|
| Back to main conference page |
Abstracts |
The effects of gopher disturbance and litter accumulation on fine-scale pattern of functional groups and diversity in secondary succession
S. Bartha
Institute of Ecology and Botany, Hungarian Academy of Sciences, H-2163 Vácrátót, Hungary, E-mail: sanyi@botanika.botanika.hu
Fine-scale spatial patterns of small mammal disturbances and local
accumulation of plant litter were studied together with the spatial pattern of vegetation
in the early stages of succession based on a consequence of old fields in the Cedar Creek
Natural History Area, MN, USA. Six stands of age 1, 4, 6, 10, 14, and 18 years were
sampled, recording the presence of living plant species, the presence of local soil
disturbances by Pocket Gophers (Geomys bursarius), and the presence of local plant
litter accumulation in 10 × 10 cm contiguous quadrates along circular transects of 2400
units. Spatial dependence between animal disturbance, litter, and vegetation were analyzed
using information theoretic models across a range of scales from 0.1 to 25 m. Significance
tests were based on Monte-Carlo methods using the algorithms of random shifts.
Results show that frequencies of local Gopher disturbances and litter accumulations change
considerably in succession and they are negatively related both in space and time. Local
species density increases with increasing disturbance and decreasing litter effect. The
strength of this relationship varies between functional groups, and it increases and
become non-linear after 18 years. There are positive spatial associations between the
majority of subordinated species and Gopher disturbances, while dominant species are
negatively related. Fine-scale litter accumulation shows complex spatio-temporal patterns
of associations to functional groups that underlines the potential importance of the
quality of litter and the indirect effects. Among the significant relationships, there
were consistent negative associations between the litter and the sedges, the annual
grasses, some annual forbs, and all perennial forbs. Other functional groups (C3 and C4
perennial grasses, mosses, and the winter annual Fabaceae species) show
inconsistent patterns of associations. The overall number of significant associations
among the studied characteristics increases with the increasing field age suggesting
increasing spatial organization of these controlling factors.
Spatio-temporal contingencies in old-field succession: Conclusions from a long term comparative study
S. Bartha1, S.T.A. Pickett2 & M.L. Cadenasso2
1Institute of Ecology and Botany of the Hungarian Academy of Sciences, H-2163 Vácrátót, Hungary, E-mail: sanyi@botanika.botanika.hu; 2Institute of Ecosystem Studies, Millbrook, NY 12545 0129, USA
Classical theories of vegetation succession predict smooth, usually monotonous or unimodal trends of population and community level attributes over time. However, field tests often find complex, stochastic patterns, and sometimes the lack of predictability. We argue that discrepancies between theories and field patterns can be resolved if (1) theories are based on spatially explicit non-equilibrium dynamics and (2) the extent and resolution of field sampling designs are appropriate to represent the inherent spatio-temporal variability and contingencies of successional processes. To illustrate our position, we re-analyzed the data of the Buell-Small Successional Study, one of the longest continuous studies of old field succession. The study includes 10 fields, high resolution continuous data from 1958, ca. 500 species of diverse growth forms, and different field treatments and times of abandonment. Each field has a regular grid of 48 plots, permitting spatially explicit analysis of community structure and dynamics. We calculated community characteristics (e.g. total cover, species richness, evenness, turnover rates) and the abundance of functional groups at the local quadrate scale at each sampling date. Local characteristics were arranged into spatio-temporal matrices which showed spatial and temporal neighborhood relationships. Auto- and cross- correlation within and between the spatio-temporal matrices were analyzed using models and randomization tests. While community characteristics vary at the local scale, considerable spatio-temporal dependence and organization is shown if spatio-temporal contexts are considered. We found significant spatial and temporal contagion of community characteristics but with variable intensities over time. Local characteristics (e. g. local total cover and local species richness) were often correlated, but strength and direction of correlation varied with field age, history, and vegetation context. Recovery of total cover after severe drought was a function of the local species richness. These results suggest that community characteristics are functionally correlated, i. e. the state of one characteristic affects the dynamics of another.
Changes of coarse scale compositional pattern diversity during old-field succession
Z. Botta-Dukát, Z. Molnár & A. Csecserits
Institute of Ecology and Botany of the Hungarian Academy of Sciences, H-2163 Vácrátót, Hungary, E-mail: bdz@botanika.botanika.hu
"Pattern diversity is a measure of the relative arrangement of subunits within an ecological unit" (Scheiner 1992). It is obvious that this definition should be specified. For this purpose the following questions must be answered:
Based on the third standpoint, Scheiner (1992) distinguished three types of pattern divesity: "spatial pattern diversity" is the arrangement of subunits in real space, "temporal pattern diversity" is the arrangement of subunits in time and "compositional pattern diversity" is the arrangement of subunits in the topological space of species. There are many methods for the analysis of spatial pattern, and many case studies were completed. However, only few attempts have been made on compositional patterns at coarse scale. Information theory functions developed by Pál Juhász-Nagy describe compositional pattern of a relevé set. Generally, different plot sizes (spatial series) are used and the changes of values in this spatial series are the characteristic features of spatial pattern. In the case of coarse scale studies the number of plots are limited, so some functions can not be estimated accurately. In this study the ecological units were old-field parcels, the subunits are plots with 1 m2 size. The arrangement of plots in the topological space was analysed by information theory functions. These functions relates to Juhász-Nagy's, but only about 50-100 plots are required for their estimation. The observed values were compared with expected random values. We found that:
Secondary succession on old fields on sandy soil
A. Csecserits1, Z. Molnár2 & T. Rédei1
1 Department of Plant Taxonomy and Ecology, Eötvös Loránd University, Budapest, Hungary, E-mail: picok@ludens.elte.hu; 2 Institute for Ecology and Botany, Hungarian Academy of Sciences, Vácrátót, Hungary
The return of the semi-natural vegetation is usually a slow process in abandoned old-fields (several hundred years). It is faster (20-30 years) in the highly abiotically controlled vegetation of sandy soils. We studied the changes in the number and dominance of the species and the appearance and disappearance of species during the succession. In Hungary, especially in the region between the Danube and the Tisza there are many abandoned oldfields and their number will increase as a consequence of Hungary's acceptance into the EU. Our sampling area was here, where we studied 54 neighbouring old-fields differing in the date of abandonment. We used the space-for-time substitution method, so we first had to specify the time of the abandonment, then the landscape-history of these old-fields with the help of old military maps and aerial photographs. On the basis of these we divided the abandoned old-fields into four age-groups, which are: 1-5 years old, 5- 10 years old, 11-23 years old and 24-33 years old. We made species-lists and estimated the dominance of the species using the old-fields as units. We made groups from the species two different ways. The first classification was made on the basis of the coenological status and social behaviour of the species. The categories are the following: generalists, specialists and weeds. The second classification was made on the basis of the physiological attributes. The categories are: annual, biennial, perennial and woody species. Our results show that the total number of species did not change between the four age-groups. This means that the rate of the extinction and the immigration was equivalent on this scale. The number of generalists species did not change, the number of the specialists increased between the first and the second age-groups and the number of the weeds decreased between the first and the second age-groups. The dominance of the different coenological classes changed in the following way: the generalists did not change, the dominance of the specialists increased between the first and the third age-groups and the dominance of the weeds decreased between the first and second and also between the second and the third age-groups. According to the classification of physiological attributes we found the following changes: the dominance of the annuals decreased between the first and third age-groups, the biennials decreased between the first and second age-groups, the perennials increased between the first and second age-groups and the dominance of the woody plants increased between the first and second age-groups. We conclude that the species composition stabilized first, and the stabilization of the dominance proportions was slower.
