Acta Botanica 1-2017 - za web.indd ACTA BOT. CROAT. 76 (1), 2017 15 Acta Bot. Croat. 76 (1), 15–26, 2017 CODEN: ABCRA 25 DOI: 10.1515/botcro-2016-0041 ISSN 0365-0588 eISSN 1847-8476 Ecological, fl oristic and functional analysis of zonal forest vegetation in Bosnia and Herzegovina Vladimir Stupar1*, Andraž Čarni2, 3 1 University of Banjaluka, Forestry Faculty, Department of Forest Ecology, S.Stepanovića 75a, BA-78000 Banjaluka, Bosnia and Herzegovina 2 Institute of Biology, Research Centre of the Slovenian Academy of Sciences and Arts, Novi trg 2, SI-1000 Ljubljana, Slovenia 3 University of Nova Gorica, Vipavska 13, SI-5000 Nova Gorica, Slovenia Abstract – Zonal vegetation is a large-scale expression of macro-climate and, due to the climatic diversity of the country, there are seven traditionally recognized zonal forest plant communities in Bosnia and Herzegov- ina. Using data from Bosnia and Herzegovina, this study aimed to reveal whether macro-climate is indeed the most important factor determining the existence of zonal forest plant communities (ZFPC). Detrended corre- spondence analysis of 398 relevés of seven ZFPCs revealed that the species turnover along the fi rst axis is strongly related to the macro-climatic gradient (annual mean temperature, mean temperature of the coldest quarter and precipitation of the warmest quarter). No correlation was detected between this gradient and topo- graphic factors (slope and aspect) and soil reaction. Floristic analysis revealed clear separation of ZFPCs in terms of diagnostic species. Functional analysis of all layers showed that competitive ecological strategy has the highest proportion, while analysis of the herb layer alone expressed a shift of CSR signatures towards the middle of the C–S axis. Ruderality was overall poorly expressed. Statistically signifi cant differences among communities were discovered along the C–S axis. In terms of life forms, statistically signifi cant differences in the proportions of Phanerophytes, Geophytes and Hemicryptophytes among communities were discovered. Our study confi rms that macro-climatic gradient is the most important determinant of the species turnover along ZFPCs. CSR signatures show that zonal forest vegetation is represented by productive communities in a terminal stage of succession. This does not refer to degraded Quercus ilex stands (maquis), which are in the middle stage of secondary succession. Keywords: Balkans, climatic gradient, chorotypes, ecological strategies, life forms, ordination, plant func- tional types, zonal communities * Corresponding author, e-mail: vladimir.stupar@sfbl.org Introduction Every plant community is a result of a complex interac- tion of various ecological factors in a given place and time. In terms of natural vegetation, there are three types of veg- etation that generally develop in accordance with the biotic, climatic and soil conditions: zonal, extrazonal and azonal (Dierschke 1994, Ellenberg 2009, Surina 2014). Zonal veg- etation is a large-scale expression of climate dominating a particular area, while it is not confi ned to specifi c soil con- ditions, i.e., it most precisely refl ects the macroclimatic con- ditions of particular regions (Kovar-Eder and Kvaček 2007). Zonal vegetation often does not represent a single homoge- neous plant association but rather a number of similar commu- nities, which can differ to some extent and, consequently, it is possible to talk of a ‘zonal vegetation group’ (Ellenberg 2009). For example, when different bedrocks and soil series occur within the same climatic zone or there is a species turnover due to minor biogeographical differences, several different, yet similar associations make up the zonal vegeta- tion group. Furthermore, a mountainous relief of a particu- lar climatic zone leads to vertical differentiation of the cli- matic factors and, accordingly, vertical differentiation of zonal vegetation communities, which are then often called ‘altitudinal belts’. Although in the last several thousand years man has de- forested great parts of Europe (Gunia et al. 2012), the po- tential natural vegetation and, consequently, the zonal veg- etation of most of temperate Europe is forest (Bohn and Neuhäusl 2004, Ellenberg 2009). The same is valid for Bos- nia and Herzegovina (B&H) (Horvat et al. 1974, Stefanović STUPAR V., ČARNI A. 16 ACTA BOT. CROAT. 76 (1), 2017 et al. 1983). B&H is situated in the western part of the Bal- kan Peninsula in SE Europe (Fig. 1). Its climate is very di- verse, since two major climatic zones overlap here: central European from the north and Mediterranean from the south. The climate of the transitional zone is highly modifi ed by the infl uence of mountain massifs (Dinaric Alps), while in the eastern part of country the infl uence of the drier conti- nental climate can be felt (Milosavljević 1976). While veg- etation studies of forests in the neighboring regions have produced synthetic overviews (Borhidi et al. 2012, Vukelić 2012, Chytrý 2013, Tomić and Rakonjac 2013), despite the fairly long tradition of phytosociological studies in B&H (Horvat 1933, 1941, Horvat and Pawlowski 1939, Tregubov 1941) and the considerable number of relevés, the ecology, nomenclature and syntaxonomical position of the majority of B&H forests still remains unsettled (Redžić 2007). With the exception of thermophilous deciduous forest communi- ties (Stupar et al. 2015), this also applies to zonal vegeta- tion, which has only been the subject of a few studies in the past (Horvat et al. 1974, Stefanović et al. 1983, Beus 1984). However, following the macro-climatic diversity, these authors generally agree that seven zonal forest plant communities (ZFPCs) can be distinguished for the territory of B&H (Tab. 1, Fig. 1). Four communities are represented by vari- ous oak forests, which occupy the lowlands and hilly region of B&H, while the other three are altitudinal belts (mon- tane, altimontane and subalpine) above the oak forests, main- ly built by various types of beech forests (pure and mixed). There are two different conceptual frameworks for the study of plant communities. The traditional approach to classifi cation is performed at the species level, while a more recent ‘functional’ approach is based on plant functional types (Duckworth et al. 2000, Shipley 2010, Škornik et al. 2010). Plant functional types are non-phylogenetic group- ings of species that show close similarities in their response to environmental and biotic factors and are derived from plant traits based on species morphology, physiology and/or life history (Pérez-Harguindeguy et al. 2013). Plant func- tional types can aid in the understanding of ecological pro- cesses, such as the assembly and stability of communities and succession, and facilitate the detection and prediction of response to environmental change on a range of scales (Duckworth et al. 2000). With plant functional types, com- parisons between communities of widely differing compo- sition can be facilitated (Diaz et al. 2004). There are two main approaches to the classifi cation of functional types Fig. 1. Location of the study area in Bosnia and Herzegovina. The potential zonal forest plant communities (ZFPCs) are indicated (Horvat et al. 1974, modifi ed after Stefanović et al. 1983). Num- bers on the map correspond to community numbers in Tab. 1. Tab. 1. Zonal forest plant communities in Bosnia and Herzegovina. Community numbers correspond to those used in Tabs. 3–4, and Figs. 1–4. Asterisk (*) denotes provisional invalid names still in use in Bosnia and Herzegovina. Commu- nity no. Forest type Related syntaxa No. of relevés No. of resampled relevés 1 Quercus ilex Quercion ilicis Br.