Thesis (MScEng (Civil Engineering))--University of Stellenbosch, 2011. / ENGLISH ABSTRACT: The hypothesis postulated in this research, namely that the effectiveness of natural and constructed buffer dune systems can be assessed by a set of indicators that defines the integrity of the dune system and triggers informed management decisions, was evaluated and proved to be essentially true. Two key objectives, namely (1) the identification of key indicators that define the buffer dune integrity; and (2) the development of a scientifically defendable and practical checklist-based method of gathering qualitative information on the identified key indicators so as to guide decision-making at municipal level formed the core of the study. The six dune integrity indicators that collectively define the risk profile of a particular site along the Southern Cape coastline are (1) the degree of protection from prevailing wave energy, (2) the characteristics of the dominant winds and sand supply during the dry season, (3) the relative height of the foredune, (4) the degree of pressure on the buffer dune due to humans, (5) the vulnerability of the type of coastline to erosion, and (6) the coastline stability considering the prevailing coastal processes. The first two indicators relate to the natural (permanent) characteristics of the site and can be defined by experts and presented in the form of a risk and vulnerability atlas layer for direct use by non-experts. The third and fourth indicators relate directly to the implementation of proactive assessment and appropriate management actions to ensure a high level of buffer dune integrity. The last two indicators allow for management intervention to reduce the vulnerability but may entail costly engineering solutions and require expert input. A conceptual risk profile assessment procedure and a decision support guideline incorporating these indicators were developed and evaluated for relevance and practicality through a series of workshops with municipal officials along the south coast of South Africa. It was seen that although some initial basic training may be required, carrying out rapid assessments of the environmental status of key components of an identified human–nature system, such as a buffer dune, is practical and achievable by non-experts. / AFRIKAANSE OPSOMMING: Die hipotese wat in hierdie navorsing gepostuleer is, naamlik dat die doeltreffendheid van natuurlike en geboude bufferduinstelsels geassesseer kan word deur ’n stel aanwysers wat die integriteit van die duinstelsel bepaal en ingeligte bestuursbesluite tot gevolg het, is getoets en bewys hoofsaaklik waar te wees. Twee sleuteldoelwitte, naamlik (1) die identifisering van sleutelaanwysers wat die bufferduinintegriteit bepaal; en (2) die ontwikkeling van ’n praktiese kontrolelys-gebaseerde metode wat wetenskaplik verdedigbaar is om kwalitatiewe inligting oor die geïdentifiseerde sleutelaanwysers in te samel ten einde besluitneming op munisipale vlak te bevorder, vorm die kern van die studie. Die ses duin-integriteitsaanwysers wat gesamentlik die risikoprofiel van ’n bepaalde terrein langs die kuslyn bepaal, is (1) die graad van beskerming teen die heersende golfenergie, (2) die kenmerke van die dominante winde en sandbron gedurende die droë seisoen, (3) die relatiewe hoogte van die voorduin, (4) die graad van druk op die bufferduin as gevolg van mense, (5) die eroderingskwesbaarheid van die soort kuslyn, en (6) die kuslynstabiliteit met inagname van die kusprosesse. Die eerste twee aanwysers het betrekking op die natuurlike (permanente) eienskappe van die terrein en kan deur kundiges bepaal word en in die vorm van ’n kaart in ’n risiko-enkwesbaarheidsatlas aangebied word vir direkte gebruik deur niedeskundiges. Aanwysers 3 en 4 hou direk verband met die implementering van tydige en deurlopende proaktiewe assessering en gepaste bestuursaksies om ’n hoë vlak van bufferduinintegriteit te verseker. Aanwysers 5 en 6 bevorder bestuursaksies om kwesbaarheid te verminder, maar kan moontlik duur ingenieursoplossings inhou en kundige insette benodig. ’n Konseptuele risikoprofielassesseringsprosedure en ’n besluitondersteuningsriglyn wat die aanwysers insluit, is ontwikkel en geëvalueer vir toepaslikheid en uitvoerbaarheid deur ’n reeks werkswinkels met munisipale amptenare aan die suidkus van Suid-Afrika. Hoewel aanvanklike basiese opleiding nodig kan wees, bly dit dat vinnige assessering van die omgewingstatus van sleutelkomponente van ’n geïdentifiseerde mens–natuurstelsel, soos ‘n bufferduin, prakties en haalbaar deur niedeskundiges is.