Importance of plant cover in reduction of nutrient losses from soil during secondary succession in deforested areas
K. Fiala1, I. Tůma1, P. Holub1, M. Tesařová2 & J. Jandák2
1 Institute of Botany, Academy of Sciences of the Czech Republic, Department of Ecology, Poříčí 3b, CZ-603 00 Brno, Czech Republic, E-mail: fiala@brno.cas.cz; 2 Institute of Pedology and Microbiology, Mendel University of Agriculture and Forestry, Zemědělská 1, CZ-613 00 Brno, Czech Republic
Environmental disturbance is usually accompanied by high losses of nutrients from ecosystems. However, rapid formation of vegetation cover during secondary succession tends to reduce this loss of nutrients. After forest decline, caused by air pollution, intensive leaching of nutrients from the soil of deforested sites occurs. We hypothesize that grass stands, which spread on deforested areas and represent the first stages of secondary succession on these disturbed sites, reduce losses of nutrients from ecosystem and improve features of soil environment. Therefore an experimental approach was applied in order to study losses of nutrients from soil of deforested sites affected by air pollution in the Moravian-Silesian Beskydy Mts (Czech Republic). The experiment involved transfer of soil blocks from adjacent forest stands to deforested sites fully exposed to pollution. The soil blocks were inserted into plastic boxes. The boxes were planted with young plants of Calamagrostis arundinacea and C. villosa in order to simulate the first stage of secondary succession. Several boxes were modified to lysimeters and soil percolates were collected for assessing by chemical analyses the leaching of nutrients and labile aluminium. Soil samples were also taken and analyzed. Data obtained in the first year of grass stand formation have shown a considerable changes in soil environment. The number of soil microorganisms increased by 30 % in variants with grasses in comparison to bare forest soil. Intensive growth of grasses followed by development of populations of soil microorganisms resulted in a decrease of the amount of nitrogen leached from soils, especially nitrates. In the course of the second growing season, 19.2 kg of N per hectare was leached from bare forest soil, whereas in variants with grasses the amount represented only about 5 kg of N per ha. In forest soil transferred to deforested site, the content of mineral nitrogen was nearly two times greater than in variants with grasses. Above all, the amount of nitrates increased four and three times in bare soil at the end of the first and second vegetative seasons, respectively. In the summer of the first year, during the maximum development of grasses, pH of percolates was more than one grade higher (mostly over 4.5 pH) and data obtained in the second year of this field experiment have shown that the average pH values of percolates in variants with grasses were higher by 0.5 pH units than those in controls (bare forest soil). Our data also indicated the reduction of leaching of base cations (Ca, Mg) from soil. The amount of leached bivalent cations (in g/l) was three to four times lower in comparison with percolates from bare forest soil. In the second year, losses of Ca were more than twice as low (about 9 kg/ha) than in the control variant without grasses. Therefore an excess of labile Al was substantially eliminated in variants with growing grasses. Our data thus confirm the importance of grass cover formation during secondary succession on deforested areas for (i) the reduction of nutrient losses from forest soil and (ii) the elimination of negative processes associated with soil acidification.
Spontaneous establishment and further development of reed communities in surface mining areas
A. Grüttner
Institute of Geobotany, Martin-Luther-Universität, Neuwerk 21, D-06110 Halle, Germany, E-mail: gruettner@botanik.uni-halle.de
Reed habitats resulting from open cast mining differ largely in extent and character. Hence the patterns and rate of succession in such newly established wetlands are of a great interest. Basically, three cases must be distinguished, i. e. (1) stagnant waters (from puddles to lakes) with rising levels, (2) stagnant waters with dropping levels, and (3) slopes or embankments which are continuously soaked by spring water. Each case is documented by an example from Germany. Case 1: "Paupitzschsee" - brown coal mining area Goitsche, N Leipzig: Drowning Phragmites stands situated at a former shore line survived and produced seeds over several years in spite of slowly increasing water level. The actual shore line provided numerous safe sites for germination and primary establishment of reed species. In addition to juveniles resulting from seed germination, sprouting shoot fragments of Phragmites were also found. Typha latifolia was able to germinate and establish under the water surface, and resulting juvenile plants were grazed heavily by waterfowl. Although hundreds of individuals per m2 were recorded, no permanent establishment was observed. Case 2: kaolin pit "Röblingen", W Halle: A new shore was exposed gradually. Quick colonization took place by (a) germination of several reed species, and (b) runners produced by Phragmites stands growing on the previous year shore line. These runners were several meters long and gave rise to vertical shoots in the same year. A closed Phragmites canopy was thus created very quickly by clonal growth. The sexual offsprings of Phragmites and other reed species therefore had a very low probability of survival and reproduction. Case 3: North slope "Cospuden" - brown coal mining area, S Leipzig: This example refers to a newly created slope with no reed communities established at its borders. Establishment by seed was observed in several reed species. Individuals of Typha latifolia outnumbered other species by far. The development of all individuals was followed over 3 years. Proportional cover of Phragmites increased from year to year. Especially one individual expanded rapidly by forming long runners. It can be concluded that in all cases reed species were able to colonize rapidly former mining areas. Runner formation of Phragmites australis was often decisive for vegetation development. Spontaneous formation of reed stands is very fast but depends, at the beginning, on the distance from the potential source of diaspores. With an exception of Case 3, the spontaneous establishment guarantees a high genetic diversity, from which the clones best adapted to the prevailing conditions can be naturally selected. This spontaneous process is considered to be advantageous in comparison with reed initiation by planting of cuttings. In lakes, which are attractive to water fowl, temporary prevention from grazing seems to be necessary for the establishment of new reed zones.
Primary succession on an exposed river bed
I. Hahn1, A. Gergely2, R. Mészáros-Draskovits1 * S. Barabás3
1Department of Plant Taxonomy and Ecology, L. Eötvös University, Budapest, Hungary, E-mail: hahn@ludens.elte.hu; 2Botanical Department, Hungarian Natural History Museum, Budapest, Hungary; 3 Ecological and Botanical Research Institute, Hungarian Academy of Sciences, Vácrátót, Hungary
Following the installation of the reservoir at Cunovo, the water level in the main Danube channel fell roughly by 3 m. In 1994 two 50-meter-long transects running from the current shoreline to the former riverbank were set up in the newly exposed riverbed at Dunaremete (1825 km). Both transects contained 25 adjacent quadrats of 2 × 2 m size. The percentage cover of each plant species was estimated in the quadrats every year in June. The elevation difference between transect endpoints was ca. 3 m. At the beginning of our studies, coarse gravel dominated the sediment at the lower end of the transects, while in the upper sections sand and fine silt deposited on the gravel were characteristic. By 1998, due to a periodical flooding silt covered the full length of the transects. Initially (1994-1997) the habitat properties were studied in the transects by using various species attributes (e. g. phytosociological classification, nature conservation value, ecological indicator values - mostly water requirement). Results suggest very fast, but spatially unbalanced succession along a gradient. The underlying environmental gradient is clearly described by the frequency distribution of the species' water requirement classes, although other - mostly pedological - aspects of the ongoing ecogenesis (e.g. soil physical structure, N content, seed bank) are also examined. By the fifth year after the water level drop (1998) the vegetation showed a marked zonation. In the lower one third of the transects a 2-4 m high, two- layered willow thicket established. In the middle, a tall-herb vegetation of 1,5 m in height developed, while further away from the current shoreline a closed grassland with numerous xerophilous species and weeds are characteristic. Species number declined considerably, most markedly in the lower willow thicket zone. Nevertheless, the fewer species produced greater ground cover reaching 100 % by the third year (1996). No unambiguous temporal trends in species diversity were revealed. Canonical correspondence analysis clearly shows the influence of terrain on the vegetation development. Starting from an early undifferentiated stage, quadrats become arranged according to their elevational position. In this context the axis associated with elevation can also be considered as a time axis.
