A dissertation submitted to the faculty of Science, University of the Witwatersrand, Johannesburg, in fulfillment of the requirements for the degree of Master of Science. / Unable to load abstract.

Thesis submitted in fulfilment of the requirements for the degree of Doctor of Philosophy. School of Animal, Plant and Environmental Science, University of the Witwatersrand, Johannesburg, South Africa. October 2016. / Mining has negative impacts on the environment in many different ways. One method developed to quantify some of these impacts is Landscape Function Analysis (LFA) and this has been accepted by some mining companies and regulators. In brief, LFA aims at quantifying the organization of vegetative and landscape components in a landscape into patches along a transect and quantifying, in a relative manner, three basic processes important to landscape functioning, namely: soil stability or susceptibility to erosion, infiltration or runoff, and nutrient cycling or organic matter decomposition. However, LFA is limited in large heterogeneous environments, such as those around mining operations, due to its localized nature, and the man hours required to collect a representative set of measurements for such large and complex environments. Remote sensing using satellite-acquired data can overcome these limitations by sampling the entire environment in a rapid and objective manner. What is required is a method of connecting these satellite-based measurements to LFA measurements and then being able to extrapolate these measurements across the entire mine surface. The aim of this research was to develop a method to use satellite-based hyperspectral imagery to predict landscape function analysis (LFA) using partial least squares regression (PLSR). This was broken down into three objectives: (1) Collection of the LFA data in the field and validation of the LFA indices against other environmental variables collected at the same time, (2) validation of PLSR models predicting LFA indices and various environmental variables from ground-based spectra, and (3) production of risk maps based on predicting LFA indices and above-ground biomass using PLSR models and Hyperion satellite-based hyperspectral imagery. Although the study was based in grasslands at two mining regions, West Wits and Vaal River, a suitable Hyperion image was only available for Vaal River. A minimum of 374 points were sampled for LFA indices, ground-based spectra, above-ground biomass and soil cores along 2880 m of LFA transect from both mine sites. Soil cores were weighed fresh before sieving with a 2 mm sieve to separate root and stone fractions. The sieved soil fraction was tested for pH, EC, SOM, and for the West Wits samples, organic nitrogen and total extractable inorganic nitrogen. There was one modification to the LFA method where grass patches were collapsed into homogenous units as it was deemed not feasible to sample 180 m transects at grass tuft scales of 10 – 30 cm, but other patch definitions followed the LFA manual (Tongway and Hindley, 2004). Evidence suggested that some of the different patch types, in particular the bare/biological soil crust – bare grass – sparse grass patch types, represented successional stages in a continuum although this was not conclusive. There also was evidence that the presence or absence of cattle play a role in some processes active in these grasslands and erosion is mainly through deflation, rain splash and sheet wash. Generally the environmental variables supported the LFA indices although the nutrient cycling index was representative of above-ground nutrient cycling but not below-ground nutrient cycling. Models derived with PLSR to predict the LFA indices from ground-based spectral measurements were strong at both mine sites (West Wits: LFA stability r2 = 0.63, P < 0.0001; LFA infiltration r2 = 0.75, P < 0.0001; LFA nutrient cycling r2 = 0.73, P < 0.0001; Vaal River: LFA stability r2 = 0.39, P < 0.0001, LFA infiltration r2 = 0.72, P < 0.0001, LFA nutrient cycling r2 = 0.54, P < 0.0001), as were PLSR models predicting above-ground biomass (West Wits above-ground biomass r2 = 0.55, P = 0.0003; Vaal River above-ground biomass r2 = 0.79, P < 0.0001) and soil moisture (West Wits soil moisture r2 = 0.45, P = 0.0017; Vaal River soil moisture r2 = 0.68, P < 0.0001). However, for soil organic matter (r2 = 0.50, P < 0.0001) and EC (r2 = 0.63, P < 0.0001), Vaal River had strong prediction models while West Wits had weak models for these variables (r2 = 0.31, P = 0.019 and r2 = 0.10 and P < 0.18, respectively). For EC, the wide range of soil values at Vaal River in association with gypsum crusts, and low values throughout West Wits explained these model results but for soil organic matter, no clear explanation for these site differences was identified. Patch-based models could accurately discriminate between spectrally well-defined patch types such S. plumosum patches but were less successful with patch types that were spectrally similar such as the bare/biological soil crust – bare grass – sparse grass patch continuum. Clustering similar patch types together before PLSR modelling did improve these patch-based spectral models. To test the method proposed to predict LFA indices from