-Bl. 1931 (1936) 5 5 2 Quercus pubescens- Carpinus orientalis Querco pubescenti-Carpinetum orientalis Horvatić 1939 (Carpinion orientalis Horvat 1958) 30 16 3 Quercus frainetto Quercetum frainetto-cerridis (Rudski 1949) Trinajstić et al. 1996 (Quercion frainetto Horvat 1954) 38 24 4 Quercus petraea- Carpinus betulus Querco-Carpinetum illyricum Horvat et al. 1974* (Erythronio-Carpinion betuli (Horvat 1938) Marinček in Wallnöfer et al. 1993) 43 26 5 pure Fagus / mixed Fagus-Abies Fagetum montanum illyricum Fukarek et Stefanović 1958*, Abieti-Fagetum dinaricum Tregubov 1957* (Aremonio-Fagion Török et al. ex Marinček et al. 1993) 231 162 6 mixed Picea- Abies-Fagus Piceo-Abieti-Fagetum Stefanović et al. 1983* (Abieti-Piceenion Br.-Bl. in Br.-Bl. et al. 1939) 191 123 7 Subalpine Fagus sylvatica Fagetum subalpinum s. lato* (Saxifrago rotundifoliae-Fagenion Marinček et al. 1993) 74 42 ZONAL FOREST VEGETATION OF BOSNIA AND HERZEGOVINA ACTA BOT. CROAT. 76 (1), 2017 17 (Shipley 2010). The fi rst is used especially by plant geogra- phers, concentrating on the morphological properties of plants, e.g., Raunkiaer’s life-forms (Raunkiaer 1934) and how, on average, such morphologies change as a function of major climatic variables. A second approach is based on the notion of ecological ‘strategies’, for which Grime’s model of CSR triangle (Grime 1974, 1977, 2001) is often used (Pugnaire and Valladares 2007). This model is a classifi ca- tion based on how plants deal with two groups of external factors, i.e., stress and disturbance, which results in three primary plant strategies: competitors (C), stress-tolerators (S), and ruderals (R) and four secondary, intermediate ones. The position of each species, as well as each relevé, can be determined in a CSR triangle. The whole community is thus given a functional signature, which can be very useful in comparative studies involving widely differing samples (Hunt et al. 2004). While the use of ecological strategies is quite common in studies of herbaceous vegetation (Hunt et al. 2004, Zel- nik and Čarni 2008, Škornik et al. 2010, Pipenbaher et al. 2013), it has lately also been gaining ground in the study of forest and woodland communities (Kilinç et al. 2010, Paušič and Čarni 2012, Juvan et al. 2013, Košir et al. 2013b, Rozman et al. 2013). Its use is most often related to studies of changes in communities’ composition as a response to disturbance, succession, environmental changes or in gradi- ent analysis. The aim of this study was to test whether zonal vegeta- tion is an expression of macro-climatic conditions or whether there are also other environmental factors involved. The un- derlying assumptions were (1) that the macro-climatic gra- dient is the most important for the species and structure turnover along the gradient of ZFPCs in B&H; and (2) that ZFPCs would demonstrate similarities in Grime’s ecologi- cal strategies, due to the fact that the communities are in terminal stages of succession, with a low level of degrada- tion, occupying habitats on moderately fertile soils. Materials and methods Data collection and preparation Seven ZFPCs were identifi ed for B&H (Tab. 1, Fig. 1, Ste fanović et al. 1983, Beus 1984): (1) eu-Mediterranean ever green Quercus ilex forests (a very small area of the warmest, southernmost part of B&H; represented by maquis); (2) sub-Mediterranean thermophilous deciduous Quercus pubescens-Carpinus orientalis forests (a major part of the lowland and hilly region in southern B&H); (3) Central Balkans thermophilous Quercus frainetto forests (relatively narrow lowland and hilly belt of eastern B&H); (4) Illyrian mesophilous Quercus petraea-Carpinus betulus forests (major part of northern B&H and marginally in cen- tral B&H); (5) montane mesoneutrophilous pure Fagus or mixed Fagus-Abies forests (altitudinal belt above oak for- ests in all B&H but mainly in the mountainous region of central B&H (Dinaric Alps); (6) alti-montane, colder and more acidophilous mixed Picea-Abies-Fagus forests (alti- tudinal belt above Community 5, mainly in the Dinaric mountainous region); and (7) subalpine Fagus sylvatica forests (uppermost forest altitudinal belt). Relevés were extracted from the Forest vegetation data- base of Bosnia and Herzegovina stored in the Global index of vegetation-plot databases (Dengler et al. 2011) with the ID EU-BA-001. This database consists of 2810 published and available unpublished forest vegetation relevés in B&H. We compiled all relevés that were assigned to one of the seven ZFPCs by their authors (Tab. 1), with the excep- tion of two zonal communities of thermophilous deciduous forests (Communities 2 and 3), for which we used the re- sults of formalized classifi cation (Stupar et al. 2015). We did not consider relevés of stands with less than 70% of canopy cover, nor those with edifi er tree species cover val- ue less than 3 (25%) on the Braun-Blanquet scale, consider- ing them structurally degraded. Only stands of high, pro- ductive forests were thus taken into account, except in the case of Community 1 (Quercus ilex stands) because well- established stands do not exist in B&H, and all relevés were made in structurally degraded maquis. All relevés were made using the standard Central European phytosociologi- cal method (Braun-Blanquet 1964). Only relevés that could be georeferenced relatively precisely and those that con- tained complete species records were taken into consider- ation. In all 612 relevés were compiled in the Turboveg da- tabase (Hennekens and Schaminée 2001) and exported to JUICE 7 software (Tichý 2002) for further analysis. Mosses, as well as taxa determined only to the genus level, were removed from the data set prior to numerical analysis. All vegetation layers were merged into one layer. Taxonomy and nomenclature followed Flora Europaea (Tu- tin et al. 1968–1993) unless a more modern taxon concept or circumscription suggested otherwise. These taxa, as well as those from taxonomically critical groups that were com- bined into aggregates (agg.) or species that included several subspecies that were not always recorded or recognized by authors and were combined under the abbreviation ‘s.l.’ (sensu lato), were listed in On-line Suppl. Tab. 1. The dubi- ous taxon Quercus dalechampii was treated as part of Quer- cus petraea agg., following Di Pietro et al. (2012). Records of Fagus moesiaca were treated as F. sylvatica (Marinšek et al. 2013). To avoid geographic overrepresentation of some vege- tation types due to oversampling of certain regions, we per- formed geographical stratifi cation and resampling of the initial data set (Knollová et al. 2005), a frequent strategy in recent national and regional level vegetation studies (Chy- trý 2013, Košir et al. 2013a, Rodríguez-Rojo et al. 2014). Stratifi cation was performed in a geographical grid with 1 km2 size. If two or more relevés assigned to the same com- munity fell in the same grid cell, only one of them was se- lected. Stratifi cation was not performed on Community 1 since it consisted of only fi ve relevés. The resulting strati- fi ed data set contained 398 relevés and 669 species. Data analysis The data set was then subjected to detrended correspon- dence analysis (DCA) in R software, version 2.10.1 (R Deve- lopment Core Team 2009) using the vegan package (http:// cc.oulu.fi /~jarioksa/softhelp/vegan.html) on presence-absence data. To extract the main gradients in species composition, STUPAR V., ČARNI A. 18 ACTA BOT. CROAT. 