Possible role of seed bank in the restoration of open sand grassland on old fields
M. Halassy & K. Török
Institute of Ecology and Botany, Vácrátót, Hungary, E-mail: hmelinda@botanika.botanika.hu
The process of succession, in general, is initiated by a disturbance (the nudation process with the Clementsian term). Some organism - termed residuals - might survive the disturbance event. Others, mainly nearby present or highly mobile species that can easily reach the disturbed site, join residuals in time. New colonists and residuals form the foundation of any developing community. The establishment of species by natural processes tends to be slow. One of the hindering factors of vegetation development is a slow colonization. Species colonizing particular site can establish from the seed bank or arrive from the surrounding areas. Comparing the vegetation and the seed bank of a particular site can lead us to a better understanding of the colonization process and thus help in restoration efforts. The vegetation and seed bank of an old field in the Kiskunság National Park, where cultivation stopped about 5 years before the study, were compared. Sampling was carried out along 5 parallel transect (each 40 m long). Vegetation cover was estimated in 1 m2 quadrats in 5 m interval, and a mixed soil sample was taken from the four sides of each quadrat (1 cm3). Reference vegetation and soil samples (15 each) were also taken in an open sand grassland patch. Soil samples were sieved and seeds identified. Almost all characteristic species of the open sand grassland community were established in the studied oldfield. However, all of them occurred only in a very limited number, except for Stipa borysthenica, a dominant grass species of the natural community (its cover reached 10 % and its frequency 71 % as maximum). Probably because of this limited occurrence, distance dependence was not provable by multivariate analyses. This supports the common opinion that the establishment of species by natural processes tends to be slow and stochastic. Similarity between the vegetation and the seed bank was low, as the latter was dominated by annual weedy species. Our study confirms that the role of seed banks in the restoration of grasslands is very limited as grassland communities usually do not form persistent seed banks. Propagules arriving from the surrounding vegetation and seeds of species already established can have major influence on colonization of particular site, and thus on the restoration process.
Blasts from the past: Re-establishment of chalk grassland disturbed during military training
R.A. Hirst1, R.F. Pywell1, P.D. Putwain2 & R.H. Marrs2
1 Institute of Terrestrial Ecology, Monks Wood, Abbots Ripton, Huntingdon, Cambs., PE17 2LS, United Kingdom, E-mail: RAHI@wpo.nerc.ac.uk; 2Applied Ecology Research Group, School of Biological Sciences, University of Liverpool, Liverpool, L69 3BX, United Kingdom
Salisbury Plain Army Training Area (SPTA) in southern England is the largest military training area in the UK covering some 38,000 hectares. The sole use of this land for military training since the end of the 19th century has protected the grasslands from many of the ecologically damaging impacts of modern intensive agriculture, and SPTA now contains around 40% of Britain's unimproved chalk grassland. Although around 20,000 hectares of the training area have been designated as a Site of Special Scientific Interest (SSSI), it is highly vulnerable to disturbance from heavy armoured vehicles. The historical aerial photographic record for Salisbury Plain, dating back to 1945, was used to select a suite of chalk grassland sites across SPTA with known disturbance histories. A field survey of the current status of these sites was executed, including the collection of vegetation and soil chemistry data and a number of other environmental variables such as aspect and slope. Ordination techniques were employed to characterise typical succession trajectories of the grassland following disturbance. Preliminary results show the background change in vegetation communities over this time period due to changing management regimes, and suggest that a heavily disturbed tank track takes approximately 30 years to re-establish a vegetation cover similar to the area surrounding it. These data will be instrumental to the Ministry of Defence in the development of sustainable management plans for SPTA.
Species spontaneously arrived after an artificially-induced vegetation development on mine spoil
M. E. A. Jochimsen
Plant Sociology and Ecology Department, University of Essen, FB 9, Universitätstr. 5, D-45117 Essen, Germany, E-mail: maren.jochimsen@uni-essen.de
In Germany, heaped-up mine spoil from deep coal mining has to be reclaimed due to legislation. This means the mounds must be integrated into the natural environment. From an ecological point of view, it seems reasonable to imitate and accelerate natural succession, as it follows the present site conditions and leads to sustainable plant communities. Species that are able to colonise the mine spoil spontaneously are nowadays rare in the countryside because of extinction. Therefore, we created seed mixtures of ruderal plants according to phytosociological relevés done in the past. The evaluation started in 1980 on the mine spoil heap Zollverein, and was repeated on Ewald (in 1982) with an enlarged number of fertilizers and/or other amendments to the soil. Furthermore, the seed mixture was improved. Each plot consisted of a circle enclosing 1 square metre, and the developing vegetation was assessed by using the scale of relative abundance (Jochimsen 1970). The 3rd site was also situated on a virgin mine spoil (Waltrop, since 1987). Though the research fields are larger there (116 m2), the estimated quadrates also comprise 1 m2, but the seed mixture is only composed of species belonging to the phytosociological alliance Dauco-Melilotion. In the control plots, the bulk of spontaneously arrived species originated from the seeded plots in their immediate neighbourhood. The other part of the immigrants was characterized by their indicator values (according to Ellenberg). It revealed that there are not only particular species groups invading the site, but also special types of resources utilized.
Managing succession for floodplain grassland restoration
Chris B. Joyce
School of the Environment, University of Brighton, Cockcroft Building, Lewes Road, Brighton, BN2 4GJ, United Kingdom, E-mail: C.B.Joyce@bton.ac.uk
Ecologically valuable floodplain grasslands are maintained by regular appropriate vegetation management, usually mowing and/or grazing, and are threatened both by changes in management and by abandonment. This paper describes the short-term succession of different floodplain grassland plant communities following the cessation of management, and examines the influence of particular vegetation management activities on community dynamics, in order to inform restoration of the floodplain grassland resource. The study was undertaken at two sites along the River Trent in the English Midlands: one a grazed inundation community (Gunthorpe) and the other characteristic of traditional hay-meadow management (Besthorpe). Plant species presence and cover at both sites was monitored in three successive growing seasons in one set of permanent quadrats subject to ongoing management and in another set in which no management activity took place. Results of ordination analysis and graphical comparisons demonstrated that grazing at Gunthorpe favoured species diversity and a dynamic community of short-lived small ruderal species that responded rapidly to perturbations. At Besthorpe, the long history of low-intensity management was expressed in a diverse community composed of perennial competitive stress-tolerating species that showed some inertia to management cessation. However, abandonment at both sites encouraged the expansion of robust competitive species, although the rate of successional change was greater in the community of short-lived plants. It is recommended that the different responses by specific plant communities to neglect and the implementation of management be considered in proposals for the restoration of floodplain grasslands.