satellite-based hyperspectral imagery, a Hyperion image matching 6 transects at Vaal River was acquired by NASA’s EO-1 satellite and downloaded from the USGS Glovis website. LFA transects were partitioned to match and extract pixel spectra from the Hyperion data cube. Thirty-one spectra were separated into calibration (20) and validation (11) data. PLSR models were derived from the calibration data, tested with validation data to select the optimum model, and then applied to the entire Hyperion data cube to produce prediction maps for five LFA indices and above-ground biomass. The patch area index (PAI) produced particularly strong models (r2 = 0.79, P = 0.0003, n =11) with validation data, whereas the landscape organization index (LOI) produced weak models. It is argued that this difference between these two essentially similar indices is related to the fact that the PAI is a 2-dimensional index and the LOI is a 1-dimensional index. This difference in these two indices allowed the PAI to compensate for some burned pixels on the transects by “seeing” the density pattern of grass tufts and patches whereas the linear nature of the LOI was more susceptible to the changing dimensions of patch structure due to the effects of fire. Although validation models for the three LFA indices of soil stability, infiltration and nutrient cycling were strong (r2 = 0.72, P = 0.004; r2 = 0.66, P = 0.008; r2 = 0.70, P = 0.005, n = 9 respectively), prediction maps were confounded by the presence of fire on some transects. The poor quality of the Hyperion imagery also meant great care had to be taken in the selection of models to avoid poor quality prediction maps. The 31 bands from the VNIR (478 – 885 nm) portion of the Hyperion spectra were generally the best for PLSR modelling and prediction maps, presumably because of better signal-to-noise ratios due to higher energy in the shorter wavelengths. With two satellite-based hyperspectral sensors already operational, namely the US Hyperion and the Chinese HJ-1A HSI, and a number expected to be launched by various space agencies in the next few years, this research presents a method to use the strengths of LFA and hyperspectral imagery to model and predict LFA index values and thereby produce risk maps of large, heterogeneous landscapes such as mining environments. As this research documents a method of partitioning the landscape rather than the pixel spectra into pure endmembers, it makes a valuable contribution to the fields of landscape ecology and hyperspectral remote sensing. / LG2017

La rapide augmentation des gaz à effet de serre (GES) dans l’atmosphère - dont le CO2 – due aux activités humaines est considérée comme responsable des changements climatiques en cours et futurs. Les écosystèmes terrestres sont potentiellement des "puits" importants de C et pourraient contribuer à l'atténuation des GES. Les prairies permanentes (steppes, savanes, prairies de montagne, ...) couvrent 40% de la surface terrestre (hors calotte glaciaire) et leurs sols représentent potentiellement un énorme "puits" permettant de stocker du C naturellement (GIEC 2001). Cependant, les processus impliqués et leur régulation restent à préciser. L’objectif de la thèse était d’analyser l’effet des pratiques de pâturage sur le stockage de C dans le sol. Cette analyse a été réalisée à partir de données acquises sur deux dispositifs ‘long terme’ en prairie permanente (SOERE ACBB) et en s’intéressant aux flux de C entre les différents compartiments de l’agro-écosystème sous différentes intensités de pâturage afin i) d’étudier notre capacité à estimer le stockage de C dans le sol après 10 ans d’application de traitement, en comparant deux méthodes (méthode utilisant des tours à flux et mesure du stock de C du sol) ii) d’apporter des connaissances sur les mécanismes et régulations agissant sur les dynamiques de stockage du C. Les résultats de la comparaison des deux méthodes de mesures testées ont indiqué une séquestration nette de C dans le sol, avec un taux de séquestration moyen mesuré avec les deux méthodes de 2.21 t C ha-1 an-1 et de 2.29 t C ha-1 an-1, sans différence significative entre traitements, mais avec une tendance à une séquestration plus élevée avec la gestion plus intensive. Chaque méthode permet d’accéder à des informations différentes. L'approche avec les tours à flux permet d'identifier des interactions entre le climat et les pratiques de gestion sur les flux de C dans les prairies. Les inventaires de sol ont permis de montrer que le carbone se stocke également dans les couches plus profondes de sol. Alors que les communautés végétales ont évolué sous l’effet des traitements différenciés de pâturage, les mesures ne montrent pas d’évolution des stocks de C totaux ni des matières organiques particulaires. L’analyse des flux de C entre les différents compartiments de l'écosystème, après 7 ans d’application des traitements, montre que les traitements avec une intensité faible ou nulle ont conduit à une réduction des flux de carbone entre les compartiments du continuum de dégradation du C, tandis que les stocks de carbone des racines et des POM ne sont pas affectés par les traitements. Une étude complémentaire conduite pour estimer les productions racinaires