Foi realizado um trabalho de pisoteio experimental na vegetação de dunas do Parque Estadual da Serra do Mar, Núcleo Picinguaba, Ubatuba, SP. Dez parcelas permanentes foram instaladas, cinco no verão e cinco no inverno, subdivididas em 6 tratamentos, que receberam pisoteio controlado nas intensidades de 25, 75, 200, 500 e 1000 passadas, além do controle (sem pisoteio). O procedimento foi repetido em dois anos consecutivos. A cobertura e a altura relativa da comunidade e a cobertura das principais espécies da área foram avaliadas periodicamente ao longo de cada ano. As diferenças entre os tratamentos foram analisadas através do Teste de Friedman. A comunidade apresentou redução na cobertura e na altura da vegetação, mas sua recuperação ocorreu rapidamente. Os efeitos do pisoteio variaram com a estação do ano e com o número de ciclo do impacto, sendo mais evidentes no inverno e no segundo ano independente da estação. A altura decaiu mais rapidamente que a cobertura e demorou mais tempo para se recuperar. Não houve diferenças significativas entre o controle e os tratamentos de menor intensidade (25 e 75 passadas) nem entre os de média e grande intensidade (200, 500 e 1000 passadas), indicando que o uso da capacidade de carga recreativa não é um bom parâmetro para a área estudada. Analisando os dados obtidos pode-se afirmar que a comunidade estudada apresenta alta resistência e resiliência aos efeitos do pisoteio. O número de espécies aumentou com os anos do experimento, mas a maioria delas tem características de plantas invasoras. Três meses após o impacto as espécies apresentavam cobertura semelhante à observada antes do pisoteio. Os dados indicam que as espécies apresentam diferentes padrões de resistência e de resiliência. A Poaceae Panicum racemosum foi a espécie mais resistente ao impacto, e foi classificada como indiferente ao pisoteio; Hidrocotyle bonariensis foi a mais sensível, mas se recuperou rápido e superou os valores iniciais, sendo identificada como favorecida pelo pisoteio. Blutaparon portulacoides apresentou resistência e resiliência intermediárias, mas foi, aos poucos, eliminada da área pelas outras espécies, sendo classificada como desfavorecida pelo pisoteio. Considerando os resultados obtidos, uma forma eficiente de manejo da área é a concentração do uso, direcionando a visitação nas áreas de ocorrência da espécie mais resistente, evitando o lado esquerdo da praia cuja vegetação permanece mais estruturada. Recomenda-se o monitoramento periódico da área para evitar alterações drásticas na estrutura e na composição da vegetação. / Experimental trampling was conducted in a coastal sand-dune community located at "Serra do Mar State Park" in Southeastern, Brazil. It was used 10 permanent plots, five trampled in summer and other five in winter. The trampling intensities were 25, 75, 200, 500 and 1000 passes besides a control area (without trampling) in each plot. The procedures were repeated in two years. Response to trampling was assessed by determining species composition, vegetation cover and height evaluation in six evaluations during each year. The differences between trampling intensities were accessed with Friedman Test. Although there was loss of vegetation cover and height the study area was recovered in a few months. There were on trampling effects between season and years of impact. Effects were more noticeable for winter plots and for the second year of trampling. Reductions in height occurred with less impact and it took longer to recover. Neither 200, 500 and 1000 pass interference showed statistical differences between them, nor the control and 25 or 75 pass interference did. This suggests that the Recreation Carrying Capacity is not a good parameter to control visitors' impact in the area. The data analyses showed that the community has good resistance and resilience. The species number increased during the study, but most of the plant species were intruders. After three months, species cover was round the same as before trampling. The species showed different patterns of resistance and resilience, Panicum racemosum was the most resistant species. It was classified as neutral to trampling impacts. Hidrocotyle bonariensis, classified as favored, was the most sensible one, but it was recovered very fast, reaching higher cover values. Blutaparon portulacoides was classified as injured because although it had an intermediate behavior on resistance, it was almost eliminated from the plots as other species showed up, A management option that comes form this work is to concentrate public use on areas where the most resistant specie occur, avoiding the beach left side where vegetation is more preserved. It is important to monitor vegetation periodically to avoid drastic changes on vegetation structure or composition.