Initiated and spontaneous vegetation development on slope sites in former brown-coal mining areas
A. Kirmer
Hochschule Anhalt, Fachbereich 1, Strenzfelder Allee 28, 06406 Bernburg, Germany, E-mail: Kirmer.Bamberger@t-online.de
In the lignite mining area "Goitsche" (Germany, North of Leipzig) spontaneous and initiated successional processes were examined. In 1994, we established permanent plots in sites on unvegetated slope composed of quaternary and tertiary sandy substrates to obtain information about successional dynamics in relation to different abiotic (substrate, pH) and biotic (diaspore input) factors. To accelerate the vegetation development, initial settings by deposition of plant cuttings from psammophyte and meadow communities were carried out. Vegetation analysis was made in all sites from 1995 to 1998. Dependent on the site conditions and the used material we could, for example, establish species-rich psammophyte communities (on quaternary sandy substrate) or initiate the development towards pioneer forests (on mixed tertiary and quaternary substrate). Course and duration of vegetation development on plots with depostition of plant cuttings were compared with untreated plots. To record and compare the overall influence of seed rain on the different slope sites, seed traps were used from July 1994 to July 1996. The results of these investigations show possibilities to influence progress and pace of succession and enable predictions about the vegetation development on sandy protosoils. Recommendations were made for site-specific initial settings when diaspore sources are not available or recultivation is required due to mining-related aspects.
Spontaneous development of nature in abandoned sedimentation ponds on toxic substrata: Hierarchy of questions for restoration ecology
P. Kovář
Department of Botany, Charles University, Benátska 2, 128 01 Prague 2, Czech Republic; E-mail: kovar@natur.cuni.cz
Study of abandoned sedimentation ponds on wastes deposited by a factory producing sulphuric acid from pyrite ore (Chvaletice) in comparison with ash-slag deposits from a power plant (Opatovice) brings results showing different ways, mechanisms and rates of vegetation succession. Biodiversity of particular groups of organisms exhibits important variations in time and space. The adaptive (genome) variability expressed by different ecological behaviour of a grass dominant (Calamagrostis epigejos) is mentioned. Toxicity-tolerant species and/or clones are naturally selected on the substrata characterized by significant heavy metal content and high salinity. Development of restoration ecology apparently needs to be (1) more experimental when we expect its higher credibility, and (2) more friendly to nature if we may support the nature-like functioning of (semi)artificial ecosystems.
Management of grasslands in the Krkonoe Mts: what do we know?
F. Krahulec
Institute of Botany, Academy of Sciences of Czech Republic, Průhonice, CZ-25243 Czech Republic, E-mail: Krahulec@ibot.cas.cz
Most of the meadows in the Krkonoe Mts are product of man activities. Only a small part of them may be considered as primary communities, occurring near the timberline and on the periphery of glacial cirques. There are rare mountain meadows on the Polish side of this mountain range, due to different land use in the past. Montane meadows are an important habitat for several rare alpine species (e.g. Campanula bohemica, Geum montanum, Viola lutea subsp. sudetica). The existence of montane meadows was possible only under their continuing use for cutting or grazing; only some of them were fertilized by dung in the past. Different habitat conditions and different management lead to the formation of many types of plant communities (the number of distinguished associations exceeds 20). The present state of the meadows is bad: traditional management mostly disappeared, nitrogen pollution changes the proportions of particular species and increased the losses of the kations. These circumstances influence us to look for new forms of management as sheep grazing and mulching protecting species composition at selected sites. Unfortunately, the representative network of sites is not complete; the detailed knowledge is available only for selected communities.
Sheep grazing as alternative management of mountain meadows
F. Krahulec1, V. Hadincová1, T. Herben1, R. Freiová1, S. Pecháčková1, H. Skálová1, I. áková 2 & M. Bílek 2
1 Institute of Botany, Academy of Sciences of Czech Republic, CZ-25243 Průhonice, Czech Republic, E-mail: Krahulec@ibot.cas.cz; 2 Institute of Animal Husbandry, CZ- 16000 Praha 6-Uhříněves, Czech Republic
Mountain meadows in the Sudeten Mts (Czech Republic) were traditionally managed by cutting; they were grazed by cattle in late autumn. These activities essentially disappeared following socio-economic and political changes after the Second World War; as a result, species richness of these grasslands declined and several dominant species increased. We tested sheep grazing as a potential alternative; careful approach had to be taken as the region is state-protected. The experiment took place in a mountain meadow (1200-1250 m) over eight years. Significant changes appeared after two years of grazing; it lead to suppression of the dominant herb, Polygonum bistorta. Grasses (Deschampsia cespitosa, Festuca rubra, Agrostis capillaris) increased. Several herbs increased as well often in spite of the fact that they were selectively grazed (Adenostyles alliariae, Melandrium rubrum); their increase was due to gaps created by sheep. The best way of management is the combination of sheep grazing and mowing that would suppress Deschampsia.
Saving the world in a day? Bog restoration in western and Central Scotland
A.R.G. Large
Department of Geography, Daysh Building, University of Newcastle NE1 7RU, United Kingdom, E-mail: A.R.G.Large@newcastle.ac.uk
Much of the 'bedrock' for our current scientific understanding of wetland habitats has come from long-term study of mires in north-west Europe and North America. In the United Kingdom however, scientific investigations have increasingly become 'locked-in' to the predominant pattern of research funding - currently no more than a three year period. Indeed, more and more, recommendations for restoration management for many sites of high conservation value are made on the basis of reconnaissance field visits and rapid 'condition surveys'. This paper presents survey results and recommendations for Coladoir Bog and Carnwath Moss, two sites in Scotland examined via such survey methods. It examines the recommendations made for each, and assesses their validity in relation to more intensive, longer-term studies. To achieve any sort of effective restoration, a range of attributes and features need to be addressed. While much good information exists for both mires, it is important to ensure that management provides good return for effort spent. For both these sites, mandatory targets include maintaining the existing area of the mire, and thus by definition the existing area of habitat. It is also considered mandatory to maintain the water table near the surface all year, as a water table to within a few centimetres of the surface is necessary for healthy growth of Sphagna. Such high water tables can adversely affect other species however, and water table monitoring will therefore be necessary to achieve sensitive management. While some of the efforts to enhance water levels on these sites have had decidedly mixed results from both a nature conservation and an aesthetic point of view, there is much of value in terms of site vegetation cover. The most valuable areas are undoubtedly the areas of 'active' bog, and these areas should serve as the foci for conservation. A disturbed community can be seen to have a number of recovery objectives. In terms of conservation management, the oft-quoted ideal or vision is to return a degraded site to its 'original' condition. In nature however, this is problematic given that, under normal ecological conditions, communities develop over time. Thus, what is envisaged as the 'original' community will, in reality, have altered to a different condition over the time that has elapsed since the original disturbance. By definition, through restoration objects can be restored to either their initial state, or to some former or antecedent condition. This distinction is particularly important for wetland sites subject, if not strictly to spontaneous succession, to natural developmental change. Finally, site-specific constraints offer a barrier to restoration success; integrated management should therefore take into account potential conflicts and aim at clearly specifying restoration priorities within sites.
The past we inherit - the future we build: Constructed wetlands for treatment of minewaters
A.R.G. Large, P.L. Younger & A.P. Jarvis
Department of Geography & Department of Civil Engineering, University of Newcastle, Newcastle Upon Tyne NE1 7RU, United Kingdom; E-mail: A.R.G.Large@newcastle.ac.uk
What can be done when new minewater discharges emerge? The answer is very little as it is difficult to catch the first 'flush' of pollution at a given site - and the first is often the worst. When strict discharge consents must be met by mine operators, active treatment is still the surest means of achieving highs standards of treatment; the relatively high costs of active treatment being offset against profits. However, active treatment is not an attractive option for long-term abatement of uncontrolled acid mine discharges from abandoned mines. Recently developed, low-cost, low-maintenance minewater treatment techniques may provide a solution. Early designs have aimed specifically at the nub of the problem i.e. the minewaters themselves. Now however, attempts are being made to integrate these features between into the surrounding landscape and to incorporate ecological function (including spontaneous ecological succession) as well as hydrochemical efficiency.
