Gradients in dry grassland and heath vegetation on rock outcrops in eastern Germany — An analysis of a large phytosociological data set

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
  The following study presents a statistical analysis of 595 relevés gathered during a survey of isolated grasslands and dwarf shrub heaths in the region of Halle, eastern Germany. Relevés were classified with UPGMA clustering based on a Bray-Curtis
  Folia Geobotanica 40. 341-356, 2005 GRADIENTS IN DRY GRASSLAND AND HEATH VEGETATION ON ROCK OUTCROPS IN EASTERN GERMANY AN ANALYSIS OF A LARGE PHYTOSOCIOLOGICAL DATA SET Karsten Wesche *), Monika Partzsch, Susann Krebes Isabell Hensen Institute of Geobotany and Botanic Garden, Martin-Luther-University Halle-Wittenberg, Am Kirchtor 1, D-06108 Halle, Germany: *) Corresponding author:fax +49 345 55 27228, e-mail Abstract: The following study presents a statistical analysis of 595 relevrs gathered during a survey of isolated grasslands and dwarf shrub heaths in the region of Halle, eastern Germany. Relevrs were classified with UPGMA clustering based on a Bray-Curtis distance matrix; the resulting classification was compared with a previously published phytosociological classification. Most of the 14 target communities were supported by both approaches, but in the case of communities on heavily disturbed sites cluster analysis corresponded to units on higher phytosociological evels alliances). DCA ordinations clearly revealed that nutrient and moisture availability controlled the community differentiation. Both factors were, however, highly correlated with the distance to the outcrop edge. Thus the outcrop size also influenced differentiation of communities, making larger outcrops richer in xerophytic and endangered species. This appears to be largely due to edge effects since the relationships diminished sharply when we controlled for progressively increasing distances from the edge in partial ordinations. Thus, microtopography of outcrops was the most important factor for community differentiation. It is therefore suggested that future research and conservation management should concentrate on the consequences of nutrient influx and changing land use on larger outcrops. Keywords: Classification, Edge effects, Ordination, Phytosociology Nomenclature: ROTHMALER et al. 2002) INTRODU TION Dry grasslands are of prime interest in European nature conservation SSYMANK et al. 1998). Their plant communities are often rich in species that are otherwise rarely found in the agricultural and forested landscapes. The eastern part of Germany is particularly well known for species of continental rather than submediterranean biogeographical affinities, many of them being endangered in Central Europe. In the eastern part of Germany, dry grasslands and dwarf shrub heaths are largely restricted to shallow soils on ancient sand dunes, calcareous sediments, and rock outcrops. North of the city of Halle, continuous erosion has exposed porphyry bedrocks at more than 200 sites, which are scattered as habitat islands in the otherwise almost completely farmed agricultural landscape Fig. 1). The typical plant communities occurring here have a high number of drought-resistant species, several of which are rare. A total of 37 vascular plant species found on the outcrops are considered endangered or critically endangered in Germany KORNECK et al. 1996).  342 K Wesche et al ~ Germany 11. Porphyry outcrop settlement road -- path Fig. 1. Location of the study region n the northern surroundings Halle, Saxony-Anhalt, Germany. As is the case of most grasslands in Central Europe, the larger part of the open vegetation on these outcrops is secondary in the sense that it srcinated due to the anthropo-zoogenic impact of traditional agriculture. Land use has changed tremendously in the last few decades and rendered usage of the outcrops economically unfeasible. This has subsequently induced changes in the vegetation and has allowed woody plants to invade the stands. Changes were accelerated by atmospheric nitrogen deposition (estimated at around 50 kg/ha'l yr-1, cf. PARTZSCH 2000) in the 1980s and early 1990s, when the region was one of the most heavily industrialized parts of the former of German Democratic Republic. The region around Halle is no exception, as traditional land use of dry grasslands and heaths for grazing and, locally, harvesting of hay and bedding for livestock, has decreased widely in Central Europe (POTT 1996). Similarly widespread are increased levels of atmospheric nitrogen deposition and not surprisingly, these two factors together are usually considered the most important threats for conservation of typical grassland vegetation (e.g. FISCHER STOCKL1N 1997, POSCHLOD et al. 1998). Many dry grasslands and heaths become increasingly fragmented, but the porphyry outcrops are extreme cases in this respect, forming small natural islands in a matrix governed by loess deposits. Thus, we find steep microtopographical gradients that allow assessing whether indeed large-scale phenomena such as land use changes or nitrogen deposition alone or effects related to small-scale topographic variation control vegetation patterns. Previous