Paper: Vegetation Restoration on Opencasts

www.minest.ee/mining txt: An Analysis of Vegetation Restoration on Opencast Oil Shale Mines in Estonia Margus Pensa,1,2,5 Arne Sellin,3 Aarne Luud,1 and Ingo Valgma4 Abstract We compared four types of 30-year-old forest stands growing on spoil of opencast oil shale mines in Estonia. The stand types were: (1) natural stands formed by spontaneous succession, and plantations of (2) Pinus sylvestris (Scots pine), (3) Betula pendula (silver birch), and (4) Alnus glutinosa (European black alder). In all stands we measured properties of the tree layer (species richness, stand density, and volume of growing stock), understory (density and species richness of shrubs and tree saplings), and ground vegetation (aboveground biomass, species richness, and species diversity). The tree layer was most diverse though sparse in the natural stands. Understory species richness per 100-m2 plot was highest in the natural stand, but total stand richness was equal in the natural and alder stands, which were higher than the birch and pine stands. The understory sapling density was lower than 50 saplings/100 m2 in the plantations, while it varied between 50 and 180 saplings/100 m2 in the natural stands. Growing stock volume was the least in natural stands and greatest in birch stands. The aboveground biomass of ground vegetation was highest in alder stands and lowest in the pine stands. We can conclude that spontaneous succession promotes establishment of diverse vegetation. In plantations the establishment of diverse ground vegetation depends on planted tree species. Key words: Alnus glutinosa (L) Gaertn., Betula pendula Roth, forest plantation, opencast mine, Pinus sylvestris L., restoration, spontaneous succession. Introduction The destruction of ecosystem through mining for minerals and other activities to meet industry demands has been an intrinsic part of modern development. Further need for mineral resources will accelerate degradation of natural habitats, which will result in reduced biodiversity (Singh et al. 2002). In the second half of the twentieth century scientists and engineers were presented with many challenges to achieve restoration, but the less difficult goal of reclamation was more often practiced on human-disturbed areas around the world (definitions as in www.ser.org). As the utilization of natural resources continues and opportunities to restore ecosystems damaged by human activities become more common, restoration is playing an increasingly important role in environmental protection (Prach et al. 2001). The public responds emotionally to lands degraded by mining activities and associates mining with land that has been left devoid of all topsoil, all vegetation, and any hope of regeneration in the short to mid time scale. The Convention on Biological Diversity (1992) signed by most states of the world in Rio de Janeiro calls for ecologically sound restoration of degraded ecosystems as measures to promote the recovery of local biodiversity. Governments have therefore frequently given resources to reestablish vegetation on degraded lands, in anticipation that this will lead to restoration of the preexisting ecological state and may add economic value to the degraded lands (Hunter et al. 1998). The traditional approach to reclamation has been to sow grass and legumes and plant trees to minimize financial and human resource expenditures. Landscape engineers and foresters often establish a low-diversity plant cover or use monospecific plantations of exotic species (Hunter et al. 1998; Rebele & Lehmann 2002). Although plantations can play a key role in restoring forest ecosystems and achieving short-term socioeconomic goals by protecting the soil surface from erosion, catalyzing development of native forests, and accelerating the recovery of genetic diversity (Singh et al. 2002), spontaneous vegetation succession, or natural recovery, as an alternative approach to restoration or reclamation has gained increasing attention (Prach & Pysˇek 1994, 2001; Prach 1994; Prach et al. 2001). It has been claimed that spontaneous succession can be more efficient than human efforts at returning degraded lands to their original state and reestablishing the self-regularity of ecosystems (Prach et al. 2001). Depending on soil conditions, the required time period for establishment of woody species on degraded mining sites in Central Europe has been, on average, 20 years (Prach 1994). The first individuals of woody species may be present at the beginning of succession (Rebele 1992; Prach 1994; Prach & Pysˇek 2001; Rebele & Lehmann 2002), and the 1 Institute of Ecology, Tallinn Pedagogical University, 15 Pargi Street, 41537 Jo˜hvi, Estonia. 