fen vegetation at the alliance level. – Applied Vegetation Science 20: 124–142. Peterka T., Kalníková V. & Plesková Z. (2017): Pseudocalliergon lycopodioides, a new bryophyte species for Montenegro. – Herzogia 30: 496–500. Horsák M., Polášková V., Zhai M., Bojková J., Syrovátka V., Šorfová V., Schenková J., Polášek M., Peterka T. & Hájek M. (2018): Spring-fen habitat islands in a warming climate: partitioning the effects of mesoclimate air and water temperature on aquatic and terrestrial biota. – Science of the Total Environment 634: 355– 365. Peterka T., Hájek M., Dítě D., Hájková P., Palpurina S., Goia I., Grulich V., Kalníková V., Plesková Z., Šímová A. & Štechová T. (2018): Relict occurrences of boreal brown-moss quaking rich fens in the Carpathians and adjacent territories. – Folia Geobotanica 53: 265–276. Bruelheide H., Dengler J., Jiménez‐Alfaro B., Purschke O., Hennekens S. M., Chytrý M., Pillar V. D., Jansen F., Kattge J., Sandel B., Aubin I., Biurrun I., Field R., Haider S., Jandt U., Lenoir J., Peet R. K., Peyre G., Sabatini F. M., Schmidt M., Schrodt F., Winter M., Adid S., Agrillo E., Alvarez M., Ambarlı D., Angelini P., Apostolova I., Arfin Khan M. A., Arnst E., Attorre F., Baraloto C., Beckmann M., Berg C., Bergeron Y., Bergmeier E., Bjorkman A. D., Bondareva V., Borchardt P., Botta‐Dukát Z., Boyle B., Breen A., Brisse H., Byun C., Cabido M. R., Casella L., Cayuela L., Černý T., Chepinoga V., Csiky J., Curran M., Dušterevska R., Dajid Stevanovid Z., De Bie E., De Ruffray P., De Sanctis M., Dimopoulos P., Dressler S., Ejrnæs R., El‐Rouf Mousa, El‐Sheikh M. A., Enquist B., Ewald J., Fagúndez J., Finckh M., Font X., Forey E., Fotiadis G., García‐Mijangos I., de Gasper A. L., Golub V., Gutierrez A. G., Hatim M. Z., He T., Higuchi P., Holubová D., Hölzel N., Homeier J., Indreica A., Isık Gürsoy D., Jansen S., Janssen J., Jedrzejek B., Jiroušek M., Jürgens N., Kącki Z., Kavgacı A., Kearsley E., Kessler M., Knollová I., Kolomiychuk V., Korolyuk A., Kozhevnikova M., Kozub Ł., Krstonošid D., Kühl H., Kühn I., Kuzemko A., Küzmič F., Landucci F., Lee M. T., Levesley A., Li C.-F., Liu H., Lopez‐Gonzalez G., Lysenko T., Macanovid A., Mahdavi P., Manning P., Marcenò C., Martynenko V., Mencuccini M., Minden V., Moeslund J. E., Moretti M., Müller J. V., Munzinger J., Niinemets Ü., Nobis M., Noroozi J., Nowak A.,

217

Onyshchenko V., Overbeck G. E., Ozinga W. A., Pauchard A., Pedashenko H., Pequelas J., Pérez‐Haase A., Peterka T., Petřík P., Phillips O. L., Prokhorov V., Rašomavičius V., Revermann R., Rodwell J., Ruprecht E., Rūsira S., Samimi C., Schaminée J. H. J., Schmiedel U., Šibík J., Šilc U., Škvorc Ž., Smyth A., Sop T., Sopotlieva D., Sparrow B., Stančid Z., Svenning J.-C., Swacha G., Tang Z., Tsiripidis I., Turtureanu P. D., Ugurlu E., Uogintas D., Valachovič M., Vanselow K. A., Vashenyak Y., Vassilev K., Vélez‐Martin E., Venanzoni R., Vibrans A. C., Violle C., Virtanen R., von Wehrden H., Wagner V., Walker D. A., Wana D., Weiher E., Wesche K., Whitfeld T., Willner W., Wiser S., Wohlgemuth T., Yamalov S., Zizka G. & Zverev A. (2019): sPlot – a new tool for global vegetation analyses. – Journal of Vegetation Science 30. – https://doi.org/10.1111/jvs.12710