This thesis investigates whether topography has a quantifiable effect on the spatial distribution of landcover within an inland sand dune ecosystem. To accomplish this, an investigation was undertaken to compare the associations between various topographic characteristics and landcover distribution within the Great Sand Hills (GSH), Saskatchewan. The methodology comprised a spatial analysis of digital topographic and landcover data. In terms of landcover diversity, results shows that there is a moderate association between topography and landcover diversity in the GSH, and that results vary with spatial scale. Similarly, results show modest predictability for one particular community type (trees) thought to be strongly related to topographic position in this region. Several environmental factors are proposed that might overprint the topography-landcover association. Overall, this thesis expands the geographic context of the link between topography and landcover, while also demonstrating the utility of high resolution topographic and landcover data and the importance of spatial scale. / x, 90 leaves : col. ill. ; 29 cm

Foi realizado um trabalho de pisoteio experimental na vegetação de dunas do Parque Estadual da Serra do Mar, Núcleo Picinguaba, Ubatuba, SP. Dez parcelas permanentes foram instaladas, cinco no verão e cinco no inverno, subdivididas em 6 tratamentos, que receberam pisoteio controlado nas intensidades de 25, 75, 200, 500 e 1000 passadas, além do controle (sem pisoteio). O procedimento foi repetido em dois anos consecutivos. A cobertura e a altura relativa da comunidade e a cobertura das principais espécies da área foram avaliadas periodicamente ao longo de cada ano. As diferenças entre os tratamentos foram analisadas através do Teste de Friedman. A comunidade apresentou redução na cobertura e na altura da vegetação, mas sua recuperação ocorreu rapidamente. Os efeitos do pisoteio variaram com a estação do ano e com o número de ciclo do impacto, sendo mais evidentes no inverno e no segundo ano independente da estação. A altura decaiu mais rapidamente que a cobertura e demorou mais tempo para se recuperar. Não houve diferenças significativas entre o controle e os tratamentos de menor intensidade (25 e 75 passadas) nem entre os de média e grande intensidade (200, 500 e 1000 passadas), indicando que o uso da capacidade de carga recreativa não é um bom parâmetro para a área estudada. Analisando os dados obtidos pode-se afirmar que a comunidade estudada apresenta alta resistência e resiliência aos efeitos do pisoteio. O número de espécies aumentou com os anos do experimento, mas a maioria delas tem características de plantas invasoras. Três meses após o impacto as espécies apresentavam cobertura semelhante à observada antes do pisoteio. Os dados indicam que as espécies apresentam diferentes padrões de resistência e de resiliência. A Poaceae Panicum racemosum foi a espécie mais resistente ao impacto, e foi classificada como indiferente ao pisoteio; Hidrocotyle bonariensis foi a mais sensível, mas se recuperou rápido e superou os valores iniciais, sendo identificada como favorecida pelo pisoteio. Blutaparon portulacoides apresentou resistência e resiliência intermediárias, mas foi, aos poucos, eliminada da área pelas outras espécies, sendo classificada como desfavorecida pelo pisoteio. Considerando os resultados obtidos, uma forma eficiente de manejo da área é a concentração do uso, direcionando a visitação nas áreas de ocorrência da espécie mais resistente, evitando o lado esquerdo da praia cuja vegetação permanece mais estruturada. Recomenda-se o monitoramento periódico da área para evitar alterações drásticas na estrutura e na composição da vegetação. / Experimental trampling was conducted in a coastal sand-dune community located at "Serra do Mar State Park" in Southeastern, Brazil. It was used 10 permanent plots, five trampled in summer and other five in winter. The trampling intensities were 25, 75, 200, 500 and 1000 passes besides a control area (without trampling) in each plot. The procedures were repeated in two years. Response to trampling was assessed by determining species composition, vegetation cover and height evaluation in six evaluations during each year. The differences between trampling intensities were accessed with Friedman Test. Although there was loss of vegetation cover and height the study area was recovered in a few months. There were on trampling effects between season and years of impact. Effects were more noticeable for winter plots and for the second year of trampling. Reductions in height occurred with less impact and it took longer to recover. Neither 200, 500 and 1000 pass interference showed statistical differences between them, nor the control and 25 or 75 pass interference did. This suggests that the Recreation Carrying Capacity is not a good parameter to control visitors' impact in the area. The data analyses showed that the community has good resistance and resilience. The species number increased during the study, but most of the plant species were intruders. After three months, species cover was round the same as before trampling. The species showed different patterns of resistance and resilience, Panicum racemosum was the most resistant species. It was classified as neutral to trampling impacts. Hidrocotyle bonariensis, classified as favored, was the most sensible one, but it was recovered very fast, reaching higher cover values. Blutaparon portulacoides was classified as injured because although it had an intermediate behavior on resistance, it was almost eliminated from the plots as other species showed up, A management option that comes form this work is to concentrate public use on areas where the most resistant specie occur, avoiding the beach left side where vegetation is more preserved. It is important to monitor vegetation periodically to avoid drastic changes on vegetation structure or composition.