2 Rovaniemi Research Station, Finnish Forest Research Institute, P.O. Box 16, 96301 Rovaniemi, Finland. 3 Department of Botany and Ecology, University of Tartu, 40 Lai Street, 51005 Tartu, Estonia. 4 Department of Mining, Tallinn Technical University, 82 Kopli Street, 10412 Tallinn, Estonia. 5 Address correspondence to M. Pensa, email margus.pensa@metla.fi 2004 Society for Ecological Restoration International 200 Restoration Ecology Vol. 12 No. 2, pp. 200 206 JUNE 2004 probability of their early establishment is higher in moderately wet, nutrient-poor sites (Prach 1994) or in sandy soils (Rebele 1992). In general the establishment of woody species on degraded lands is highly variable due to the many stochastic factors that affect vegetation succession. Promoting spontaneous succession for ecological restoration therefore requires careful consideration of the target ecosystem and time scale in which the goals of restoration are to be achieved (Prach et al. 2001). If the goal is to restore a certain type of plant community, an engineered restoration process is preferred and succession should be directed toward intended objectives. When selecting suitable tree species for plantations, variable results rendered by different species under the same environmental conditions must be borne in mind (Singh et al. 2002). For example, Montagnini et al. (1995) have shown that among 20 indigenous tree species used for reclaiming forests on degraded lands in Brazil, four species tended to have more positive effects on soil properties than others. On abandoned agricultural lands in India restored with different bamboo species, the diversity of ground vegetation was greatest under the canopy of a specific species (Arunachalam & Arunachalam 2002). In this study we selected three tree species, Betula pendula (silver birch), Alnus glutinosa (European black alder), and Pinus sylvestris (Scots pine), growing in 30-year-old plantations established on calcareous spoils of opencast oil shale mines in Estonia, and evaluated their impact on understory and ground vegetation. Reclamation of the Estonian oil shale opencast mining areas is regulated by the Ministry of Environment. It has been determined that the goals of reclamation should encompass aspects of both economic (forestry) and ecological importance. However, studies carried out on the reclaimed oil shale opencast mining areas have mainly focused on questions related to forest management (Kaar et al. 1971; Kaar 2002). The outcome of reclamation has been measured in the volume increment of growing stock in stands. Except for pedogenesis (Reintam & Kaar 1999; Reintam 2001; Reintam et al. 2002), other ecological aspects of reclamation have been scarcely studied on the oil shale opencast mining areas. Laasimer (1973) and Reintam and Kaar (2002) studied spontaneous succession in the oil shale opencasts, but they also emphasized the economic outcome, suggesting establishment of plantations as the only way to achieve rapid recovery of vegetation on mined areas. If spontaneous succession is permitted, however, valuable ecosystems can develop over decades, which may serve as refuges for rare species and communities (Kirmer & Mahn 2001). This proposition is supported by the results of Holl (2002), who demonstrated that spontaneous succession can create diverse plant communities on an abandoned coal mine in the United States. The need for comparisons between engineered reclamation and spontaneous succession has also been emphasized by Prach et al. (2001) and Pysˇek et al. (2001). Therefore, we additionally sampled mine spoils where spontaneous succession has occurred since 1950. Three specific questions are addressed by this study: (1) are there differences in characteristics of the tree layer between plantations and spontaneous stands; (2) does biomass and diversity of ground vegetation in plantations differ from that of spontaneously developed stands; and (3) have tree species