76 (1), 2017 398 relevés, together with the selected ecological variables were projected onto the two-dimensional ordination space of DCA. Unweighted average species ecological indicator values (EIVs) for soil reaction (Pignatti et al. 2005) and se- lected climatic variables available from the WorldClim da- tabase (Hijmans et al. 2005) were used as explanatory eco- logical variables. The signifi cance of EIVs correlation with the DCA relevé scores was tested using the modifi ed per- mutation test proposed by Zelený and Schaffers (2012). Climate variables best explaining variation in species com- position were selected through forward selection in canoni- cal correspondence analysis (CCA) in CANOCO 4.5 software (Microcomputer Power, Ithaca, NY, US). Other explanatory variables (altitude, inclination of slope, aspect, chorotypes, latitude and longitude, life forms and ecological strategies) were selected and projected based on the strength of corre- lation with the fi rst and second DCA axis. The signifi cances of correlations between these explanatory variables and DCA relevé scores were calculated using the Kendall tau coeffi cient in Statistica software (StatSoft, Inc.; v 7.0, http:// www.statsoft.com). Chorological spectra of the zonal forest communities (Gajić 1980, Pignatti et al. 2005) were calcu- lated for each relevé using presence-absence data. Endemic Illyrian species were separated from southeast European species in a broader sense. In order to reveal the fl oristic differences among the seven ZFPCs, we calculated their diagnostic species in the resampled data set using phi coeffi cient in the JUICE 7 pro- gram (Chytrý et al. 2002), after standardization to a relevé group size equal to a seventh of the total data set size (Tichý and Chytrý 2006). Fisher’s exact test was calculated giving a zero fi delity value to species whose phi values were not statistically signifi cant (P>0.001). The threshold phi value for a species to be considered diagnostic was set at 0.25. The functional study of zonal forest communities was performed using data of plant functional traits, i.e., life forms (Raunkiaer 1934) and ecological strategies (Grime 1977). Plants were classifi ed according to Grime’s primary and secondary strategies into seven functional types using data from the BIOLFLOR database (Klotz et al. 2002). We thus obtained data for about 80% of species. Ecological strategies for the remaining species were calculated using the dichotomous key suggested by Véla (2002). Averages of each strategy type weighted by cover-percentage were calculated and standardized for each relevé as suggested by Hunt et al. (2004). Ecological strategy scores for each group of relevés were then calculated with the CSR Signa- ture Calculator 1.2 program (Hunt et al. 2004) and repre- sented on a CSR ternary plot. We did a separate analysis for all layers and for the herb layer alone because herb func- tional signatures respond more quickly to environmental changes in the course of late succession (Paušič and Čarni 2012). Plant life forms obtained from Pignatti et al. (2005) and supplemented by our expert knowledge were used for the calculation of life forms spectra for each relevé using presence-absence data, and mean proportions were com- pared between ZFPCs. The occurrence of an individual functional trait in each of the ZFPCs was compared using the Scheffé post hoc test for normal distributions and Krus- kal-Wallis test by ranks and median for non-normal distri- butions using Statistica software. A normality check was performed using Lilliefors test. Results Gradient analysis within ZFPCs in Bosnia and Herzegovina Forward selection suggested that the climatic variables with the highest explanatory value of the variation in spe- cies composition are annual mean temperature (BIO1), mean temperature of the coldest quarter (BIO11) and pre- cipitation of the warmest quarter (BIO18). The fi rst DCA axis represents the main gradient in the data set, and all three climatic variables are signifi cantly related to the fi rst DCA axis at P<0.001 (Tab. 2, Fig. 2). It runs from the cold- Tab. 2. Correlations (Kendall-Tau coeffi cient) between detrended correspondence analysis (DCA) relevé scores and explanatory vari- ables. BIO1 – annual mean temperature; BIO11 – mean temperature of the coldest quarter; BIO18 – precipitation of the warmest quarter; Eur – European and Eurasian; SEur – south and southeast European; EuriMed – Euri-Mediterranean; StMed – Steno-Mediterranean; Illyr – Endemic Illyrian; Bor – Boreal; SEOro – South European orophytes; Cosm – Widespread species; P – Phanerophytes; NP – Nano- phanerophytes; Ch – Chamaephytes; H – Hemicryptophytes; G – Geophytes; T – Terophytes; C – Competitors; S – Stress-tolerators; R – Ruderals. Asterisk (*) denotes signifi cant correlation at P<0.001. Ecological factors Geographical factors Altitude Aspect Slope BIO1 BIO11 BIO18 Latitude Longitude DCA 1 –0.521* 0.046 0.050 0.569* 0.566* –0.451* –0.200* 0.367* DCA 2 –0.151* –0.034 –0.103 0.118* 0.103 –0.011 0.028 –0.041 Chorotypes Eur SEur EuriMed StMed Illyr Bor SEOro Cosm DCA 1 0.088 0.203* 0.445* 0.309* –0.172* –0.313* –0.510* –0.155* DCA 2 0.077 –0.034 0.035 –0.004 –0.066 –0.009 –0.138* 0.077 Life forms Ecological strategies P NP Ch H G T C S R DCA 1 0.182* 0.020 0.045 0.080 –0.317* 0.109 0.045 –0.048 0.014 DCA 2 0.010 –0.059 0.110 0.057 –0.121* 0.126* –0.049 –0.067 0.109 ZONAL FOREST VEGETATION OF BOSNIA AND HERZEGOVINA ACTA BOT. CROAT. 76 (1), 2017 19 est and most mesophilous subalpine beech forests (Com- munity 7) on the left side of the diagram to the most xero- thermophilous Quercus ilex maquis (Community 1) on the far right side of diagram. It is positively correlated with an- nual mean temperature and mean temperature of the coldest quarter, and negatively with the precipitation of the warm- est quarter (Fig. 2, Tab. 2). These are strong indicators that the fi rst DCA axis represents a macro-climatic gradient that runs from wet summers, cold winters and lower annual temperatures to dry summers but warmer winters and high- er overall annual temperatures. There is also a high nega- tive correlation with altitude, which is again related to mac- ro-climatic factors. The positive correlation with longitude and negative correlation with latitude refl ects the fact that south and east parts of B&H are warmer and dryer than the north and west. Correlations between the fi rst axis and slope and aspect are not statistically signifi cant (Tab. 2). A modifi ed permutation test also showed that the correlation between DCA 1 and EIVs for soil reaction was not statisti- cally signifi cant (R2=0.125, P=0.195). In terms of choro- types, DCA axis 1 shows a positive correlation with the proportions of south- and southeast European, Euri-Medi- terranean and Steno-Mediterranean chorotypes, and a nega- tive correlation with Boreal and south European orophytic chorotypes, while the correlation with the most abundant European and Eurasian chorotypes is not statistically sig- nifi cant. Correlations between the fi rst axis and life forms are signifi cant only for the proportions of phanerophytes (positive) and geophytes (negative). There is no signifi cant correlation between ecological strategies and DCA axes (Tab. 2). Floristic differentiation Analysis revealed differences in the fl oristic composi- tion among the seven ZFPCs. Tab. 3 shows a simplifi ed fre- quency-fi delity synoptic table of the ZFPCs of B&H based on a data set of 398 resampled relevés (full version is given in On-line Suppl. Tab. 2). Fig. 2. Detrended correspondence analysis (DCA) spider plot of 398 relevés of zonal forest plant communities with climatic vari- ables, ecological indicator values (EIVs) for soil reaction and alti- tude passively projected. The surface variable is annual mean tem- perature. The length of the fi rst DCA axis is 7.02 SD units, the length of the second axis is 2.85 SD units. Centroids of clusters are indicated by numbers that refer to Tab. 1. Tab. 3. Frequency-fi delity table of zonal forest plant communities (ZFPCs) in Bosnia and Herzegovina. Frequencies of species are pre- sented as percentages with phi values multiplied by 100 shown in superscript. Diagnostic species (phi values higher than 0.25) for each community are shaded (only ten species with the highest phi value for every community are presented). Proportions of chorotypes and altitudinal ranges of each community are also shown. Community numbers correspond to those used in Tab. 1 and Fig. 1. Community number 1 2 3 4 5 6 7 No. of relevés 5 16 24 26 162 123 42 Altitudinal range (m) 0–150 100–650 150–700 200–900 700–1400 800–1600 1400–1800 Chorotype (% of all species in a community) Widespread species 3 1 3 5 7 7 4 European and Eurasian 13 29 56 67 59 55 49 south and southeast European 21 30 21 16 13 10 12 Eurimediterranean 29 29 13 5 1 1 0 Stenomediterranean 33 7 0 0 0 0 0 Endemic Illyrian 1 3 1 0 1 2 3 Boreal 0 1 6 6 10 13 13 South European orophytes 0 0 0 1 9 12 19 ZFPC 1 Quercus ilex 100 96.5 6 – . . . . . Arbutus unedo 80 88 . . . . . . Teucrium polium ssp. capitatum 80 84.1 6 – . . . . . Centaurium erythraea 80 83.2 . . 8 – . . . Pistacia terebinthus 100 82.6 38 18.3 . . . . . Juniperus phoenicea 60 75 . . . . . . Pistacia lentiscus 60 75 . . . . . . Juniperus oxycedrus ssp. macrocarpa 40 60.3 . . . . . . STUPAR V., ČARNI A. 20 ACTA BOT. CROAT. 76 (1), 2017 Community number 1 2 3 4 5 6 7 No. of relevés 5 16 24 26 162 123 42 Altitudinal range (m) 0–150 100–650 150–700 200–900 700–1400 800–1600 1400–1800 Crepis sancta 40 60.3 . . . . . . Cistus salvifolius 40 60.3 . . . . . . ZFPC 2 Quercus pubescens 20 – 100 84.5 12 – . . . . Cornus mas . 88 83.6 8 – 8 – . . . Acer monspessulanum . 62 76.7 . . . . . Sesleria autumnalis 20 – 81 76.5 . . 1 – 1 – . Frangula rupestris . 56 72.4 . . . . . Viola hirta . 75 69.4 17 – 12 – . . . Rubus ulmifolius . 50 67.9 . . . . . Carex halleriana 20 – 62 64.2 . . . . . Brachypodium sylvaticum . 75 61.1 12 – 19 – 5 – 6 – 7 – Juniperus oxycedrus 40 – 69 60 . . . . . ZFPC 3 Quercus frainetto . 19 – 100 90.3 . . . . Chamaecytisus hirsutus agg. . . 58 70.4 4 – . 1 – . Quercus cerris . 50 23 100 69.8 27 – 1 – 1 – . Thymus pulegioides . . 54 65.7 . 4 – 1 – 2 – Lychnis coronaria . . 46 64.8 . . . . Lathyrus niger . 12 – 58 63 4 – . . . Euphorbia cyparissias . 19 – 58 61.7 . . . . Dianthus armeria . . 42 61.6 . . . . Carex caryophyllea . 19 – 58 57.8 8 – . . . Silene viridifl ora . 12 – 46 55.4 . . . . ZFPC 4 Carpinus betulus . . 42 19.6 100 77.3 11 – . . Cruciata glabra . . 4 – 69 67.5 11 – 8 – . Prunus avium . . 25 – 65 60.7 6 – 1 – . Luzula luzuloides . . . 58 59.9 1 – 8 – 12 – Pteridium aquilinum . . 38 – 77 57.7 12 – 15 – . Stellaria holostea . . 8 – 50 56.6 1 – . 7 – Melampyrum pratense . . 17 – 50 54.5 . 1 – 2 – Erythronium dens-canis . . . 27 49 . . . Crataegus monogyna . 44 – 29 – 69 45.3 15 – 2 – . Hieracium racemosum . . . 23 45.2 . . . ZFPC 5 Galium odoratum . . . 12 – 76 46.5 59 – 36 – Acer pseudoplatanus . . . 12 – 81 44.6 69 – 52 – Cardamine bulbifera . . . 12 – 67 42.3 30 – 55 – Lonicera xylosteum . . . . 31 38.9 15 – 2 – Cardamine enneaphyllos . . . . 62 38.1 35 – 62 – Daphne mezereum . . . . 51 35.1 37 – 38 – Salvia glutinosa . . 4 – 4 – 31 34.9 14 – 2 – Glechoma hirsuta . . 17 – 19 – 50 34.6 26 – 12 – Rubus hirtus . . 12 – 42 – 57 33.1 36 – 14 – Cardamine trifolia . . . . 28 32.9 20 – 2 – ZFPC 6 Picea abies . . . . 51 – 100 59.7 69 – Tab. 3. – continued ZONAL FOREST VEGETATION OF BOSNIA AND HERZEGOVINA ACTA BOT. CROAT. 76 (1), 2017 21 Analysis of functional traits In the standard CSR triangle plot, communities are posi- tioned in the upper right part of the diagram (Fig. 3a). This means that zonal forest plant communities are dominated by species with considerable competitive capacity. However, there is apparent divergence of Communities 1, 2 and 4 from the main cluster along the C–S axis, whereby statistically signifi cant differences were discovered (Tab. 4). The high- est ratio of stress tolerant-species occurs in Community 1, much less in Community 2, while the most competitive abi- lity is expressed by Community 4. CSR signatures of the herb layer are shifted to the middle of the C–S axis (Fig. 3b). The most signifi cant differences in functional traits of a life form were detected for Phanerophytes and Geophytes between Communities 1–4 and Communities 5–7, and for Hemicryptophytes between Communities 1 and 5 and the rest of the data set (Tab. 4). The highest ratios of Phanero- Fig. 3. CSR triangle of the ordination of zonal forest plant com- munities in Bosnia and Herzegovina according to ecological strat- egy scores (proportion of competitors – C, stress-tolerators – S and ruderals – R components in each community). Community numbers correspond to those used in Tab. 1 and Fig. 1. (a) All lay- ers together; (b) Herb layer. Community number 1 2 3 4 5 6 7 No. of relevés 5 16 24 26 162 123 42 Altitudinal range (m) 0–150 100–650 150–700 200–900 700–1400 800–1600 1400–1800 Abies alba . . . 8 – 73 – 98 52.5 76 – Athyrium fi lix-femina . . . 8 – 38 – 63 49 14 – Galium rotundifolium . . . . 9 – 41 46.5 12 – Oxalis acetosella . . . . 64 – 83 46.1 69 – Lonicera nigra . . . . 19 – 46 43.9 17 – Sorbus aucuparia . . . . 33 – 63 42.1 50 – Senecio nemorensis s.l. . . . . 38 – 59 40.8 38 – Prenanthes purpurea . . . 4 – 46 – 68 38.9 67 – Lamiastrum galeobdolon . . . 27 – 56 – 73 38.8 52 – ZFPC 7 Saxifraga rotundifolia . . . . 12 – 24 – 76 66.7 Luzula sylvatica . . . . 3 – 20 – 67 65.6 Adenostyles alliariae . . . . 5 – 23 – 64 61.8 Valeriana montana . . . . 4 – 8 – 45 55.3 Cicerbita alpina . . . . 1 – 10 – 43 53.9 Veronica urticifolia . . . . 4 – 23 – 52 52.8 Ranunculus platanifolius . . . . 2 – 5 – 38 52.4 Astrantia major . . . . . 1 – 31 51.8 Homogyne alpina . . . . . . 29 50.5 Veratrum lobelianum . . . . . . 24 46 Species diagnostic for more than one community Phillyrea latifolia 100 78.2 50 28.4 . . . . . Clematis fl ammula 80 64.5 50 33 . . . . . Asparagus acutifolius 60 40.5 81 62.1 . . . . . Carpinus orientalis . 88 68.3 54 34.5 . . . . Fraxinus ornus 60 – 100 50.4 83 36.5 31 – 2 – 1 – . Genista tinctoria . . 54 50.6 35 26.7 1 – . . Quercus petraea . . 71 43.7 100 71.3 1 – 1 – . Acer campestre . 12 – 42 27.5 58 45.1 4 – . . Fagus sylvatica . . 12 – 46 – 100 39.8 100 39.8 100 – Vaccinium myrtillus . . . 4 – 12 – 57 40.3 57 40.6 Tab. 3. – continued STUPAR V., ČARNI A. 22 ACTA BOT. CROAT. 