Diversity of wetland vegetation in the Bulgarian high mountains, main gradients and context-dependence of the pH role

Abstract

We fill a gap in understanding wetland vegetation diversity and relationship with environmental determinants in Bulgarian high mountains. A total of 615 phytosociological samples were taken from springs, mires, wet meadows and tall-forb habitats throughout Bulgaria, of which 234 releve´s are from mire and spring vegetation above timberline. The vegetation was classified by TWINSPAN and the resulting vegetation types were reproduced by the formal definitions using the combination of Cocktail species groups based on phicoefficient of joint co-occurrence of the species. Nine vegetation types of springs and fens have been clearly delimited above the timberline. All vegetation types include Balkan endemic species, the representation of which varies. Fens generally harbour more Balkan endemics than do springs, with the exception of speciespoor high-altitude Drepanocladetum exannulati. The gradient structure of the vegetation was revealed by DCA and by CCA with forward selection of environmental factors. The major determinants of vegetation variation strongly differ above and below the timberline and likewise between springs and fens. The baserichness gradient controls the floristic variation of Bulgarian submontane fens, whereas the complete data set including both submontane and subalpine fens is governed by the altitude gradient from lowland and basin fens to subalpine fens rich in Balkan endemics. When focusing on sites above the timberline only, the first DCA axis separates fens from springs without organic matter. The major species turnover in springs follows the variation in water pH and mineral content in water, whereas fen vegetation variation is primarily controlled by succession gradient of peat accumulation. Altitude remains an important factor in all cases. Weak correlation between water pH and conductivity was found. This correlation was even statistically insignificant in fens above the timberline. Water pH is not influenced by mineral richness in Bulgarian high mountains, since it is buffered by decomposition of organic matter in fens. In springs, pH reaches maximum values due to strong aeration caused by water flow. The plant species richness decreases significantly with increasing altitude. The increase of species richness towards circumneutral pH, often found in mires, was not confirmed in Bulgarian high mountains. The correlation between species richness and pH was significant only when arctic-alpine species and allied European high-mountain species were considered separately. The richness of boreal species was independent on pH. Some of them had their optima shifted to more acidic fens as compared to regions below the timberline. Our results suggest that subalpine spring and fen vegetation should be

Wet meadows of the alliance Molinion and their environmental gradients in Slovenia

“…2+ , pH and EC could be used to characterize the acidity-alkalinity gradient, which had a primary role in segregation of the plant communities and of these pH also explains the largest part of the vegetation variation (Hájková et al 2006). Other important gradients for the wet meadow vegetation, beside the acidity are humidity and nutrient gradient (Tables 2, 4).…”

“…Wet meadows and other similar wetland-types have been the object of several studies throughout Europe in the last years (Hájek & Hájková 2004;Havlová et al 2004;Hölzel & Otte 2004;Hrivnák 2004;Botta-Dukát et al 2005;Grootjans et al 2005;Stančić 2005;Zelnik 2005a,b;Hájková et al 2006;Havlová 2006), since the threat to the biodiversity of these ecosystems is still increasing and numerous sites have been destroyed, respectively.…”

Mire vegetation of the Muránska Planina Mts — formalised classification, ecology, main environmental gradient and influence of geographical position

“…However, only half of the relevés were assigned to the associations by the formal definitions. This rate is somewhat higher than in several other studies that applied national definitions to the regional mire data sets Hájková et al 2006;Dítě et al 2007). Kuželová & Chytrý (2004) demonstrated changes in interspecific associations (i.e.…”

“…We accepted a characterisation of each vegetation type by the formal definition using the combination of defined species groups created by the Cocktail method (see Bruelheide 2000;Kočí et al 2003;Hájková et al 2006;Chytrý 2007) with the statistical tendency of joint occurrences in vegetation, calculated by the phi-coefficient (Chytrý et al 2002). In the first step, we used the defined groups of species with the statistical tendency of joint occurrences in vegetation (called sociological species groups further in the text) and formal definitions of mire associations created based on 4117 phytosociological relevés from the West Carpathians ; see Appendix 1 and 2).…”

Vegetation Sampling for Wetland Delineation: A Review and Synthesis of Methods and Sampling Issues

“…From a historical perspective, the relevé method of community classification (Mueller-Dombois and Ellenberg 1974) has strongly influenced notions of plot size, and variations on this approach are still widely used (Peet 1981;Wells 1996;Dierssen and Dierssen 2005;Hajkova et al 2006). Mueller-Dombois and Ellenberg (1974) indicated that the sample area should be large enough to contain all of the species belonging to a plant community and should be relatively uniform in total plant cover.…”