Factors controlling heathland vegetation change after treatment for bracken control
M.G. Le Duc1 & R.H. Marrs2
1School of Biological Sciences, 2Applied Ecology Research Group, University of Liverpool, Derby Building, Liverpool L69 3BX, United Kingdom, E-mail: mled@liverpool.ac.uk
An 18-year experiment designed to test a range of bracken control and heathland restoration treatments on a Breckland heath in south-eastern England is discussed. The treatments applied included cutting once or twice yearly, spraying with the herbicide asulam, and spraying and cutting combined. Calluna seeds were added to half the plots to aid heathland establishment. The experiment was split after six years and bracken control was continued/re-applied on only half the plots. The vegetation change was recorded in two stages, firstly in 1978-1986, when species biomass was sampled, and secondly in 1986-1996 when species cover was measured. Initially Calluna establishment was good in some treatments, especially where Calluna seed was added and bracken was controlled. Other plots developed either a grass heath flora dominated by Agrostis capillaris, Deschampsia flexuosa, Dicranum scoparium, Festuca ovina and Rumex acetosella or were dominated by clonal species, Calamagrostis epigejos or Carex arenaria. Unconstrained ordination showed that there were significant changes in the vegetation with time. Several treatments were important in influencing the vegetation, especially those involving asulam application. When variation partitioning was carried out using constrained ordination a different explanation emerged. During the first stage (1978-1986) the most important factors were the management treatments applied, and both time elapsed and spatial factors, with very little overlap between these variables in the amount of variation explained. In the latter period (1986-1996) time on its own became irrelevant. Effectively, all the variation explained could be accounted for by other variables .The most important of these were spatial effects, weather, the amount of bare ground caused by disturbance and the cover of bracken litter (an index of bracken recovery). The implications of these results in interpreting vegetation change are discussed.
Plant succession on lignite wastes in eastern Pyrenees
J.M. Ninot, P. Herrero, A. Ferré & R. Guardia
Department of Plant Biology, University of Barcelona, Spain, E-mail: ninot@porthos.bio.ub.es
This paper deals with the vegetation establishment on waste areas located in the montane and submontane belts of the eastern Pyrenees, where sub-Mediterranean oak woodland, Scotch pineyards and meso-xerophilous pastures make up most of the landscape. The waste areas consist of marl and lime regolith, very poor in nutrients and structure. They were mainly sown with pasture species and left to spontaneous recolonization with little or no further intervention. Forty vegetation samples were taken 10 years after abandonment, based on the point quadrat procedure. They were analysed on the basis of their species composition (ecological groups and multifactorial analysis) and structure aspects (coverage, diversity, etc.). Successional processes have led to various communities, from poorly covered surfaces formed by a mixture of opportunists and stress resistant species to dense grasslands heavily dominated by one (or few) competitive, introduced grass species. Only few areas seem to evolve towards "target communities", i. e. dense, diverse and mainly formed by spontaneously arrived species.
Different forms of spontaneous succession on fen areas within the Biebrza National Park
H. Piorkowski
Institute of Land Reclamation and Grassland Farming, 05-090 Raszyn, Poland, E-mail: H.Piorkowski@imuz.edu.pl
The Biebrza river valley is located in northeast Poland. It occupies approximately 600 km2 and was formed in Pleistocene on a water outflow from the glacier. Hydrogeological conditions, together with relief and climate contributed to forming of widespread complexes of fens temporary grouped in three basins: upper, central and lower. This unique hydrological system is the reason for specific specific features of the valleys at the periphery of the glacier. Human interest in Biebrza river valley goes as back to the Middle Ages. Because the valley is not easily accessible, its use was always extensive and adapted to natural hydrological rhythm. Fishery and farming were the most common occupations within local people. Bottom of the valley was used as pastures and hay meadows. High humidity of sites prevented forest and shrub communities from expansion, and together with mowing of meadows, sedge and reed beds enabled to maintain a specific open landscape. The cultural landscape of the valley made possible coexistence for many plant and animal species providing large biodiversity. For its high natural values of the landscape and abundance of fauna and flora, the area was claimed the Biebrza National Park (BNP) by the Polish Parliament in 1993. It is the largest national park in Poland. High rates of socio-economic changes in the 20th century, especially in its second half influenced the environment of natural valleys in a principal way. Successive emigration from the Biebrza region, abandonment of traditional landuse, and changes of water conditions catalysed successional processes and started shrub encroachment onto open wetlands. Spontaneous succession is nowadays considered the main problem in the national park. Salix, Betula and Alnus are the most expansive trees; Populus tremula and Pinus are less expansive, the latter only sporadically on transitional bogs. Different parts of the valley are colonized by different species. Shrub encroachment has various spatial pattern, depending on site factors and autecology of the present species. In succession dominated by Salix, younger stands form a thick front of trees penetrating from the edge towards the centre of the valley. Betula trees (single or groups) occur far from the edge of the valley. The succession with Salix appears to be the most dynamic. Trees and shrubs can penetrate into the wettest areas and form a thick cover gradually shifting into the centre of river valley where they serve as foci for further expansion. Special type of succession was observed in the vicinity of mineral islands. Expanding shrubs form concentric structures outside the core areas. Successional dynamics is principally affected by local factors. Comparison of shrub ranges on aerial photographs from 1962 and 1997 made it possible to distinguish areas with different rate of succession. Four main areas can be distingusihed in BNP, i. e. open wetlands extremely endangered by succession (the rate exceeds 10 m/year), open wetlands endangered by succession, areas with stable range of shrubs, and areas with regression. The BNP authoritiesare at the beginning of the long-lasting fight against expanding shrubs. Up till now, there were some short term works, experiments and projects focused on gathering information about the phenomenon as well as about methods of the shrub removal (mowing, grazing and burning). Preliminary holistic analysis of the problem shows that there is no simple universal solution to stop shrub encroachment in BNP.
Application of remote sensing and GIS for monitoring spontaneous vegetation succession in wetland areas: Scrub encroachment in the Biebrza National Park
H. Piórkowski1 & A. Schmidt2
1 Institute for Land Reclamation and Grassland Farming, IMUZ Falenty, 05-090 RASZYN, Poland; 2 DLO-Staring Centre and DLO-Institute for Forestry and Nature Research, P.O.Box 47, Wageningen, The Netherlands, E-mail: schmidt@sc.dlo.nl
Time series of satellite images (1987 and 1997) and aerial photographs (1962/1963 and 1997) have been used for monitoring scrub encroachment in the Biebrza National Park. The preliminary results are presented here together with some tentative conclusions. By means of visual interpretation and automatic classification of Landsat TM images of May 1987 and May 1997 the scrub encroachment in the Biebrza National Park has been mapped and monitored. Only the final stages of the scrub encroachment process could be identified on these satellite images. The early stages of the scrub encroachment process can be distinguished on a Landsat-TM image of August 1997. By means of visual stereoscopic interpretation of panchromatic photographs 1: 16 000 from 1962/1963 and true colour photographs 1: 20 000 from 1997 various stages of the scrub encroachment process have been mapped and monitored. Furthermore, a distinction has been made between different spatial patterns of scrub encroachment process and between different tree species. The results of both the aerial photograph and satellite interpretations of 1997 have been verified in the field. Both aerial photographs and satellite images are very useful for the identification of areas with scrub encroachment. Satellite images are very helpful for giving a general overview of the scrub encroachment process. The interpretation of satellite images is less time consuming than interpretation of aerial photographs and therefore less expensive. Because of automatic classification procedures the satellite interpretation is relatively objective. The aerial photographs are very useful for giving a detailed information on the scrub encroachment process. Different spatial patterns and differences in tree species can be distinguished. This gives more insight into the scrub encroachment process. A combination of both remote-sensing techniques is recommended. By comparing the satellite images with the aerial photograph and the field situation it seems that the more information on vegetation, humidity and possibly as well on soils can be derived from the image. This would give more insight into the underlying causes of the scrub encroachment process such as changes in groundwater table and mineralization of the soil.