used to establish plantations had different effects on ground vegetation biomass and diversity. Methods Study Area The study was conducted at Ku¨ttejo˜u (59 200 N, 26 590 E) and Narva (59 180 N, 27 450 E) oil shale opencast mining areas in northeastern Estonia. Mean annual temperature is 4.5 C in both areas; annual precipitation is 600 mm in Ku¨ttejo˜u and 660 mm in Narva. Oil shale is the main mineral resource of Estonia, and it is mostly used for energy production. Mining for oil shale began at the beginning of the twentieth century, but its extraction and consumption was most intensive in the 1980s, declining sharply after the collapse of the Soviet Union in 1991. A reclamation program for opencast oil shale mines was initiated in the late 1960s when large areas around Narva were opened to mining activities. As of 2003, about 9000 ha of oil shale opencast mines have been reclaimed by leveling the mine spoils and planting a variety of tree species on them (Reintam et al. 2001). The most frequently planted tree species have been Pinus sylvestris(covering approximately 85 90% of the reclaimed area) and Betula pendula. In total 54 species have been used for reclamation, but most of them have only been planted experimentally on small plots (Kaar et al. 1971). The reclamation program did not set targets for restoring the opencast spoils in Ku¨ttejo˜u, where mining activities had ceased by the 1950s. Since then, the Ku¨ttejo˜u mining area (200 ha) has been revegetated by means of spontaneous succession to a sparse forest with an average stand age of 30 years. Background of the Mining Areas Ku¨ttejo˜ u opencast mine operated on former agricultural lands from 1925 to 1947. Oil shale layers were mined by hand, and the overburden was stripped by a 3-m3 steam excavator equipped with a spoiler. The width of the mining pit was 20 m, forming spoils with maximum slopes of 20 . The thickness of the oil shale layer was 2.6 m, and the overburden thickness varied from 5 to 12 m, with an average stripping factor of 1.9 m3 /t. Consequently, the height of the ground in the mined out area is 0 2 m below the original ground surface. As a spoiler and mechanical shovel were used for stripping, the estimated swell factor of the material was 1.3. Maximum size of the material mined was 1.4 m due to bucket size, while the average size was 0.15 m. Due to the lack of reclamation requirements and poor mobility of the spoiler, the spoil depth varies from 0.5 to 2 m. The organic matter content of the spoil, which originates from kukersite oil shale, is on average 4 6% and is evenly distributed in the spoil. Vegetation Restoration on Opencast Oil Shale Mines JUNE 2004 Restoration Ecology 201 Mining activities at Narva opencast mine began in 1963, and the mine is still operating today. The area was previously covered with swamp forests and bogs. The oil shale layers are mined by blasting; the overburden is blasted and stripped by 15 m3 and 90 m long boom draglines. The widths of the mining pits are 50 m, forming spoils with maximum slopes of 3 . The thickness of the oil shale layer is 2.0 m, and the overburden thickness varies from 11 to 17 m, with an average stripping factor of 4.1 m3 /t. Consequently, the average height of the ground in the mined out area is 2 4 m above the original ground surface. The higher surface level and water pumping, which is required for opencast operations, results in lower water availability and spoil seepage than at the Ku¨ttejo˜u mine. The Narva mining area has been a test site for a variety of new stripping methods. As a result, in some areas, the peat layer is located at the bottom of the spoil and in other areas, on the top. The organic component of the remaining oil shale layers is mostly at the bottom of the spoil. Dragline, and in some cases a mechanical shovel, was used for stripping the blasted overburden rock, and therefore the swell factor of the material is 1.4. Maximum size of the material mined was 3.1 m, while the average size was 0.3 m. Spoil depth is uneven and ranges from 0.5 to 1.5 m, but can be as much as 6 m in test sections. The organic matter content of the spoil, which also originates from kukersite oil shale, is on average 2% (4% at the bottom and 1% at the surface of the spoil). Due to the stripping method used, the material is unevenly distributed in the