76 (1), 2017 phytes and Nanophanerophytes are found in the fi rst two communities, with a decline in Phanerophytes towards Community 7 (Fig. 4). Beech-dominated forests (Commu- nities 5–7) have twice the ratio of Geophytes compared to oak-dominated forests (Communities 1–4). There is the highest proportion of Chamaephytes in Community 1, while the biggest share of Therophytes is found in Commu- nities 1 and 3. Discussion Our study strongly suggests macro-climatically based differentiation of ZFPCs in B&H. While gradient analysis revealed a major infl uence of climatic factors on the species turnover, there is little or no impact of topographic (slope and aspect) or edaphic conditions (Fig. 2, Tab. 2), which is congruent with the traditional understanding of zonal vege- tation (Dierschke 1994, Ellenberg 2009, Surina 2014). Macro-climatic gradient is supported by the signifi cant cor- relation with the geographic factors (altitude, longitude and latitude), which are all determinants of climate on a larger scale (Ellenberg 2009, Adams 2010). A recent study of zon- al forests of Korean Peninsula similarly showed that the main factor discriminating individual forest types was tem- perature (annual mean temperature, as well as temperature extremes) followed by precipitation, altitude and aridity (Černý et al. 2015). While chorotypes indicate units that de- scribe distribution patterns shared by several species, ex- plaining origins and history of development of particular fl oras (Pignatti and Pignatti 2014, Passalacqua 2015), some studies suggest that there is a correlation between choro- types and climatic conditions (Ferrer-Castán and Vetaas 2003, Abbate et al. 2012). In the case of our study, the pro- portion of ‘warmer’ (Euri-Mediterranean and Steno-Medi- terranean) chorotypes increases in a direction of the main gradient, while the proportion of ‘colder’ (Boreal and Oro- phytic) chorotypes decreases (Tab. 3), which again suggests the macro-climatic gradient. Floristic analysis showed a clear separation of seven ZFPCs in B&H. Community 1 (Quercus ilex maquis, Tab. 3, col. 1) occupies only a small area (about 20 km2) in the extreme south of B&H. It is represented by the secondary succession stage of the eastern Adriatic eu-Mediterranean zonal association Fraxino orni-Quercetum ilicis (Kutleša and Lakušić 1964). Mean annual temperatures are above 15 °C (Fig. 2) and elevations 0–150 m. Diagnostic species are mainly of Steno-Mediterranean, Euri-Mediterranean and south and southeast European chorotypes. Diagnostic herb species are mainly indicators of rocky habitats due to struc- tural degradation. Although in the last 50 years this com- munity has expanded its distribution area in B&H, due to the abandonment of coppicing, burning and goat breeding, there are still no high stands (Drešković et al. 2011). The situation is similar, though a little better, in neighboring Croatia (Vukelić 2012). Community 2 (Quercus pubes- Fig. 4. Proportions of life forms in zonal forest plant communities in Bosnia and Herzegovina. Community numbers correspond to those used in Tab. 1 and Fig. 1. P – Phanerophytes; NP – Nano- phaneorphytes; Ch – Chamaephytes; H – Hemicryptophytes; G – Geophytes; T – Therophytes. Tab. 4. Comparison of differences in functional traits occurrences between each pair of zonal forest plant communities (ZFPCs). Statisti- cal signifi cance established using Kruskal-Wallis test by ranks and median, except for R, H and G, for which the Scheffé post hoc test for a normal distribution was used. Community numbers correspond to those used in Tab. 1 and Fig. 1. FT – Functional type; C – Competi- tors; S – Stress tolerators; R – Ruderals; P – Phanerophytes; NP – Nanophanerophytes; Ch – Chamaephytes; H – Hemicryptophytes; G – Geophytes; T – Therophytes. *** p<0.001; ** p<0.01; * p<0.05. FT Compared pairs of ZFPCs 1– 2 1– 3 1– 4 1– 5 1– 6 1– 7 2– 3 2– 4 2– 5 2– 6 2– 7 3– 4 3– 5 3– 6 3– 7 4– 5 4– 6 4– 7 5– 6 5– 7 6– 7 C *** * ** * ** *** * S ** *** ** * ** *** *** * *** ** *** R ** *** ** P * *** * ** *** *** ** *** *** *** *** NP *** *** *** * *** Ch ** * ** *** * * ** H *** *** ** ** *** ** *** * *** *** *** *** G *** ** *** *** *** *** *** *** *** *** *** *** *** * T * ** *** ** *** ZONAL FOREST VEGETATION OF BOSNIA AND HERZEGOVINA ACTA BOT. CROAT. 76 (1), 2017 23 cens-Carpinus orientalis forests, Tab. 3, col. 2) is repre- sented by high, not degraded stands of zonal thermophilous deci duous forest of the association Querco pubescenti-Car- pinetum orientalis (Stupar et al. 2015) found in the low- lands and hilly area of sub-Mediterranean B&H. However, due to the negative human impact, they have mainly been replaced by the secondary scrub community Rusco acu- leati-Carpinetum orientalis (Muratspahić et al. 1991), while their present distribution area is restricted to small patches of mainly private old groves (Stupar et al. 2015). Mean annual temperatures are between 12 and 15 °C (Fig. 2), occupying elevations between 100 and 650 m. Diagnos- tic species are thermophilous and xerophilous plants of mainly S/SE European or Euri-Mediterranean distribution, although some widespread nemoral herbs and shrubs ap- pear with high frequency, e.g.: Brachypodium sylvaticum, Dactylis glomerata, Veronica chamaedrys, Crataegus mo- nogyna, Hedera helix etc., indicating more mesophilous microclimatic conditions under the closed canopy. Com- munity 3 (Quercus frainetto forests, Tab. 3, col. 3) is found in the eastern parts of B&H, where the infl uence of the con- tinental drier climate increases. This is a zonal community of Central Balkans lowlands and hilly areas (Horvat et al. 1974) and is represented by the fairly heterogeneous asso- ciation Quercetum frainetto-cerridis (Tomić and Rakonjac 2013, Stupar et al. 2015). Although we included in the anal- ysis only stands with 70% and more canopy cover, these forests are mainly found in the proximity of human settle- ments, so they are frequently degraded by grazing and browsing, as well as occasional fi res. Mean annual temper- atures are between 10 and 12 °C (Fig. 2), and elevations are between 150 and 700 m. They are dominated by a mixture of thermophilous, acid tolerant and widespread nemoral species, as well as some more light-demanding herbs, indi- cating wood pasture and cutting. Community 4 is repre- sented by sessile oak-common hornbeam forests, which are considered a zonal forest vegetation community for the area of NW Balkans (Quercus petraea-Carpinus betulus forests, Tab. 3, col. 4). In B&H they are found mainly in the hilly area of northern parts of the country but can also penetrate deeper into the south (e.g., Central Bosnian basin, Fig. 1). Habitats of these forests have been under intensive anthro- pogenic infl uence since the Neolithic (Horvat et al. 1974), so it is hard to fi nd stands of high, productive forests with Quercus petraea cover value in the tree layer more than 3 on the Braun-Blanquet scale (25–50% of cover) and more than 70% of overall canopy cover. They are often, due to bad management, structurally degraded to Carpinus betulus coppice. Mean annual temperatures are between 9 and 10 °C (Fig. 2) and they occupy elevations between 200 and 900 m. Several types of these forests occur in B&H but nei- ther the syntaxonomy nor nomenclature has been settled (Lakušić et al. 1978, Redžić 2007). In addition, there is dis- agreement about the syntaxonomical position of these for- ests on the regional level. While Croatian and Slovenian authors assign these forest to the Illyrian alliance Erythro- nio-Carpinion betuli (Vukelić 2012, Košir et al. 2013a), some other authors question this