Ecological restoration of mining wastes: From spontaneous succession to sustainable landscapes
P. Putwain, M. Le Duc & D. Fee
School of Biological Sciences, The University of Liverpool, United Kingdom, E-mail: philp@liverpool.ac.uk
During the 20th century quarrying for minerals such as kaolin (china clay), sand and aggregates has caused significant destruction of heathland landscapes in southern England. Similarly in north west England deep-mined colliery spoil has been tipped on a variety of seminatural habitats. Over a period of several decades spontaneous processes of plant species assembly have resulted in the development of heatland and woodland ecosystems on these mining and quarry wastes. Natural succession proceeds over many decades but often many heatland and woodland species characteristic of typical regional vegetation communities are missing from successional communities on mining wastes. Constraints on immigration of diaspores may be one reason for the species-deficient successional communities. Accelerated restoration of heatland and woodland vegetation, consisting of a characteristic assemblage of species appropriate to a local area, is the initial step in the eventual restoration of heatland and woodland ecosystems within a wider landscape mosaic. Deliberate intervention to introduce plant propagules and assist plant establishment is a key step in creative ecological restoration. We have undertaken creative restoration of heatland vegetation at several sites in southern England. Case studies will be used to illustrate techniques of restoration and to demonstrate trajectories of species assembly using ordination analysis of species abundance data. However restoration of heatland on mineral wastes will only be an appropriate landscape objective if there is a sustainable system of long-term management. This approach must be acceptable to local inhabitants and economically viable and presents a considerable challenge for the mineral industry, local authority planners and land managers. Other experiments have examined the potential to introduce late successional woodland ground flora species to early successional woodland on colliery spoil. Betula and Salix species may be amongst the primary colonies of colliery spoil and in less than 20 years provide a shaded environment suitable for the deliberate introduction of ground flora. Factors determining the success of initial establishment have been investigated. Our results demonstrate that accelerated creative restoration is a practical methodology for enhancing derelict landscapes.
Succession in man-made sites: the data and the theory
P. Pyek1 & K. Prach2
1 Institute of Botany, Academy of Sciences of Czech Republic, Průhonice, CZ-25243 Czech Republic, E-mail: pysek@ibot.cas.cz; 2 Faculty of Biological Sciences, University of South Bohemia, Braniovská 31, CZ-370 01 České Budějovice, Czech Republic; E-mail: prach@tix.bf.jcu.cz
Data on succession collated from 15 seres starting on bare ground in human-made habitats (i. e. sites disturbed by various mining activities, bulldozed sites, ruderal urban sites, exposed bottom of a destroyed water reservoir, and abandoned fields) located in the western part of the Czech Republic, Central Europe, are analyzed. Period for which the data on succession were available ranged from 12 to 60 years. In previous studies, succession over the range of habitats was compared from various aspects, such as (1) changes in total cover, (2) rate of succession, (3) changes in species traits during succession, (4) differences between primary and secondary seres, (5) role of clonal plants in succession, (6) role of woody plants, and (7) characteristics of dominant species. In the present contribution, the results are summarized and general trends are reported.
The establishment of trees on closed landfill sites for community forest use
H. Rawlinson1, N. Dickinson2 & P.D. Putwain1
1School of Biological Sciences, University of Liverpool, Nicholson Building, Liverpool, L69 3BX, United Kingdom; E-mail: hrawlin@liverpool.ac.uk; 2 School of Biological and Earth Sciences, Liverpool John Moores University, Bryom Street, Liverpool, L3 3AF, United Kingdom
The Mersey and Red Rose Forests are just two of the twelve community forest regions in England and Wales. Their regions cover a large area of the Northwest within which there is the legacy of over 1300ha of closed landfill sites. Many of these sites contribute little or no value to the local environment due to, their negative influence on the landscape and in some cases, ongoing pollution. All of these sites however, have the potential for community woodland developments that would bring benefits to the local environment, economy and community. The main aim of the research is to support the development of a cost-effective methodology for investigations before community woodland establishment on closed landfill sites. The intention is to determine 'occupational exposure limits' for tree survival in this highly variable environment. The research involves the intensive monitoring of 22 ha of trial woodlands on 11 sites across the Northwest of England. Sites were chosen on the basis that they were representative of a wide range of waste types, history and ownership's. The physical, biological and chemical characteristics of soil at each site will be studied and the influence on tree growth and survival will be assessed. Investigation into the role of landfill gas, soil structure and moisture availability on tree performance is also taking place. Fieldwork has been conducted to estimate tree mortality and growth in the first growing season and landfill gas emissions are being recorded. Operational problems were experienced when installing gas probes and alternative methods for obtaining this data have now been established. Initial tree failure was relatively low and a number of reasons were explored. The establishment of trial woodlands provides a useful legacy for further research as the woodlands develop. This study may also be incorporated in a national network of field study sites across the country that have been proposed by English Partnerships and Waste Disposal Authorities.
Land use and succession processes on loess-covered dolomite rocks in an old culture landscape near Budapest
T. Rédei1, S. Barabás1, A. Csecserits1& A. Kun2
1 Department of Plant Taxonomy and Ecology, Eötvös Loránd University, Budapest, E-mail: redy@botanika.botanika.hu; 2 Institute of Ecology and Botany, Hungarian Academy of Sciences, Vácrátót
Potential vegetation maps made with the methods of the classical vegetation science may strongly depart from the reconstructed original vegetation of the study area. The actual vegetation pattern of an intensively used landscape is created by human disturbances and natural succession. Drastic changes can occur both in the biotic and the abiotic components as a result of prolonged use. To analyse human impact, it is essential to have an idea about the original vegetation, which can only be reconstructed by using historical data. The following sources of information are available: 1. historical maps and written documents, 2. remnants of the original vegetation found in the floristic composition, species area, structure of plant communities and the landscape mosaic, 3. vegetation of similar habitats, 4. and the soil, which can stay intact long after the disappearance of the original vegetation. The poster aims at demonstrating the utility of the above mentioned methods through a case study. The studied landscape, the southern part of the Buda hills, has been intensively used over the past centuries. The scenery is mainly determined by low dolomite hills partly covered with periglacial loess. The hill tops are covered with species-rich dolomite vegetation, while the valleys are dominated by species-rich but fragmented and disturbed steppe grasslands and scrub. Woodlands are rare in the area. The lower sites are covered with buildings. Using the methods mentioned above we constructed the actual, potential and reconstructed vegetation maps of a part of the area and attempted to describe the past degradation and successional processes. We conclude that loess vegetation was more widespread over the landscape in the past, primarily represented by the mosaic of opening oak forests and steppes on loess. Early deforestation and persistent agricultural management caused great changes in the composition of vegetation. Intensive grazing caused erosion in loess cover, which led to the spread of secondary dolomite grasslands, forming the present scenery of the landscape.