spoil and the estimated swell factor is up to 1.5 at the bottom. This is an additional factor that causes water runoff from the spoil. The spoil chemistry at the Narva mining area was analyzed by Vaus (1970) before reclamation (i.e., before tree growth had influenced pedogenesis). Similar data were not available for the Ku¨ttejo˜u mine; therefore we took spoil samples in the most recently abandoned parts of the opencast. The samples were analyzed for pH, total nitrogen, and P2O5, and compared with those taken at Narva (Table 1). It must, however, be recognized that the spoil chemistry at Ku¨ttejo˜ u has been affected by vegetation establishment over decades and the results do not represent the pre-successional conditions in this area. Sampling Four different types of forest stands were sampled: (1) a natural stand formed by spontaneous succession, (2) a plantation of Pinus sylvestris (referred to as pine stand), (3) a plantation of Betula pendula (birch stand), and (4) a plantation of Alnus glutinosa (alder stand). In June 2002 we established five 10 3 10 m randomly located sample plots in each stand type (20 plots total). According to forest inventory data the stands were on average 30 years old, which was confirmed by taking tree cores with an increment borer from within each plot. Within the plots we counted all trees with diameter at breast height (dbh) greater than 10 cm and estimated density of understory (number of shrubs and tree saplings/ 100 m2 ). For analyzing ground vegetation we marked five small (0.4 3 0.5 m) randomly located quadrats within each plot. Thus, ground cover in each stand type was represented by 25 quadrats (100 quadrats total). We harvested all vascular plants from within the quadrats and dried them at 60 C for 48 hr. We determined total aboveground biomass and biomass by species for each stand type (g/m2 ). We also calculated mean number of species, mean species diversity (H) expressed by Shannon Wiener’s index weighted with biomass and volume stock of trees. Shannon Wiener’s index of diversity was calculated as follows: H ¼ Xn i¼1 pi log2ðpiÞ where n is the number of species and pi is the proportion of the ith species in total biomass. Data Analysis To compare the effects of restoration technique (spontaneous succession vs. plantations) and tree species on characteristics of the understory and ground vegetation and tree layer, we applied univariate ANOVA using stand type as a fixed factor. Post hoc comparisons were carried out using the LSD test. The assumptions of normality and homogeneity of variances were checked using the Kolmogorov Smirnov D-statistic and the Brown Forsythe tests, respectively. All statistical computations were made with the aid of the computer package Statistica’98 (StatSoft Inc., Tulsa, OK, U.S.A.) at the significance level a 5 0.05 (Statistica 1998). Results The density of trees with dbh greater than 10 cm was significantly lower in the natural stands than in the plantations, whereas species composition of the natural stands was more diverse compared with the plantations (Tables 2 & 3). There was one dominant tree species (abundance .90%) in all three plantations and a few co-dominant species with abundance above 5% (Table 3). Unlike the plantations, no one species dominated in the tree layer of the natural stands. Although nearly half of the trees were birches, which is the most common species in secondary Table 1. Characteristics of spoil chemistry in studied areas (mean ± SD). Variable Narva n Ku¨ttejo˜u n pHKCl 7.1 ± 0.2 18 7.4 ± 0.1 6 Total N (%) 0.1 ± 0.05 15 0.2 ± 0.06 6 P2O5 (mg/kg) 16.6 ± 10.7 18 20.7 ± 8.1 6 Data on Narva were published by Vaus (1970). Vegetation Restoration on Opencast Oil Shale Mines 202 Restoration Ecology JUNE 2004 succession in oldfields in Estonia, other species were also frequently represented (Table 3). The sparse tree layer in natural stands had a positive effect on the density of the understory (Table 2). In other stand types there were fewer than 50 saplings/100 m2 , the only exception being a plot in a birch stand where Caragana arborescens had been densely planted (300 saplings/100 m2 ) under the tree layer. This is why mean understory density was high in the birch stands (Table 2); however, the birch stands also had the highest degree of variation, with densities ranging from 20 