view, arguing that the group is weakly characterized by diagnostic species (Will- ner and Grabherr 2007, Borhidi et al. 2012). This is sus- tained by our results, in the sense that this ZFPC is the only one in B&H that completely lacks endemic Illyrian species (Tab. 3). Diagnostic species are mainly represented by me- sophilous European and Eurasian elements. Community 5 (pure Fagus / mixed Fagus-Abies forests, Tab. 3, col. 5) represents the fi rst altitudinal belt above oak forests and is made up of mesoneutrophilous montane pure beech, as well as mixed fi r-beech forests. The syntaxonomy and nomen- clature of these forests have not yet been the subject of thorough analysis but two main types, or widely understood associations, are included: Fagetum montanum illyricum and Abieti-Fagetum dinaricum (Beus 1984). They belong to the alliance Aremonio-Fagion (Marinšek et al. 2013). Mean annual temperatures are between 6 and 8 °C (Fig. 2) and they are mainly found at elevations between 700 and 1400 m. Floristically and ecologically similar to this type is Community 6 (mixed Picea-Abies-Fagus forests Tab. 3, col. 6), which also has a considerable amount of Boreal and south European orophytic elements, while annual mean temperatures are between 5 and 7 °C (Fig. 2). In the view of Beus (1984), which is supported by our results, the main difference between Community 6 and the previous commu- nity is that Community 6 has a higher proportion of spruce (100% frequency) and acidophilous Vaccinio-Piceetea spe- cies, while the proportion of mesoneutrophilous Aremonio- Fagion species is remarkably lower. In addition, it comes as an altitudinal belt above Community 5 (elevations between 800 and 1600 m) and can be found exclusively in the cen- tral range of Dinaric Alps, while Community 5 extends more to the north and south (Fig. 1). This is explained by the fact that the hot dry summers of the northern and south- ern chains of Dinaric Alps do not favor spruce (Beus 1984). The syntaxonomy and nomenclature of Community 6 have also not been settled, so in B&H this complex type is iden- tifi ed under the invalid provisional name Piceo-Abieti-Fag- etum. Community 7 is the highest altitudinal belt, repre- sented by subalpine Fagus sylvatica forests (Tab. 3, col. 7). This is yet another group the syntaxonomy and nomencla- ture of which in B&H has not been analyzed and it is known by the generic name of Fagetum subalpinum s. lato. These forests belong to the suballiance Saxifrago rotundifoliae- Fagenion of the alliance Aremonio-Fagion (Marinšek et al. 2013). These are mainly mesoneutrophilous pure beech for- ests found mainly on the mountains of the central and southern chains of the Dinaric Alps (Fig.1). Mean annual temperatures are between 5 and 6 °C (Fig. 2), with eleva- tions between 1400 and 1800 m. This community harbors the highest proportion (32%) of Boreal and south European orophytic elements. Functional analysis of ecological strategies showed small overall differences among ZFPCs. Based on the com- munities’ locations on the CSR triangle, competitive strate- gy (C) is dominant in all seven communities, followed by CSR and SC strategies (Fig. 3). Considering the course of succession (Fig. 3a), site production is high (Grime 1974), while the importance of ruderality is insignifi cant, which indicates late stages of succession, as pointed out by Pré- vosto et al. (2011). The terminal stage of zonal communi- STUPAR V., ČARNI A. 24 ACTA BOT. CROAT. 76 (1), 2017 ties is supported by the position of the communities on the CSR plot when only the herb layer was analyzed (Fig. 3b). In this plot, stress tolerance becomes more important since shading and lack of nutrients in the herb layer coincide with the development of large long-lived forest trees (Grime 1977, 1988). Only Community 1 (Quercus ilex maquis), which has the largest share of ruderals, expresses condi- tions of moderate productivity and middle to late stage of secondary succession (maquis). The increased proportion of stress-tolerators in Community 1 and also, to a much lesser extent, in Community 2 is related to drought stress during the summer season in the Mediterranean limestone region of B&H. On the other hand, Community 4 (Quercus petraea-Carpinus betulus forests) has an increased propor- tion of competitors, which indicates higher productive ca- pacities. Also, this community lies in the middle of the macro-climatic gradient (Fig. 2), which altogether suggests the most favorable conditions for forest vegetation develop- ment in B&H. Changes in the life form proportions of different com- munities are expressed in the greater share of woody spe- cies in the more thermophilous vegetation types, which cor- responds with the statement that ‘the Phanerophyte is the plant type that belongs to warm regions’ (Raunkiaer 1934). Similar results were obtained in the study of the gradient from warm to mesic temperate forests on the Galičica mountain range in Macedonia (Čarni et al. 2016). The pro- portion of Geophytes is higher in Communities 5–7 (beech dominated forests), which have spring ephemerals (mainly bulbous Geophytes), also abundant in oak communities (Popović et al. 2016), in addition to harboring a consider- able number of non-ephemeral rhizomatous ferns and spe- cies from the family Liliaceae. This is congruent with the notion that rhizomatous Geophytes are adapted to life in re- gions with a severe unfavorable period (e.g., hard winters), but have at the same time a long period of vegetation (Raunkiaer 1934). The proportion of Therophytes is insig- nifi cant except in Communities 1 and 3, which can be ex- plained by water shortage in the conditions of harsher Med- iterranean and continental climates (Kavgaci et al. 2012, Raju et al. 2014). The same can be said for the proportion of Chamaephytes in Community 1. On the other hand, ac- cording to Bloch-Petersen et al. (2006) plant life forms dif- ferentiate not only due to climatic variations, but seem also to relate to humane disturbance and management. Prasad (1995) studied the effects of grazing on the plant species and life form composition and found that the percentage of Therophytes was higher on grazed areas than on protected areas, which refl ects the disturbance through grazing in Communities 1 and 3. To conclude, the results of our analysis support the hy- pothesis that zonal forest vegetation in B&H is an expres- sion of macro-climatic conditions and that there are no re- markable differences in CSR plant strategies among the ZFPCs in B&H (excluding Community 1). However, hav- ing in mind that the currently protected forest area in B&H is very small and covers only a few types of ZFPCs (Stupar 2011), in order to facilitate further research into the ecologi- cal, fl oristic and functional characteristics of ZFPCs there is a need for establishment of a national network of protected areas which would cover all ZFPC types, thus preserving the representative stands in as natural conditions as possible (Milanović et al. 2015). Particular emphasis should be giv- en to thermophilous forests, especially to Community 1 (stands of Quercus ilex) which should be converted from maquis to high forests. Acknowledgements The research was partly carried out by the BH-SI bilat- eral project »Vegetation of thermophilous oak forests of the Pre-Pannonian area of the Western Balkans« no. BI-BA/ 12-13-015. AČ also acknowledges the fi nancial support of the