vegetation ecological research on the outcrops has concentrated on two main aspects, namely the effects of habitat isolation, and vegetation response to land use change and atmospheric pollution (synthesis in SETTELE et al. 1996, AMLER et al. 1999). Several studies dealt with the population ecology of rare plant species on these habitat islands (JACKEL 1999, DANNEMANN 2000), and with the principal dispersal modes (FISCHER et al. 1996, HENSEN MOLLER 1997, TACKENBERG 2001), but concentrated mainly on fragmentation and not on small-scale differences. Plant community ecology was mainly studied using traditional methods of phytosociology. Comprehensive survey data sets are available from the 1960s SCHUBERT 1960, MAHN 1965), and a second survey in the 1990s  Gradients in dry grassland and heath vegetation 4 S Porphyry outcrop N f71 1117 rFrl]rlT ... Slsymbno Faicano ConvovUOa~ens~sFestuco Ffltpendulo Gaho yen=- Festuco Tanaceto Atnollc turn vulgans- vulcan=- AgrostJetum Thyme- Euphorblo- 'tt~ e plant Agtopyretum Agropyretum Pea valestacae- He ctotnch tenu~s Festucetum Callunetum ruplcolae- Festuca Arrhenather- Arrhenather ntenss tepenhs angu~hfoha St~petum etum Huec~ 1937 cmefeae R Schube fJrachypodJ- rupJcola ~tum elat/ons etum Oberd ex community repen s Th Felferdy 1943 communPty capdfatae pratens,s etumpmnatr ¢ommunl{y Braun 1915 A F,schet M~hn et N Muller el Gor~ Mahn 1959 em Mahn Mahn 1959 1960 Mahn 1959 1985 Schubert 1969 Mahn 1965 1965 propos 1962 So=lone- St~ymbnon Planlag,n,. Festucton GemsOon CtrstoDannomcl- afflona s TX Festucfon classification Festucmn pallentts ptlosae BrachypOdton p~nnatt Arrhenathenon elahons et al ex yon alliance ConvolvuIo Agrepyr~on Gors 1966 va/eslacae ambJgu0us Passarge Khka 1931 DUVlgneaud Hada~ et Khka in Khka et W Koch 1926 Rochow 1951 Khka 1931 1964 co~ Zolyoml 1942 HadaC 1944 nora cons 1966 prop Calluno- Artemt~letea vulcan= Festuco Brometea Koeleno-Corynephoretea Uhcetea Festuco Brometea Mohnm Arrnenatheretea S~symbnetea lass Lohmeyer et al ex yon Rochow 1951 Br -BI el Tx in Br -BI 1949 Khka n KIika el Novak 1941 Br BI et Tx Br -BI et TX In Br -BI 1949 Tx 1937 Komeek 1974 ex We=theft etal 1946 edaphic conditions loess ;ubsoll loess loess porphyry porphyry loess (d=sturbed) aCdqC ~Oll ype Chemosem Para-rendzln~ Cam hi=el Syrosem Ranker Rendzma Chemosem ....................... t ................ s o° ~° ~ument content high ,ntermed,ate ow very low intermed,ate high very h,9h vegetation structure very high ~ow very dense dense dense enslty dense scattered very dense scattered zery scattered very dense dense ~pecles rlchnes low high very hlgtt low very ow yen/low low low leeght Of stand <80cm < 7ecru <3ecru < 15cm < (0era <20cm <40cm > 100cm > I(IUcm Fig. 2. Simplified pattern of plant communities on a typical large porphyry outcrop of medium age All communities analyzed in the present paper are listed except for the Cardario druhae Agropyretum TH. MULLER et GORS 1969 that occupies similar sites as the other two Agropyretum communities. Site conditions of the units are summarized according to PARTZSCI4 2000). made it possible to assess vegetation changes over the 40 year period (PARTZSCH 2000). Apparent changes included a general increase in vascular plant species richness due to the invasion ofruderal as well as adventive species (PARTZSCH MAHN 2001 ), but also a loss of several rare species. Phytosociological table work also revealed a general increase in the number of species and communities in colTelation to outcrop size and, to a lesser extent, outcrop age (PARTZSCH 2001, PARTZSCH MAHN 2001, PARTZSCH et al. 2003). Further habitat factors like relev6 position on the outcrop could not be analyzed simultaneously with age and size effects, since phytosociological table work poses strong restrictions to the number of analyzed dimensions, especially in large data sets. Here, statistical approaches offer more flexibility. The principal aim of the present study is to provide a statistical analysis of the factors driving community differentiation in dry grasslands, so a reassessment of the available  344 K Wesche et al phytosociological classification scheme was kept to a minimum. We opted for a simple standard method of numerical classification, which is also included in many freeware packages, and compared this to the available phytosociological framework (PARTZSCH 2000), thereby also checking the relevance of the previous analyses. Classification of plant communities served as a background for an analysis of the ecological factors that drive the differentiation of community mosaics. Size and age of outcrops as well as gradients in nutrient supply are of crucial importance in this respect. STUDY RE The study area is located in the eastern part of Germany (51°31'-51°35 ' N and 11°50'-1 l°56 ' E), at an altitude of 75-135 m. Mean annual precipitation is between 450-550 mm, and annual mean temperatures range between 8.5-9.5 °C (January: 0 °C; July: 18 °C). The geomorphology is defined by former glacial and periglacial activities, and by the erosive impact of the river Saale, which cuts through the lower third of the study area (Fig. 1). The bedrock is acidic porphyry, which is widely covered