to 300 saplings/100 m2 . Species richness of the understory followed the same pattern as that of the tree layer with the natural stands having significantly more species per plot than the plantations (Table 2). Although trees were more sparsely distributed in the natural stands, the difference in volume stock between the plantations and natural stands was not as clear. When compared with pine stands, natural stands would provide similar timber production, but when compared with birch and alder stands, the potential for timber production was much less (Table 2). In alder stands, the biomass of ground vegetation was significantly greater than that in the other two plantations as well as in the natural stands (Table 2). Species richness was significantly higher in the alder stands than that in the pine and birch stands, where it averaged 7 species/m2 (Table 2). In natural stands, species richness was highly variable and did not differ clearly from that of any other stand type (Table 2). Shannon Wiener’s diversity index was smallest in the birch stands, while among other stand types its variation was not significant (Table 2). Calamagrostis arundinacea (red grass) alone accounted for more than half the cumulative biomass in birch stands, while five species in natural, four in alder, and two in pine stands accounted for more than half the cumulative biomass (Table 3). Discussion The studied sites had different historical backgrounds that might have influenced the measured characteristics of vegetation. Mining activity at the Ku¨ttejo˜ u mine ended 20 years earlier than at Narva; therefore Ku¨ttejo˜u was exposed longer to species immigration. A longer immigration period was found to enhance biodiversity in studies of island colonization (Begon et al. 1996). Differences in vegetation between Narva and Ku¨ttejo˜ u opencast mines, however, cannot be explained merely by species immigration. First, although less exposed to immigration, the ground vegetation in alder stands had the same average species richness as that of natural stands. Second, although the Ku¨ttejo˜ u opencast mine was older than Narva, the average age of the tree layer was 30 years in both areas. Thus, there was a time lag of about 20 years in the development of woody stands at the Ku¨ttejo˜u mine. A similar time lag was characteristic of the formation of a tree layer by spontaneous succession in lignite opencast mines in Central Europe (Prach & Pysˇek 1994, 2001; Prach 1994). The second aspect to be considered when comparing study areas is their different locations; the Ku¨ttejo˜ u mine is situated at about 50 km west of the Narva mine. Although this distance does not result in climatic variation, it may affect propagule availability due to differences in surrounding plant communities. The Ku¨ttejo˜u mine is located in a diverse agricultural landscape (fields interspersed with small areas of forests and gardens), while the Narva mine is surrounded by natural ecosystems (swampy deciduous forests interspersed with pine bog forests). A more diverse landscape is probably the main reason why ground vegetation at Ku¨ttejo˜u opencast has the greater total number of plant species, but spontaneous succession seems to be a necessary condition for maintaining that diversity. Although Betula pendula was the most abundant tree species in the natural stands, its dominance did not exceeded 50% and groups of other woody species were interspersed within gaps. The stochastic nature of spontaneous succession is thought to be the main reason for such spatial patterns (Rebele 1992). This variability creates the required conditions for development of forest understory (shrubs and tree saplings) and ground vegetation. Sparse tree distribution was the main reason for the small volume of growing stock in the natural stands, indicating that their economic value is relatively low compared with plantations. In the plantations, contrary to the natural stands, the tree layer was dense and dominance of the Table 2. Comparison of the vegetative characteristics among the studied stand types (mean ± SD). Natural Pine Birch Alder Number of tree species/100 m2 3 ± 1* 1 ± 0.4† 2 ± 1† 1 ± 0.5† Number of trees/100 m2 9 ± 4* 25 ± 11† 28 ± 2† 26 ± 7† Volume stock of trees (m3 /100 m2 ) 1.0 ± 0.6* 1.9 ± 1.1*‡ 3.4 ± 0.8† 2.8 ± 0.5†‡ Number of understory species/100 