Slovenian Research Agency (P1-0236). We would like to express our gratitude to the Institute of Biology, Research Center of the Slovenian Academy of Sciences and Arts (ZRC SAZU) and Forestry Faculty, University of Banja Luka for logistic support. References Abbate, G., Iberite, M., Bonacquisti, S., Giovi, E., Iamonico, D., Scassellati, E., 2012: Taxonomical and chorological diversity of native woody fl ora of Italy at regional scale. Bocconea 24, 169–175. Adams, J., 2010: Vegetation-climate interaction. Springer Berlin Heidelberg, Berlin, Heidelberg. Beus, V., 1984: Vertical zonation of forests in the light of real and primary vegetation in Yugoslavia. Radovi ANUBiH 76, Odje- ljenje prirodno-matematičkih nauka 23, 23–32 (in Serbo-Cro- atian). Bloch-Petersen, M., Brandt, J., Olsen, M., 2006: Integration of European habitat monitoring based on plant life form compo- sition as an indicator of environmental change and change in biodiversity. Geografi sk Tidsskrift-Danish Journal of Geogra- phy 106, 61–74. Bohn, U., Neuhäusl, R. (eds.), 2004: Map of the natural vegetation of Europe. Landwirtschaftsverlag, Münster. Borhidi, A., Kevey, B., Lendvai, G., 2012: Plant communities of Hungary. Akadémiai Kiadó, Budapest. Braun-Blanquet, J., 1964: Pfl anzensoziologie, Grundzüge der Vegetationskunde, 3rd ed. Springer Verlag, Wien. Čarni, A., Matevski, V., Juvan, N., Kostadinovski, M., Košir, P., Marinšek, A., Paušič, A., Šilc, U., 2016: Transition along gra- dient from warm to mesic temperate forests evaluated by GAMM. Journal of Plant Ecology 9, 421–433. Černý, T., Kopecký, M., Petřík, P., Song, J.-S., Šrůtek, M., Valachovič, M., Altman, J., Doležal, J., 2015: Classifi cation of Korean forests: patterns along geographic and environmental gradients. Applied Vegetation Science 18, 5–22. Chytrý, M. (ed.), 2013: Vegetation of the Czech Republic 4. For- est and scrub vegetation. Academia, Praha (in Czech). Chytrý, M., Tichý, L., Holt, J., Botta-Dukát, Z., 2002: Determina- tion of diagnostic species with statistical fi delity measures. Journal of Vegetation Science 13, 79–90. Dengler, J., Jansen, F., Glöckler, F., Peet, R. K., De Cáceres, M., Chytrý, M., Ewald, J., Oldeland, J., Lopez-Gonzalez, G., Finckh, ZONAL FOREST VEGETATION OF BOSNIA AND HERZEGOVINA ACTA BOT. CROAT. 76 (1), 2017 25 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. Journal of Vegetation Science 22, 582–597. Diaz, S., Hodgson, J. G., Thompson, K., Cabido, M., Cornelissen, J. H. C., Jalili, A., Montserrat-Martí, G., Grime, J. P., Zarrin- kamar, F., Asri, Y., Band, S. R., Basconcelo, S., Castro-Díez, P., Funes, G., Hamzehee, B., Khoshnevi, M., Pérez-Harguin- deguy, N., Pérez-Rontomé, M. C., Shirvany, F. A., Vendrami- ni, F., Yazdani, S., Abbas-Azimi, R., Bogaard, A., Boustani, S., Charles, M., Dehghan, M., de Torres-Espuny, L., Falczuk, V., Guerrero-Campo, J., Hynd, A., Jones, G., Kowsary, E., Kazemi-Saeed, F., Maestro-Martínez, M., Romo-Díez, A., Shaw, S., Siavash, B., Villar-Salvador, P., Zak, M. R., 2004: The plant traits that drive ecosystems: Evidence from three continents. Journal of Vegetation Science 15, 295–304. Dierschke, H., 1994: Pfl anzensoziologie. Eugen Ulmer Verlag, Stuttgart. Di Pietro, R., Viscosi, V., Peruzzi, L., Fortini, P., 2012: A review of the application of the name Quercus dalechampii. Taxon 61, 1311–1316. Drešković, N., Đug, S., Stupar, V., Hamzić, A., Lelo, S., Mu ra- tović, E., Lukić-Bilela, L., Brujić, J., Milanović, Đ., Kotrošan, D., 2011: Natura 2000 – Bosnia and Herzegovina. Centar za okolišno održivi razvoj, Sarajevo (in Serbian-Croatian-Bos- nian). Duckworth, J. C., Kent, M., Ramsay, P. M., 2000: Plant functional types: an alternative to taxonomic plant community descrip- tion in biogeography? Progress in Physical Geography 24, 515–542. Ellenberg, H., 2009: Vegetation ecology of Central Europe, 4th ed. Cambridge University Press, Cambridge. Ferrer-Castán, D., Vetaas, O. R., 2003: Floristic variation, choro- logical types and diversity: do they correspond at broad and local scales? Diversity and Distributions 9, 221–235. Gajić, M., 1980: Overview of plant species in Serbia with phyto- geographical indications. Glasnik Šumarskog fakulteta, Beo- grad A 54, 111–141 (in Serbo-Croatian). Grime, J. P., 1974: Vegetation classifi cation by reference to strate- gies. Nature 250, 26–31. Grime, J. P., 1977: Evidence for the existence of three primary strategies in plants and its relevance to ecological and evolu- tionary theory. The American Naturalist 111, 1169–1194. Grime, J. P., 1988: The C-S-R model of primary plant strategies – origins, implications and tests. In: Gottlieb L. D., Jain S. K. (eds.), Plant Evolutionary Biology, 371–393. Chapman and Hall, London. Grime, J. P., 2001: Plant strategies, vegetation processes, and eco- system properties, 2nd edition. Wiley, New York. Gunia, K., Päivinen, R., Zudin, S., Zudina, E., 2012: Forest Map of Europe. European Forest Institute. Retreived December 20, 2015 from http://www.efi .int/portal/members/membership_ service/benefi ts/forest_map_of_europe Hennekens, S. M., Schaminée, J. H. J., 2001: TURBOVEG, a comprehensive data base management system for vegetation data. Journal of Vegetation Science 12, 589–591. Hijmans, R. J., Cameron, S. E., Parra, J. L., Jones, P. G., Jarvis, A., 2005: Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology 25, 1965–1978. Horvat, I., 1933: Research of vegetation of mountains in Herce- govina and Montenegro. Ljetopis Jugoslavenske akademije znanosti i umjetnosti 46, 101–113 (in Croatian). Horvat, I., 1941: Research of vegetation of mountains Biokovo, Orjen and Bjelašnica. Ljetopis Jugoslavenske akademije zna- nosti i umjetnosti 53, 163–172 (in Croatian). Horvat, I., Glavač, V., Ellenberg, H., 1974: Vegetation Südosteu- ropas. Gustav Fischer Verlag, Stuttgart. Horvat, I., Pawlowski, B., 1939: Research of vegetation of moun- tain Vranica. Ljetopis Jugoslavenske akademije znanosti i um- jetnosti 51, 149–152 (in Croatian). Hunt, R., Hodgson, J. g., Thompson, K., Bungener, P., Dunnett, N., Askew, A., 2004: A new practical tool for deriving a func- tional signature for herbaceous vegetation. Applied Vegetation Science 7, 163–170. Juvan, N., Košir, P., Marinšek, A., Paušič, A., Čarni, A., 2013: Dif- ferentiation of the Piceetalia and Athyrio-Piceetalia forests in Slovenia. Tuexenia 33, 25–48. Kavgaci, A., Arslan, M., Bingöl, Ü., Erdoğan, N., Čarni, A., 2012: Classifi cation and phytogeographical differentiation of orien- tal beech forests in Turkey and Bulgaria. Biologia 67, 461– 473. Kilinç, M., Karavin, N., Kutbay, H. G., 2010: Classifi cation of some plant species according to Grime’s strategies in a Quer- cus cerris L. var. cerris woodland in Samsun, northern Tur- key. Turkish Journal of Botany 34, 521–529. Klotz, S., Kühn, I., Durka, W. (eds.), 2002: BIOLFLOR – Eine Datenbank zu biologisch-ökologischen Merkmalen der Ge- fäßpfl anzen in Deutschland. Bundesamt für Naturschutz, Bonn. Knollová, I., Chytrý, M., Tichý, L., Hájek, O., 2005: Stratifi ed re- sampling of phytosociological databases: some strategies for obtaining more representative data sets for classifi cation stud- ies. Journal of Vegetation Science 16, 479–486. Košir, P., Casavecchia, S., Čarni, A., Škvorc, Ž., Živković, L., Biondi, E., 2013a: Ecological and phytogeographical differen- tiation of oak-hornbeam forests in southeastern Europe. Plant Biosystems 147, 84–98. Košir, P., Čarni, A., Marinšek, A., Šilc, U., 2013b: Floodplain