by slightly alkaline loess deposits, but these are eroded in several places, exposing isolated rock outcrops. Erosion has been accelerated by agriculture, so about 100 of the total of 204 outcrops studied have formed in the last 150 years (as far as what can be inferred from historical maps). Outcrop size varies from as small as 30 m 2 to 33,200 m 2. Edaphic conditions are characterized by patchy mosaics of acidic Rankers and Syrosems, acidic Cambisols, neutral to slightly alkaline (Para-) Chernosems and occasional alkaline Rendzinas. The edaphic and microclimatic conditions have resulted in a variety of small-scale vegetation mosaics, where communities of rather different biogeographical affinities (subatlantic, subcontinental, submediterranean) intersect. Fig. 2 summarizes the principal patterns and introduces the phytosociological units analyzed in the present paper. The region was settled as early as 6500 BP and agricultural use has intensified ever since, especially on the Chernosems. The fertile soils found over loess deposits have been cultivated with crops, whereas less productive sites have been used as grazing pasture for sheep. Such fanning practices have been (and still are) an important precondition for the existence of dry grasslands and dwarf shrub heaths. METHODS A total of 104 porphyry outcrops were visited in 1992-1996 and all visually discernible vegetation types were sampled. Tabular rearrangement of relev6s yielded a total of 50 plant communities (PARTZSCH KRUMBIEGEL 1996), of which 14 can be regarded as typical of dry grasslands and heaths; these were selected for the present study. The remaining communities included woody scrub and heavily disturbed stands dominated by annuals, which are of minor interest for nature conservation. The method followed the phytosociological approach with subjective placement of vegetation plots. Sampling was not random but rather comprehensive, and yielded a total of 595 relev6s with 322 vascular plant species for the 14 common vegetation types analyzed here (Fig. 2). Plot size was usually 16 m 2, but in small patches of grasslands on acidic soil Thymo-Festucetum cinereae) smaller plots had to be used in 8 cases. As these did not differ in species composition and richness  Gradients in dry grassland and heath vegetation 45 Table 1. Supplementary data available for the analysis of outcrop vegetation. Variables Units For relev6s Age class Position Size Distance to edge of outcrop Light Temperature Continentality, i.e. biogeographical affinities Moisture Soil reaction Nutrients Richness of vascular plants of plants typical of dry grasslands of red data list species For species Light Temperature Continentality, i.e. biogeographical affinities Moisture Soil reaction Nutrients 4-level ordinal scale (< 10; < 60; < I00; > 100 yrs) Categorial scale, north vs. south of the river Saale Raw values (m 2) transformed (y = log[x + 1]) Raw values in meters Mean Ellenberg indicator value (ordinal scale) Mean Ellenberg indicator value (ordinal scale) Mean Ellenberg indicator value (ordinal scale) Mean Ellenberg indicator value (ordinal scale) Mean Ellenberg indicator value (ordinal scale) Mean Ellenberg indicator value (ordinal scale) All species Species of Festuco Brometea and Koelerio Co~nephoretea According to red data list of Saxony-Anhalt Ellenberg indicator value (ordinal scale) Ellenberg indicator value (ordinal scale) Ellenberg indicator value (ordinal scale) Ellenberg indicator value (ordinal scale) Ellenberg indicator value (ordinal scale) Ellenberg indicator value (ordinal scale) (U-test richness; P = 0.4), small relev6s were not removed from the data set. Plant cover was estimated using the 9-level cover-abundance scale of Braun-Blanquet (including subdivisions 2m, 2a, 2b; REICHELT & WILMANNS 1973). To allow flexible data handling, initial Braun-Blanquet values were translated to metric figures of average cover prior to analysis (VAN DER MAAREL 1979). The survey character of the work restricted the range of supplementary data to the most easily assessed variables (Table 1). Multivariate classifications were based on a distance matrix calculated with the Steinhaus/Bray-Curtis coefficient, which is the quantitative version of the S6rensen coefficient; group average (UPGMA) clustering was used as it is the most widely applied hierarchical agglomerative clustering strategy. Many species of the dry grasslands and heaths are characteristic of these habitats but do not reach high cover, so the emphasis was put on small values by transforming all cover values to their square-root prior to analysis. Species that appeared in less than two relev6s (out of 595) were excluded. We plotted fusion levels against number of clusters (MCGARIGAL et al. 2000), and found a steep decrease of the curve in the 5 uppermost levels. Further smaller steps occurred at the 16th and 27th highest fusion levels. Thus, we selected a fusion level that resulted in 28 clusters, which were compared to a previously published phytosociological classification of the same data set (PARTZSCH 2000) by means of cross-tabulation.
Related Search
Similar documents
View more
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!