m2 9 ± 3* 5 ± 2† 5 ± 1† 3 ± 2† Understory density (saplings/100 m2 ) 98 ± 54 38 ± 17 85 ± 120 16 ± 19 Biomass of ground vegetation (g/m2 ) 52.9 ± 20.8* 32.4 ± 13.7† 59.2 ± 23.6* 104.0 ± 41.3‡ Number of species in ground vegetation/m2 12.4 ± 6.9*† 7.0 ± 2.5* 7.0 ± 1.9* 12.2 ± 2.8† Diversity of ground vegetation, H 1.73 ± 0.44* 1.52 ± 0.32* 0.86 ± 0.47† 1.56 ± 0.46* Values within rows with different symbols are significantly different at a 5 0.05. Due to high degree of variation, the significance of differences was not estimated for density of understory. Vegetation Restoration on Opencast Oil Shale Mines JUNE 2004 Restoration Ecology 203 Table 3. List of vascular plants in the forest stands growing on the opencast oil shale mines in Estonia. Stand Type Species Natural Pine Birch Alder Total number 54 27 28 36 Tree layer 52 4 3 Alnus glutinosa 91.5* Alnus incana 6.9* Betula pendula 47.7 90.8* 1.5 Larix decidua 8.8* Malus domestica 2.3 Pinus sylvestris 4.5 91.2* 6.3* Populus tremula 29.5 1.4 Salix caprea 15.9 1.4 Understory 11 10 9 7 Alnus glutinosa 10.8* Alnus incana 2.7 Betula pendula 15.4 44.5 17.5* 5.4 Caragana arborescens 71.1* Frangula alnus 1.9 0.6 0.2 5.4 Larix decidua 1.3* Padus avium 4.9 0.6 Picea abies 2.3 12.9 3.6 13.5 Pinus sylvestris 0.9 18.1* 0.7 Populus tremula 15.5 1.3 0.5 Rhamnus cathartica 3.5 Ribes alpinum 17.6 Ribes nigrum 54.1 Rosa vosagiaca 1.3 Salix sp. 5.4 11.0 4.7 8.1 Sorbus aucuparia 26.2 8.4 1.2 Viburnum opulus 4.4 0.5 Ground vegetation 41 17 18 29 Calamagrostis arundinacea 24.0 66.0 13.4 Calamagrostis epigeios 9.5 Dactylis glomerata 1.3 Deschampsia caespitosa 6.1 0.5 Epilobium angustifolium 4.7 0.5 0.5 7.5 Epipactis helleborine 0.4 Eupatorium cannabinum 7.6 Festuca gigantea 0.3 10.9 Filipendula ulmaria 1.8 Fragaria vesca 14.0 13.6 7.6 7.0 Geum rivale 4.6 Geum urbanum 3.3 Hieracium pilosella 4.2 Medicago lupulina 6.0 Mycelis muralis 1.1 2.0 0.4 0.2 Orthilia secunda 34.2 11.7 Phragmites communis 1.1 Poa pratensis 1.7 Pyrola rotundifolia 8.6 3.8 0.1 Rubus idaeus 2.5 3.2 2.8 Rubus saxatilis 16.7 4.6 1.2 2.0 Solanum dulcamara 2.6 Solidago virgaurea 2.1 Stellaria media 2.9 Taraxacum officinale 0.7 5.3 1.1 3.2 Tussilago farfara 2.3 5.1 5.3 0.6 Urtica dioica 24.7 Valeriana officinalis 1.0 0.4 Veronica chamaedrys 2.2 (continued) Vegetation Restoration on Opencast Oil Shale Mines 204 Restoration Ecology JUNE 2004 planted species was very high (.90%). As previous studies (Kaar et al. 1971; Kaar 2002) have shown, environmental conditions on the oil shale opencast spoil are ideal for growth of Pinus sylvestris and Betula pendula as well as Alnus glutinosa. Thus, dense stands of these species suppress the growth of other trees, which otherwise could spontaneously establish on mined areas. Although in some cases establishment of plantations on degraded lands may promote growth of other species (Lugo 1997; Singh et al. 2002), our results indicate that on calcareous and stony spoils of the opencast oil shale mines, spontaneous succession may have several advantages in terms of increased plant diversity. This increase in plant diversity may result in an increase in the diversity of other organisms, as variability in stand structure and species composition creates habitats for a variety of organisms. Our results also revealed differences in the ability of the tree species to modify growing conditions for other plants as measured by the effect of woody species on ground vegetation. Alnus glutinosa, a species that forms symbiotic relationships with nitrogen-fixing bacteria from the genus Frankia (Wall & Huss-Danell 1997; Sprent & Parsons 2000), promotes the most ground vegetation growth (Table 2). Vigorously growing herbaceous vegetation in turn suppresses woody seedlings, which seems to be the main reason why the understory density was lowest in the alder stands. Low-ground vegetation biomass in the pine stands is probably related to poor soil-forming capabilities of coniferous monocultures as shown by Kilian (1998). The greatest biomass of ground vegetation observed in the alder stands was related to high species richness. Although mean species richness in the natural stands was also 12 species/m2 , it varied widely due to the high spatial heterogeneity of the stand structure. The variation in species richness of the ground vegetation among plots in the natural