The aim of this study was to map the prevalence of aeolian sand dunes in Västerbotten, northern Sweden to find dunes formed during earlier deglaciation phases, which was then preserved in cold-based conditions during the youngest stadial of the Weichselian glaciation. These preserved dunes were expected to be covered by a layer of till and have a rather faint morphology compared to dunes formed during the Holocene. Consequently, high resolution LIDAR-derived images from the Swedish mapping, cadastral and land registration authority (Lantmäteriet) was used to detect the dunes throughout an area covering most of Västerbotten, above the highest coastline and below the mountain range. The analysis resulted in several new findings of aeolian dunes compared to the findings in the quaternary soil map created by the Geological Survey of Sweden. Despite the fact some of the dunes at the LIDAR-derived image had a strange faint appearance, it was quickly evident during the fieldwork that the aeolian sand was not covered by till on any of the dunes visited. The common view during the last decades has been that cold-based ice will not erode or alter the morphology of the landscape beneath the ice. However, this study suggests that hypothesis may not be entirely correct, and therefore it can be hard to use geomorphological implications to reconstruct past glacial environments. Thus, it can be concluded that even under cold-based conditions, preservation of aeolian sand dunes in Västerbotten is probably not very common.

The importance of understanding the ecological functioning of coastal dune systems is emphasized as being fundamental to the correct management of the dune landscape. Dune vegetation along the Eastern Cape coastline, from Cape St Francis in the west to Kei Mouth in the east was described in terms of the distribution and phytochorological affinities of the taxa. At a regional scale species distribution was strongly influenced by both the climate, particularly rainfall, and the phytochorological affinities of the taxa. Seven communities were defined using TWINSPAN, and the interrelationships between these communities in anyone area appeared to be linked to a successional gradient. Dune Slacks are thought to play a key role in this successional sequence, and a temporal study of this community led to a conceptual model of plant succession in these dunefields. Climate, particularly rainfall and wind, are major factors influencing plant succession. Wind-borne sand causes the slacks to migrate in an easterly direction under the influence of the predominantly westerly winds, although easterly winds, mainly in summer months may reverse these trends. Autogenic changes appeared to be important in this succession, and a comparative study of a good example of a primary succession at Mtunzini in Natal was undertaken to elucidate the main mechanism of change. Eight communities that were identified here were concluded to be distnbuted along a gradient of increasing age, with successional changes predictable, linear and directional. Species were grouped in distinct zones along the continuum and edaphic changes (decrease in soil pH, increase in organic matter and exchangeable bases) were related to the community based changes in species composition. The mechanism of change supported the facilitation model of plant succession which is a modification of the original Clementsian concept. Similar results were found in the Eastern Cape, but due to the harsh environment, multiple pathways of succession exist. Data from this study lent support to the model of plant succession developed earlier, and confinned that the dune slacks played an important role in this facilitation by acting as centres of diversity. The foredunes were found to have an indirect role in protecting these slacks from salt spray and sand movement. The central theme of the management studies was to investigate the ecological consequences of recreational pressure within the dune environment. Current levels of beach utilization at East London were lower than other beaches in South Africa, but a general trend of increasing utilization due to sociopolitical changes can be expected. The suitability of questionnaire surveys to assess aspects such as the adequacy of facilities, perceptual carrying capacity and the beach users opinion of natural vegetation and preference for particular beaches was demonstrated. The dune vegetation was found to be sensitive to human trampling, but at current levels, the ecological carrying capacity will not be exceeded since results of the aerial census counts and questionnaire survey revealed that few people entered sensitive zones such as the coastal forest. More detailed long term studies on the susceptibility of dune vegetation to both trampling and off-road vehicle impacts revealed a low resilience of dune plant communities to these effects. Although susceptibility differed between the three communities tested, generally the greatest amount of damage occurred after the first few passages, and vehicles caused a more significant decrease in height when compared to trampling. Recovery rates were slow and low levels of repeated damage were sufficient to retard or prevent the recovery of the plants. Stricter control of vehicle use on beaches is therefore required, and in high use zones the ecological carrying capacity should be increased by providing access tracks if possible, or if not possible, by restricting access. A historical account of the process of dune stabilization showed that although first initiated in 1845, indigenous species were only used in the past three decades. The use of alien species has resulted in problems such as a reduction in the ecological integrity and aesthetic appeal of coastal systems. The techniques applied in the stabilization of drift sands with indigenous vegetation have been successful, as revealed by a quantitative survey of 17 sites in the Eastern Cape. Sites were grouped by multivariate analysis on the basis of their species composition, and variability between sites was dependent on the types of species planted. Selection of suitable species is therefore important and is discussed with respect to their natural distribution along the coast. The long term objective of stabilization should be the creation of functional, diverse, aesthetic ecosystems, since the intrinsic and economic value of the dune landscape for tourism lies therein. However, detailed studies should be undertaken prior to implementing a manipulative process such as dune stabilization, since ecological processes may be disrupted. An understanding of such processes is therefore important if one wishes to effectively manage the dune landscape.