Species richness across a secondary succession gradient in a dry heathland
I. Sedláková1, M. Chytrý2 & L. Tichý2
1Institute of Botany, Bělidla 4a, CZ-603 00 Brno, Czech Republic, E-mail: sedlak@brno.cas.cz; 2Department of Systematic Botany & Geobotany, Masaryk University, Kotlářská 2, CZ-611 37 Brno, Czech Republic, E-mail: chytry@sci.muni.cz
Secondary succession in a species-rich dry heathland was studied in the southern part of the Czech Republic over 7-year period following experimental sod- cutting, burning and cutting. Sod-cutting caused the most severe disturbance with removal of all above-ground biomass and topsoil; burning destroyed most of the above-ground biomass and exposed patches of mineral soil; cutting did not influence basal parts of plants and litter layer at all. Every year species composition was recorded in eight permanent plots using 25 × 25 cm grids. Species richness pattern across the successional gradient was analysed in plots of 0.0625, 0.25, 1.0 and 9 m2. All treatments resulted in an increase in species richness. In plots of 9 m2, species richness peaked 3-4 years after sod- cutting (maximum 39-50 species), 4-5 years after burning (max. 33-36 species) and 4-6 years after cutting (max. 31 species). In smaller plots, the species richness steadily increased over the entire observation period. Similarly as in 9 m2 plots, the highest maxima of species richness in smaller plots were associated with sod- cutting and lowest with cutting. The results suggest that (1) species richness in undisturbed heathlands is limited by light and availability of regeneration niches (exposed mineral soil); (2) after a disturbance, species richness first increases, but later decreases as the vegetation cover regenerates; (3) the more intense disturbance, the faster and higher the species richness increases; (4) on larger scales species richness peaks earlier than on smaller ones.
Vegetation changes in restoration of Central European alluvial meadows
J. Strakrabová (Müllerová)1 & K. Prach2
1Institute of Botany, Academy of Sciences of Czech Republic, Průhonice, CZ-25243 Czech Republic, E-mail: mullerova@ibot.cas.cz; 2Faculty of Biological Sciences, University of South Bohemia, Braniovská 31, 37005 České Budějovice, E-mail: prach@tix.bf.jcu.cz
Main transitions between dominant species observed in the process of abandonment and restoration of alluvial grasslands in Central Europe are reviewed. The authors' results and field experience as well as literature data and personal communication were used to construct the scheme. Main vegetation types of alluvial meadows classified according to dominant species were related to the soil moisture and nutrient content. This scheme was used to express main trends in dominant species exchange after the abandonment of alluvial meadows and possible restoration after the re-establishment of regular management (i.e. mowing). Degraded, regularly managed and/or restored vegetation types were characterized by average number of species per phytosociological relevé. Rare experimental data dealing with studied problematic indicate that the process of restoration can be relatively easy and fast.
Restoration of the Morava river floodplain meadows
J. effer & V. Stanová
DAPHNE Centre for Applied Ecology, Hanulova 5/d, 844 40 Bratislava, Slovak Republic, E-mail:daphne@changenet.sk
The Morava River floodplains are located on the western edge of Slovakia and situated along the former "iron curtain" which was closed up to 1990. Total area of the floodplain on the Slovak side of the river, recognized under the Ramsar Convention is 5,100 ha. There are various types of wetlands with the largest complex of floodplain meadows in Central Europe with an area of 2,823 ha. Meadows of the alliance Cnidion venosi are a result of the extensive flood regime of the territory and sustainable use by local communities for approximately 1,000 years. These areas provide necessary foraging and nesting habitat for numerous bird species. Serious threats to the meadow ecosystem biodiversity have been caused by the plowing of 500 ha of meadows during the communist regime 1960-1989. The plowing was concentrated on the middle section of the Ramsar site, and the arable land was regularly fertilized and herbicized. Some fields are still in use, but most were abandoned few years ago and are being invaded by weeds and alien species, especially Aster novi-belgii agg. Intensive use of chemicals caused an increase of pollution in the Morava River and a decrease of species richness. These fields have a low biodiversity and have become a barrier for nesting and migrating bird species. For example the occurrence of Corncrakes was not registered in that section of the river. The only method that has been used so far for meadow restoration in the region was sowing with a commercial mixture of hybrid species mainly grasses. The vitality of hybrid grass species is limited and they are not well adapted to floodplain conditions. Several projects focused on different aspects of floodplain meadow restoration are being carried out. In 1994, a project aimed at elaboration of sufficient method for restoration and management of meadows was started. We used permanent plots to test two techniques: sowing of natural seed mixture collected from quality grasslands and transplantation of turfs. The experimental design consisted of 2 treatments of 2 replicates each. The size of plots was 5 x 5 m. In 1994, 11 species were collected and sowed into the arable land in the density of 4 g.m-2 and 6 g.m-2, respectively After the first year, only 4 species started to grow. After four years, 9 species grew. There was no evidence of germination for two species - Sanquisorba officinalis and Serratula tinctoria. A total of 68 species grew after sowing of seeds during four years, of which 25 were ruderal species. There was a strong penetration of species from plots with transplanted turf. A total of 83 species grew after transplanting of turfs during four years; 25 of these were ruderal species. Dominant and subdominant species were highly successful in terms of survival. Species that adapted well were Poa pratensis, Alopecurus pratensis, Carex praecox, Centaurea jacea, Ranunculus repens and Glechoma hederacea. Most dominant ruderal species in both treatments were Aster novi-belgii agg. and Cirsium arvense. Four years of field experiments based on permanent plots made it possible to suggest the mode of restoration of 130 ha arable lands into native floodplain meadows. The restoration started in 1998. Harvester collected the seeds from quality meadows. Yield was 20 kg of seeds per hectare containing about 15 plant species. Transfer of turfs was used as an additional method to increase the number of species from the seed bank. Monitoring of large-scale restoration started in 1999.
Calcareous grassland ecosystem restoration after burning caused by military use on Pakri islands, Estonia
L. Truus & M. Tobias
Institute of Ecology of the Tallinn Paedagogical University, Tallinn, Estonia, E-mail: laimi@eco.edu.ee
Changes in land use are followed by changes in landscape pattern, most clearly reflected in vegetation. Regular sheep and cattle grazing supported the development of calcareous grasslands (alvars) on thin rendzic leptosol on limestone and coastal deposits in western and northern Estonia. Nowadays, grazing is commonly ceased on this type of grasslands and in the course of succession development. Grasslands have overgrown with tall grasses (Avenula pratensis), shrubs (Juniperus communis, Corylus avellana, Rhamnus cathartica etc.) and trees (Pinus sylvestris, Fraxinus excelsior, Sorbus aucuparia). Distribution of extensive calcareous grasslands on Pakri Islands at the north-western coast of Estonia is a result of such pasture management that continued until 1940s. Unlike other unmanaged grasslands, they were not overgrown by brushwood, because these areas were used as a training field for military combat planes. This practice continued until 1991. Growth of tree and shrub layer was reduced because of fires that followed the shooting and bombing practise. Species composition of calcareous grasslands has changed, but large areas covered by them were preserved. Nowadays the patches of calcareous grasslands on Pakri Islands are the largest in Estonia (up to 10 hectares). Vegetation and soil were studied on permanent plots in 1994, a few years after the fires caused by bombing stopped. A few tussock-forming species (Sesleria cerulea, Festuca ovina) dominated the vegetation cover while other vascular plants and mosses surrounded the tussocks. Two years later, in 1996, herbaceous plants covered soil surface more equally and the domination of Sesleria caerulea and Festuca ovina decreased. In the following 2-3 years, juvenile shrubs and trees arrived, indicating the onset of overgrowing.