stands, however, followed the same pattern as in forest plantations. In plots with high dominance of Pinus sylvestris, the number of species was low, while within the plots where deciduous tree species dominated, species richness of the ground vegetation was higher. Similar results were obtained by Pitka¨nen (1998), who found that the number of coniferous and broad-leaved tree species had a significant effect on the diversity of vegetation in managed boreal forests. The high dominance of Betula pendula in the tree layer was accompanied by high dominance of Calamagrostis arundinacea in the ground vegetation. This was obviously the main reason for the low-diversity index of the ground vegetation in the birch stands, while in the other stand types no one species accounted for 50% or more of the total aboveground biomass (Table 3). We can conclude that spontaneous succession is a useful technique for restoration of small areas of the calcareous and stony spoils of opencast mines and may replace the typical reclamation technique of planting tree monocultures where diversity is the goal. Spontaneous succession enhances establishment of diverse vegetation and may therefore create habitats for a wide range of organisms. If economic value is set as a priority, however, creation of plantations or scattered colonization foci may assist in overcoming dispersal barriers and direct succession toward defined targets (Robinson & Handel 2000). Selection of tree species for planting has a significant impact on development of the rest of the plant community. Among active reclamation practices in oil shale opencast mines, planting of Alnus glutinosa gave the best results in terms of total number of vascular species and aboveground biomass of ground vegetation. Planting density may also affect the development of vegetation in plantations. Further studies are needed to reveal whether the within-stand variability in diversity is connected to the density of the tree layer and whether there are temporal fluctuations in diversity and biomass of ground vegetation in oil shale opencast mines. Acknowledgments We are grateful to Elga and Ene Rull who helped us with the field sampling. Sarah Wilkinson and Michael Dunderdale checked the English language of the manuscript. The company Estonian Oil Shale is thanked for the permission to carry out the study in the Narva opencast. Plant Biochemistry Laboratory of the Estonian Agricultural University analyzed the chemistry of spoil samples. Prof. Karel Prach and two anonymous reviewers made valuable comments and annotations on the manuscript. The study was funded by the Ministry of Education, Republic of Estonia (project no. 0282119s02). LITERATURE CITED Arunachalam, A., and K. Arunachalam. 2002. Evaluation of bamboos in ecorestoration of ‘jhum’ fallows in Arunachal Pradesh: ground vegetation, soil and microbial biomass. Forest Ecology and Management 159:231 239. Table 3. Continued Stand Type Species Natural Pine Birch Alder Vicia cracca 3.0 Other 1.2 2.2 0.2 0.3 The values are percentages of total abundance for tree layer and understory, and percentages of total aboveground biomass for ground vegetation (only values .0.1% are shown). For each stand type, number of species in tree layer, in understory, and in ground vegetation as well as total number of species are shown in bold. Asterisks indicate planted species. Nomenclature of species according to Kukk (1999). Vegetation Restoration on Opencast Oil Shale Mines JUNE 2004 Restoration Ecology 205 Begon, M., J. L. Harper, and C. R. Townsend. 1996. Ecology. Individuals, populations and communities. Blackwell Science, Oxford, United Kingdom. Convention on Biological Diversity. 1992. Convention on biological diversity signed at the United Nations Conference on Environment and Development, 5 June 1992. Rio de Janeiro, Brazil (available from: http://www.biodiv.org/convention/articles.asp). Holl, K. D. 2002. Long-term vegetation recovery on reclaimed coal surface mines in the eastern USA. Journal of Applied Ecology 39:960 970. Hunter, I. R., M. Hobley, and P. Smale. 