A study of the dynamics and functions of a coastal dunefield at the Bushman’s River Mouth in Kenton-on-Sea was conducted to provide essential information for the formulation of a management strategy based on an ecosystem approach. Bi-monthly tacheometric surveys were undertaken at the Westbourne Road Car Park dunefield to investigate the sedimentological processes. Changes in the topography, the amount and rate of sand accretion were monitored. Accretion and progradation were the dominant long-term sedimentological processes at the site. Long-term spatial variation in these processes corresponded to the temporal pattern of sedimentation in the river mouth while short-term variation resulted from dynamics inherent to the types of dunes present. Factors affecting aeolian accretion at the site are topography, river mouth configuration, dune types and the presence of vegetation as well as the seasonal wind regime. A vegetation study, recording species present, cover and abundance of the plants along the surveyed transects, was also conducted to examine the progressive establishment of vegetation. Five plant communities were discerned namely, saltmarsh, pioneer, enriched pioneer, open scrub, closed scrub and scrub-thicket communities. The spatial distribution of these communities coincided with the north-westerly development of the sandbank. A successional trend in community change occurred along a gradient of increasing diversity, complexity and age. Initially, autogenic changes wrought by pioneer species facilitated the recruitment and survival of other species and commenced succession at the study site. Multiple successional pathways were discussed in relation to subsequent species establishment that occurred after facilitation. The amalgamation of ecological data with a social study in the form of public perception surveys allowed for a greater understanding of the present interactions between the ecosystem and the demands placed on it. The area primarily functions as a recreational site that offered numerous beach and water-related activities. A lack of information regarding the ecology and management programmes influenced public perception of the environment and management issues in general. Different management perspectives and the use of the information presented in this study were also discussed.

Bibliography: pages 144-155. / Dunes are an integral part of the Walvis Bay environment. Their dynamics significantly influence the functioning of the coastal system and have major implications for both the ecological and socio-economic environments. In light of residential, commercial and industrial requirements for the Walvis Bay area, dune encroachment needs to be seriously considered in the future planning of land-use and resource utilisation. The overall objective of this report, therefore, is to provide a better understanding of the dominant role dunes and sand drift play in the study area and in so doing promote the more judicious future planning and management of the region. In order to achieve this aim, several specific objectives were addressed, these being: to identify and clarify all dune types occurring in the study area, and discuss the geomorphological and ecological characteristics of each; to provide an analysis of the long-term stability of dunes in the Walvis Bay area by means of aerial photograph interpretation; to highlight all factors that contribute to dune encroachment and to assess the ramifications of this movement upon the ecological and socio-economic environments; * to emphasise elements that need to be considered with any future planning of land-use and development; * to review dune management practices in the Walvis Bay area and provide recommendations for the more effective management of the system in the future.