Restoration of the site conditions of acidificated fen meadows by turf-stripping and trenches
D. van der Hoek
Department of Nature Conservation and Plantecology, Wageningen Agricultural University, Bornsesteeg 69, 6708PD Wageningen, The Netherlands; E-mail: Dick.vanderhoek@staf.ton.wau.nl
Until 50 years ago, nutrient-poor, base-rich fen meadows were common in The Netherlands. Fen meadows are semi-natural ecosystems derived from fens by extensive agricultural exploitation for hay-making without fertilization but with some degree of drainage. One of the typical plant communities in fen meadows is Junco-Molinion. Almost everywhere in The Netherlands the relicts of oligotrophic fen meadows have become increasingly eutrophicated and acidified. An important factor in this process has been the lowering of groundwater, caused by exploitation of deep groundwater for drinking water and by drainage for agriculture. Lowering of groundwater levels has resulted in increased availability of nutrients in the soil. In stead of the upward seepage of calcareous groundwater, precipitation water is infiltrating which has further contributed to the acidification. Consequently many plant species of these communities have become endangered or extinct. Special water management is needed to maintain or restore the characteristic abiotic site conditions. First, measures have to be taken to decrease loss of water due to pumping of deep groundwater. Second, measures are needed inside a nature reserve to diminish eutrophication and acidification. These measures may include turf-stripping and trench-cutting. Turf-stripping removes the eutrophicated topsoil, and has a favourable effect in that by lowering the surface level wetter conditions are created in the remaining topsoil. Shallow trenches increase surface run off of precipitation excess in winter preventing downward seepage and acidification processes. The effects of these measures have been monitored in several pilot projects. We present the results of hydrological and soil investigations of the effectiveness of both measures in a fen meadow and the results of a laboratory experiment on the relation between the buffer capacity of the topsoil and the groundwater quality. The effectiveness of trenches depends on the depth, the distance between them and on the water level. Deep trenches (about 70 cm) at a distance of 25 m cause infiltration of precipitation all year long and discharge of lithotrophic groundwater. Shallow trenches with a depth of 25 cm and a distance of 10 m drain the riparian grasslands. They discharge atmotrophic water only during wet periods when higher groundwater levels exist which is to the advantage of the upward seepage. Turf-stripping stimulates surface run off of precipitation water, which makes the groundwater more lithotrophic. In the control plots the pH and buffercapacity of the topsoil decreased since the start of the experiment eight years ago. The pH-values are relatively high in the topsoil of the stripped plots and decreased less. The experiment in the laboratory indicated that, if they are under influence of lithotrophic water, pH and buffercapacity of the topsoil of stripped plots will increase substantially more than at the control plots. We conclude that if trenching is combined with turf-stripping, the characteristic wet, nutrient poor site conditions will be restored. These conditions are only sustainable if the topsoil is under the influence of lithotrophic groundwater. Eight years after the turf-stripping, typical pioneer plant species such as Juncus bulbosus, Carex oederi, Drosera intermedia are still found in all stripped plots. In plots where the wet, nutrient poor and lithotrophic conditions were restored, species such as Dactylorhiza majalis, Carex panicea, Lycopodium inundatum were re-established. Some more demanding species, such as Parnassia palustris, had arrived in some of these plots but disappeared again. We conclude that these characteristic species will only persist in plots with the highest buffer capacity where the upward seepage of lithotrophic groundwater dominates the infiltration of atmotrophic water.
Quantifying hydrochory in a small Dutch lowland river
S. Verbeek & R. van Diggelen
Laboratory Of Plant Ecology, University of Groningen, PO Box 14, NL-9750 AA Haren, Netherlands, E-mail: S.K.Verbeek@biol.rug.nl
The absence of viable seeds of target communities in the soil seed bank often appears to be a bottleneck when trying to restore degraded meadows. However, in the case of alluvial flood plain meadows hydrochory is often assumed to supply large amounts of propagules during the flooding period. The present study was set up to test this assumption in a highly fragmented landscape in the Netherlands. The contribution of a regularly flooded alluvial meadow with well-developed plant communities to seed transport by the river was measured during a whole year. It appeared that the river carried only small quantities of viable seeds despite the availability of huge amounts of propagules in an adjacent nature reserve. These results suggest that species dispersal by hydrochory to 'restored' sites may be often overestimated. At present it is not possible to use these results for a quantitative estimation of recovery rates but it seems possible to make some qualitative predictions.
Spontaneous vegetation development during the first eight years after topsoil removal
R. Verhagen, J. Klooker, J.P. Bakker & R. van Diggelen
Laboratory of Plant Ecology, University of Groningen. P.O. Box 14, 9750 AA Haren, The Netherlands, E-mail: h.m.c.verhagen@biol.rug.nl
Removal of the top soil (up to 50 cm) is a recently developed technique to lower the soil nutrient status and improve the prospects of restoration for heathland or species-rich grassland in agricultural areas. We assume that regeneration is only possible when abiotic conditions lie within the tolerance of target communities, when viable diaspores of goal species are available, and when there are enough regeneration gaps in the vegetation. The present study was set up to investigate the effect of top soil removal on these elements and their role in the process of regeneration. Data will be presented on vegetation development in nine sites where the top soil has been removed less than 10 years ago. Comparison between sites makes clear which trends are general after topsoil removal, and which developments are unique for a particular site. By comparing the actual abiotic conditions and management practices with optimal conditions for target communities, the potential vegetation development is estimated. Next, the actual vegetation development is compared with the potential vegetation composition. It appears that there is a large discrepancy between the actual and potential vegetation composition. Studies on seedbank dynamics and species dispersal enable an insight into underlying mechanisms. It is concluded that top soil removal is an effective means of soil impoverishment but that regeneration is a slow process. It is expected that the "new nature reserves" will remain greatly undersaturated for the target communities for a long time.
Predictability of primary succession in former lignite coal mining areas
G. Wiegleb1 & B. Felinks2
1Brandenburgische Technische Universität, LS Allgemeine Ökologie, PO Box 101344, D-03013 Cottbus, Germany, E-mail: wiegleb@tu-cottbus.de; 2 Umweltforschungszentrum Leipzig-Halle GmbH, Projektbereich Naturnahe Landschaften und Ländliche Räume, Permoserstr. 15, D-04318 Leipzig, Germany
Based on a comprehensive 4-year survey, the spontaneous vegetation development of the Lusatian post-mining landscape is described. Indirect gradient analysis showed that there is a broad overlap of vegetation types, and that no successional trends can be inferred on the basis of floristic composition alone. Multivariate analyses showed that variables like pH, organic carbon, phosphate and water capacities have an obvious influence on species composition. However, it is not clear whether this is cause or consequence of the respective vegetation colonization. Autocorrelation effects hamper the interpretation of the complex data set. Partitioning between spatial, temporal and habitat effects was not yet possible based on the data set used. All successional sequences observed or inferred so far are combined into a conceptual model of spontaneous vegetation development. In the first instance various types of initial vegetation are possible. Initial vegetation is subsequently followed by a large number of stages and types forming a complex network. Some of these stages are stabilized by a dense vegetation layer or competitive power of the dominant species thus forming temporary final stages. The direction of succession is influenced to varying portions by neighborhood effects, facilitation, disturbance. Predictability is to be differentiated as to the degree of precision desired. In general it is easy to predict that 90 % of the area will become woodland in less than 100 years time. The average duration of the stages and approximate transition probabilities of vegetation types such as Corynephorus canescens and Calamagrostis epigejos grassland can be inferred from the present konowledge, too. However, actual transitions on a small spatial scale cannot be exactly predicted because of the above mentioned stochastic events. A short final note will be made on comparison of predictability of plant and animal colonization in post-mining landscapes.