Unpublished manuscripts (in English):

Peterka T., Syrovátka V., Dítě D., Hájková P., Hrubanová M., Jiroušek M., Plesková Z., Singh P., Šímová A., Šmerdová E. & Hájek M. (unpubl.): Is variable plot size a serious constraint in broad-scale vegetation studies? A case study on fens. [submitted to Journal of Vegetation Science]

Publications in Czech reviewed journals (in Czech):

Štechová T., Holá E., Gutzerová N., Hradílek Z., Kubešová S., Lysák F., Novotný I. & Peterka T. (2010): Současný stav lokalit druhů Meesia triquetra a Paludella squarrosa (Meesiaceae) v České republice. – Bryonora 45: 1–11. Peterka T. (2013): Doplněk k rozšíření druhu Paludella squarrosa na Českomoravské vrchovině. – Bryonora 52: 31–35. Štechová T., Peterka T., Lysák F., Bradáčová J., Holá E., Hradílek Z., Kubešová S., Novotný I., Bartošová V., Velehradská T. & Kučera J. (2014): Významné mechorosty rašelinišť na Českomoravské vrchovině na prahu 21. století. – Acta rerum naturalium 17: 7–32. Peterka T. (2014): Rostlinná společenstva pěnovců v údolí Kaviny na Poličsku. – Východočeský sborník přírodovědný, Práce a studie 21: 117–124. Peterka T. (2014): Nález kapradě hřebenité (Dryopteris cristata) ve Žďárských vrších. – Acta rerum naturalium 17: 85–88. Novák P., Peterka T., Roleček J. & Švarcová M. (2015): Nález ostřice Davallovy (Carex davalliana) v Lubenském lese na Litomyšlsku a poznámky k vegetaci nové lokality. – Východočeský sborník přírodovědný, Práce a studie 22: 111–119. Kučera J., Dřevojan P., Ekrtová E., Holá E., Koval Š., Manukjanová A., Peterka T., Procházková J., Štechová T., Táborská M., Tkáčiková J., Vicherová E. & Zmrhalová M. (2016): Zajímavé bryofloristické nálezy XXV. – Bryonora 57: 83–91. Kučera J., Dřevojan P., Hradílek Z., Kubešová S., Laburdová J., Lysák F., Manukjanová A., Koval Š., Peterka T., Soldán Z., Štechová T. & Zmrhalová M. (2016): Zajímavé bryofloristické nálezy XXVI. – Bryonora 58: 73–78. Peterka T., Dítě D., Hájková P. & Hájek M. (2016): Ověření výskytu suchopýru štíhlého (Eriophorum gracile) ve Žďárských vrších. – Východočeský sborník přírodovědný, Práce a studie 23: 47–56. Ďurčanová P., Jiroušek M. & Peterka T. (2017): Pozoruhodné nálezy Sphagnum affine a ďalších rašeliníkov v severnej časti západných Karpát na Slovensku. – Bryonora 59: 58–66. Dřevojan P., Holá E., Jandová L., Košnar J., Kubešová S., Kučera J., Manukjanová A., Mikulášková E., Müller F., Peterka T., Štechová T. & Štěrbová J. (2018): Zajímavé bryofloristické nálezy XXX. – Bryonora 62: 76–83. Dřevojan P., Novák P., Doležal J., Peterka T. & Zukal D. (2018): Komentované fytocenologické snímky z České republiky 3. – Zprávy České botanické společnosti 53: 177–187.

Unpublished manuscripts (in Czech):

Novák P. & Peterka T. (unpubl.): Poznámky k rozšíření a ekologii mokrýše vstřícnolistého (Chrysosplenium oppositifolium) na východním okraji areálu. *submitted to Východočeský sborník přírodovědný, Práce a studie]

218