This study examines foredune evolution along a 2100 m section of coast adjacent to the Tugela River. The foredunes vary in both height and shape along the study area and form the southern most extension of the Tugela foredune-ridge plain. Sand accumulation and erosion was measured at regular intervals over a 30 month period by tacheometric surveys. The foredunes and beaches are comprised of over 99 % sand. The sediment was predominantly composed of quartz and feldspar with subordinate lithic fragments. The quartz grains display conchoidal fractures and mechanical v-shaped pits and curved grooves. The beach and dune sand is well sorted and slightly negatively skewed with a mean grain size of 1.62 ф. The vegetation structure and floristic composition of the foredunes are explored. A range of factors influencing foredune morphology and evolution, including canopy density, height and distribution, wind velocity and a variety of ecological and environmental processes are examined. Ridge and swale morphology as well as alongshore variation in the dunefield could not be related to biological processes. The development of a foredune-ridge topography depends on a large sediment supply from the Tugela River over the long-term. Periods of high discharge introduce a fresh source of sediment to the littoral zone. Reworking of fluvial sediment landwards results in wide beaches. Onshore winds transport the sand from the beaches to the foredunes. Scaevola thunbergii encourages rapid vertical accretion and hummock dunes are formed. Lateral extensive invasion by seedlings may result in the hummock dunes joining to form coast parallel foredunes. Under periods of reduced sediment discharge erosion of the shoreline results in steep narrow beaches. Despite a negative beach budget foredunes continue to accrete vertically. Marine erosion results in either the complete destruction of embryo foredunes or their landward shift. Natural breaks in the dune crestline were attributed to changes in the delivery of sediment to the beaches. The processes operating in the study area conform to Psuty's (1988,1989) sediment budget model of foredune development. Sediment availability to the coastline produces characteristic morphologies. / Thesis (M.Sc.)-University of Natal, Durban, 1998.

Rehabilitation is a dynamic process influenced by factors related to more than one field of ecology. It is therefore necessary to consider all these components when assessing the rehabilitation, although in the initial stages the successful revegetation of the disturbed areas is the most important criterion. Richards Bay Minerals, on whose mining site this project was carried out, is dredge mining heavy minerals on the north coast of Natal, where they have rehabilitated mined areas since 1978. This project has been carried out to establish: 1) The success of their dune forest rehabilitation using quantitative techniques. 2) The available seed bank in their rehabilitation stands. 3) The similarities in the succession taking place in rehabilitation stands compared to the revegetated stands in the vicinity of Richards Bay. 4) The best methods for creating alternative vegetation communities, especially grasslands, with a high species diversity on the mined tailings. This study reviews only the success of rehabilitation of the natural vegetation but other studies focusing on the insect, reptile, mammal and bird populations are also being undertaken by other researchers. No particular method of determining the success of vegetation rehabilitation has been chosen by restoration ecologists. Therefore in this study a broad range of quantitative techniques were used to show whether successional changes are occurring in the vegetation and physical environment. The results obtained from sampling the rehabilitated vegetation have shown that both the species richness and diversity are increasing as the returned vegetation matures. Levels of soil properties such as Sodium, Phosphate, Calcium and percentage organic matter have also risen with increasing stand age. Community complexity is also increasing with stand age, and TWINSPAN and DECORANA plots have separated out the differently aged stands based on their differences. A "pilot" study was done on the seed bank present in the rehabilitation stands. This has shown the presence of large amounts of early successional, mostly herbaceous species. Seeds of later successional and woody species were scarce which may be a result of the sampling intensity used. However seeds of late successional ground cover species were found in the older stands. Comparisons between the natural revegetation of disturbed areas in the vicinity of Richards Bay and the rehabilitation stands revealed similarities in both species composition and complexity. Species richness and diversity values are comparatively similar for the younger revegetated and older rehabilitation stands, and lWlNSPAN and DECORANA analysis techniques clustered the samples recorded from these areas in close proximity on their relative plots. The oldest revegetated sites contain a number of species found in the rehabilitated vegetation but as Acacia karroo has thinned-out in these stands many of these other species are now mature individuals. Attempts at rehabilitating an area of grassland at Richards Bay Minerals has not produced satisfactory species diversity and experimental manipulations were used to try and increase the diversity of the existing Eragrostis curvula dominated community. Of the several treatments used for the manipulation, a combination of burning and further topsoiling was the most successful in reducing Eragrostis importance and in increasing the species richness. Grassland topsoil spread directly onto the bare tailings produced an extensive vegetation covering over a short period but species richness was not significantly greater than for the existing Eragrostis dominated grassland, and further treatments and management needs to continue if this technique is to be employed. Only a limited amount of alien infestation of the rehabilitated areas was evident from the sampling undertaken during this research. As the removal of alien plants is an ongoing process and the rehabilitation stands are continuously monitored to identify any new invaders, this is not expected to become a problem. From the results of work done overseas and the rehabilitation carried out in South Africa it appears that it is possible to return natural vegetation communities on mined areas. That this is a lengthy process is to be expected but by manipulating the vegetation and continuously monitoring the process it may be possible to speed up development. Areas in need of further research have been identified based on the findings of this project. This will help to reinforce the undertaking of management proposals that will enhance the vegetation recovery and the success of the rehabilitation programme.