for- est communities along the Mura River (NE Slovenia). Acta Botanica Croatica 72, 71–95. Kovar-Eder, J., Kvaček, Z., 2007: The integrated plant record (IPR) to reconstruct Neogene vegetation: The IPR-vegetation analysis. Acta Palaeobotanica 47, 391–418. Kutleša, L., Lakušić, R., 1964: Flora and vegetation of Klek Pen- insula. Godišnjak Biološkog instituta Univerziteta u Sarajevu 17, 61–115 (in Serbo-Croatian). Lakušić, R., Pavlović, D., Abadžić, S., Grgić, P., 1978: Prodrome of plant communities of Bosni and Herzegovina. Godišnjak Biološkog instituta Univerziteta u Sarajevu 30, 5–88 (in Ser- bo-Croatian). Marinšek, A., Šilc, U., Čarni, A., 2013: Geographical and ecologi- cal differentiation of Fagus forest vegetation in SE Europe. Applied Vegetation Science 16, 131–147. Milanović, Đ., Stupar, V., Kulijer, D., Kotrošan, D., Hamzić, A., 2015: Natura 2000 in Bosnia and Herzegovina: where are we at the moment?. Glasnik Šumarskog fakulteta Univerziteta u Banjoj Luci 23, 95–134 (in Serbo-Croatian-Bosnian). Milosavljević, R., 1976: Climate of Bosnia and Herzegovina. Ph.D. Thesis, University in Sarajevo (in Serbo-Croatian). Muratspahić, D., Redžić, S., Lakušić, R., 1991: Association Rus- co-Carpinetum orientalis Bleč. et Lkšić 1966 in the Neretva River valley. Glasnik Republičkog zavoda za zaštitu prirode i Prirodnjačkog muzeja u Podgorici 24, 7–12 (in Serbo-Croa- tian). Passalacqua, N. G., 2015: On the defi nition of element, chorotype and component in biogeography. Journal of Biogeography 42, 611–618. STUPAR V., ČARNI A. 26 ACTA BOT. CROAT. 76 (1), 2017 Paušič, A., Čarni, A., 2012: Functional response traits and plant community strategy indicate the stage of secondary succes- sion. Hacquetia 11, 209–225. Pérez-Harguindeguy, N., Díaz, S., Garnier, E., Lavorel, S., Poort- er, H., Jaureguiberry, P., Bret-Harte, M. S., Cornwelf, W. K., Craine, J. M., Gurvich, D. E., Urcelay, C., Veneklaas, E. J., Reich, P. B., Poorter, L., Wright, I. J., Ray, P., Enrico, L., Pau- sas, J. G., de Vos, A. C., Buchmann, N., Funes, G., Quétier, F., Hodgson, J. G., Thompson, K., Morgan, H. D., ter Steege, H., van der Heijden, M. G. A., Sack, L., Blonder, B., Poschlod, P., Vaieretti, M. V., Conti, G., Staver, A. C., Aquino, S., Cornelis- sen, J. H. C., 2013: New handbook for standardised measure- ment of plant functional traits worldwide. Australian Journal of Botany 61, 167–234. Pignatti, E., Pignatti, S., 2014: Phytogeographical and ecological indicators. In: Plant life of the Dolomites, 631–648. Springer Berlin Heidelberg. Pignatti, S., Menegoni, P., Pietrosanti, S., 2005: Valori di bioindi- cazione delle piante vascolari della fl ora d’Italia. Braun-Blan- quetia 39, 1–97. Pipenbaher, N., Kaligarič, M., Mason, N. W. H., Škornik, S., 2013: Dry calcareous grasslands from two neighboring biogeograph- ic regions: relationship between plant traits and rarity. Biodi- versity and Conservation 22, 2207–2221. Popović, Z., Bojović, S., Matić, R., Stevanović, B., Karadžić, B., 2016: Comparative ecophysiology of seven spring geophytes from an oak-hornbeam forest. Brazilian Journal of Botany 39, 29–40. Prasad, S., 1995: Species composition, life form, biological spec- trum and importance value index of grasslands of Barbigha, Munger, Bihar. Environment and ecology 13, 309–312. Prévosto, B., Kuiters, L., Bernhardt-Römermann, M., Dölle, M., Schmidt, W., Hoffmann, M., Uytvanck, J. V., Bohner, A., Kreiner, D., Stadler, J., Klotz, S., Brandl, R., 2011: Impacts of land abandonment on vegetation: successional pathways in European habitats. Folia Geobotanica 46, 303–325. Pugnaire, F., Valladares, F. (eds.), 2007: Functional plant ecology, 2nd ed. CRC Press, Boca Raton. Raju, V. S., Krishna, P. G., Suthari, S., 2014: Environmental as- sessment of climate of a habitat through fl oristic life-form spectra, a case study of Warangal North Forest Division, Telangana, India. Journal of Natural Science 2, 77–93. Raunkiaer, C., 1934: The life forms of plants and statistical plant geography. Oxford University Press, Oxford. Redžić, S., 2007: The syntaxonomy of the vegetation of the conti- nental Dinaric Alps (W. Balkans). Collection of papers devot- ed to academician Kiril Micevski. Macedonian Academy of Sciences and Arts, Skopje, 249–280. Rodríguez-Rojo, M. P., Fernández-González, F., Tichý, L., Chy- trý, M., 2014: Vegetation diversity of mesic grasslands (Ar- rhenatheretalia) in the Iberian Peninsula. Applied Vegetation Science 17, 780–796. Rozman, A., Diaci, J., Batič, F., 2013: Functional analysis of veg- etation on alpine treeline ecotone in the Julian and Kamnik- Savinja Alps in Slovenia. European Journal of Forest Re- search 132, 579–591. Shipley, B., 2010: From plant traits to vegetation structure: chance and selection in the assembly of ecological communities. Cambridge University Press, Cambridge. Škornik, S., Hartman, K., Kaligarič, M., 2010: Relation between CSR functional signatures of dry grasslands from two con- trasting geological substrates. Annales, Series Historia Natu- ralis 20, 101–112. Stefanović, V., Beus, V., Burlica, Č., Dizdarević, H., Vukorep, I., 1983: Ecological and vegetational reonization of Bosnia and Herzegovina. Šumarski fakultet, Posebna izdanja 17, 1–49 (in Serbo-Croatian). Stupar, V., 2011: Nature conservation in Bosnia and Hercegovina – problems and perspective. Glasnik Šumarskog fakulteta Univerziteta u Banjoj Luci 15, 7–15. Stupar, V., Milanović, Đ., Brujić, J., Čarni, A., 2015: Formalized classifi cation and nomenclatural revision of thermophilous de- ciduous forests (Quercetalia pubescentis) of Bosnia and Her- zegovina. Tuexenia 35, 85–130. Surina, B., 2014: Forest vegetation of tectonic dolines Pihlja and Vitra above the Vinodol valley (Liburnian karst, NW Croatia). Šumarski list 138, 259–269. Tichý, L., 2002: JUICE, software for vegetation classifi cation. Journal of Vegetation Science 13, 451–453. Tichý, L., Chytrý, M., 2006: Statistical determination of diagnos- tic species for site groups of unequal size. Journal of Vegeta- tion Science 17, 809–818. Tomić, Z., Rakonjac, L., 2013: Forest phytocoenoses of Serbia: Manual for foresters, ecologists and biologists. Univerzitet Singidunum, Fakultet za primenjenu ekologiju Futura i Insti- tut za šumarstvo, Beograd (in Serbian). Tregubov, V. S., 1941: Les forêts vierges montagnardes des Alpes Dinariques – Massif de Klekovatcha-Guermetch: Étude Bota- nique et Forestière. Causse, Graille et Castelnau, Montpellier. Tutin, T. G., Heywood, V. H., Burges, N. A., Moore, D. M., Valen- tine, D. H., Walters, S. M., Webb, D. A. (eds.), 1968–1993: Flora Europaea, Vols. 1–5. Cambridge University Press, Cam- bridge. Véla, E., 2002: Biodiversité des milieux ouverts en région médi- terranéenne: le cas de la végétation des pelouses sèches du Lubéron (Provence calcaire). Ph.D. Thesis, University of Aix- Marseille III, Marseille. Vukelić, J., 2012: Forest vegetation of Croatia. Šumarski fakultet, Sveučilište u Zagrebu, DZZP, Zagreb (in Croatian). Willner, W., Grabherr, G. (eds.), 2007: Die Wälder und Gebüsche Österreichs: Ein Bestimmungswerk mit Tabellen (in zwei Bänden). Spektrum Akademischer Verlag. Zelený, D., Schaffers, A. P., 2012: Too good to be true: pitfalls of using mean Ellenberg indicator values in vegetation analyses. Journal of Vegetation Science 23, 419–431. Zelnik, I., Čarni, A., 2008: Distribution of plant communities, eco- logical strategy types and diversity along a moisture gradient. Community Ecology 9, 1–9.