1998. Afforestation of degraded land—pyrrihic victory over economic, social and ecological reality. Ecological Engineering 10:97 106. Kaar, E. 2002. Coniferous trees on exhausted oil shale opencast mines. Forestry Studies 36:120 125. Kaar, E., L. Lainoja, H. Luik, L. Raid, and M. Vaus. 1971. Restoration of oil shale opencast mines. Ministry of Forestry and Nature Conservation, Tallinn, ESSR (in Estonian). Kilian, W. 1998. Forest site degradation—temporary deviation from the natural site potential. Ecological Engineering 10:5 18. Kirmer, A., and E.-G. Mahn. 2001. Spontaneous and initiated succession on unvegetated slopes in the abandoned lignite-mining area of Goitsche, Germany. Applied Vegetation Science 4:19 27. Kukk, T. 1999. Eesti taimestik. Vascular plant flora of Estonia. Estonian Academy of Sciences, Tartu and Tallinn, Estonia (in Estonian with English summary). Laasimer, L. 1973. Recovery of vegetation in leveled oil shale opencast mines. Forestry Studies 10:168 185 (in Estonian). Lugo, A. E. 1997. The apparent paradox of reestablishing species richness on degraded lands with tree monocultures. Forest Ecology and Management 99:9 19. Montagnini, F., A. Fanzeres, and S. Guimaraes da Vinha. 1995. The potentials of 20 indigenous species for soil rehabilitation in the Atlantic forest region of Bahia, Brazil. Journal of Applied Ecology 32:841 856. Pitka¨nen, S. 1998. The use of diversity indices to assess the diversity of vegetation in managed boreal forests. Forest Ecology and Management 112:121 137. Prach, K. 1994. Succession of woody species in derelict sites in Central Europe. Ecological Engineering 3:49 56. Prach, K., and P. Pysˇek. 1994. Spontaneous establishment of woody plants in central European derelict sites and their potential for reclamation. Restoration Ecology 2:190 197. Prach, K., and P. Pysˇek. 2001. Using spontaneous succession for restoration of human-disturbed habitats: experience from Central Europe. Ecological Engineering 17:55 62. Prach, K., S. Bartha, C. B. Joyce, P. Pysˇek, R. van Diggelen, and G. Wiegleb. 2001. The role of spontaneous vegetation succession in ecosystem restoration: a perspective. Applied Vegetation Science 4:111 114. Pysˇek, P., K. Prach, J. Mu¨llerova, and C. Joyce. 2001. The role of vegetation succession in ecosystem restoration: introduction. Applied Vegetation Science 4:3 4. Rebele, F. 1992. Colonization and early succession on anthropogenic soils. Journal of Vegetation Science 3:201 208. Rebele, F., and C. Lehmann. 2002. Restoration of a landfill site in Berlin, Germany by spontaneous and directed succession. Restoration Ecology 10:340 347. Reintam, L., 2001. Changes in the texture and exchange properties of skeletal quarry detritus under forest during thirty years. Proceedings of the Estonian Academy of Sciences. Biology. Ecology 50:5 13. Reintam, L., and E. Kaar. 1999. Development of soils on calcareous quarry detritus of open-pit oil-shale mining during three decades. Proceedings of the Estonian Academy of Sciences. Biology. Ecology 48:251 266. Reintam, L., and E. Kaar. 2002. Natural and man-made afforestation of sandy-textured quarry detritus of open-cast oil-shale mining. Baltic Forestry 8:57 61. Reintam, L., E. Kaar, and I. Rooma. 2001. Development of forest soil systems on quarry detritus of open-cast oil-shale mining. Pages 645 656 in Y. Villacampa, C. A. Brebbia, and J.-L. Uso´, editors. Ecosystems and sustainable development. III. Wessex Institute of Technology, Wessex, United Kingdom. Reintam, L., E. Kaar, and I. Rooma. 2002. Development of soil organic matter under pine on quarry detritus of open-cast oil-shale mining. Forest Ecology and Management 171:191 198. Robinson, G. R., and S. N. Handel. 2000. Directing spatial patterns of recruitment during an experimental urban woodland reclamation. Ecological Applications 10:174 188. Singh, A. N., A. S. Raghubanshi, and J. S. Singh. 2002. Plantations as a tool for mine spoil restoration. Current Science 82:1436 1441. Sprent, J. I., and R. Parsons. 2000. Nitrogen fixation in legume and nonlegume trees. Field Crops Research 65:183 196. Statistica. 1998. Statistica for Windows. I. General conventions and statistics I. StatSoft, Inc., Tulsa, Oklahoma. Vaus, M. 1970. The properties of spoil in relation to plant growth in oil shale opencast mines. Ministry of Forestry and Nature Conservation, Tallinn, ESSR (in Estonian). Wall, L. G., and K. Huss-Danell. 1997. Regulation of nodulation in Alnus incana Frankia symbiosis. Physiologia Plantarum 99:594 600. Vegetation Restoration on Opencast Oil Shale Mines 206 Restoration Ecology JUNE 2004