indique que la réponse des racines (stocks et production) et des stocks de matières organiques particulaires pourrait être en partie découplée de la réponse du compartiment aérien de la végétation. A l’issue de cette étude, il nous apparaît qu’une approche plus intégrative du fonctionnement de l’écosystème est nécessaire pour accroître notre capacité de prédiction de l’impact des pratiques sur le stockage du C en prairie. / The fast increase of greenhouse gases in the atmosphere, such as carbon dioxide, due to human activities is consider as the main cause of actual climate change. Terrestrial ecosystem are considered as a huge "sink" of C and may contribute to decrease greenhouse gases. Permanent grasslands cover 40% of land and their soil may contribute to sequester C (GIEC 2001). However, the processes involved and their regulations remain to be specified. The aim of the thesis was to analyze the effect of grazing management on soil C storage. This analysis was made from data acquired on two long term permanent grassland sites (SOERE ACBB) and by studying C fluxes between the different agroecosystem compartments under different grazing intensities for i) estimating our capacity to measure soil C storage after 10 years of grazing treatments by comparing two methods (soil inventories vs net carbon storage measurements), ii) to provide knowledge on the mechanisms and regulations affecting the dynamics of soil C sequestration. Comparing results of both methods, measurements indicated a net C sequestration in soil, with an average sequestration rate of 2.21 t C ha-1 yr-1 and 2.29 t C ha-1 year-1 and no significant difference between treatment but a tendency to a higher sequestration with more intensive management. Each method provides access to different information. The approach with flux towers allows a better understanding of the role and interactions between climate and practices on C fluxes in grasslands. Soil inventories showed carbon is store in deeper soil layers. While plant communities have evolved as a result of differentiated grazing treatments, measurements show no changes in total C stocks and particulate organic matter. Analysis of C fluxes after 7 years of differentiate grazing treatments, showed that treatments with low or zero grazing intensity led to a reduction of carbon fluxes between the compartments of the continuum of degradation, while carbon stocks in roots and POM were not affected by treatments. A complementary study conducted to estimate root production indicates that the response of roots (stocks and production) and stocks of particulate organic matter may be partly decoupled from the response of the aerial vegetation compartment. This study indicates that a more integrative approach on ecosystem functioning is necessary to increase our ability to predict the impact of management practices on C storage in grassland.

Os afloramentos rochosos ferruginosos denominados bancadas lateríticas podem ser considerados uma paisagem peculiar no Centro-Oeste do Brasil, próximos às cidades de Corumbá e Ladário, MS. Embora situada em uma região populosa, próxima a fazendas e sítios urbanos, esta vegetação ainda é desconhecida. Neste estudo avaliamos a composição florística e a estrutura da comunidade, bem como suas relações florísticas com outros afloramentos rochosos. O levantamento foi conduzido em três locais - Banda Alta (19°08´S, 57°34´W, 85 m altitude); São Sebastião do Carandá (19°06´S, 57°31´W, 90 m) e Monjolinho (19°16´S, 57°31´W, 65-150 m). A vegetação encontrada nestes afloramentos rochosos ocorre como uma cobertura contínua sobre o substrato rochoso ou como ilhas de solo. Para o estrato contínuo, a vegetação foi amostrada em parcelas de 1 m² distribuídas aleatoriamente. A própria ilha de solo foi considerada uma unidade amostral. A composição de espécies e os valores de cobertura foram avaliados em cada parcela ou ilha de solo. Avaliaram-se 199 parcelas e 164 ilhas de solo neste estudo. Além da abundância das espécies o tamanho e outras características das ilhas de solo foram incluídas neste levantamento. Os dados foram explorados através de tabelas fitossociológicas e de técnicas estatísticas de classificação e ordenação. As relações florísticas foram exploradas nos níveis taxonômicos de famílias e gêneros pela comparação de listas de espécies publicadas de outros afloramentos. Foram amostradas 188 espécies pertencentes a 58 famílias nas três bancadas lateríticas. Deste total, 66 espécies foram encontradas nas parcelas e 181 nas ilhas de solo. A estrutura da comunidade é influenciada, principalmente, pela presença de fina camada de substrato e pela área insular. Processos envolvendo competição exclusiva e diferentes respostas ao regime hídrico aparentemente podem ser as maiores fontes da separação espacial e padrões de abundância. A flora das bancadas lateríticas mostrou similaridade com a flora de afloramentos graníticos da Caatinga, em nível de famílias e gêneros. Condições climáticas e outros fatores podem influenciar as disjunções florísticas observadas mas o substrato rochoso parece ser o fator determinante, com exceção das bancadas lateríticas. Esperava-se que as bancadas fossem similares a outros afloramentos ferruginosos mas isto não ocorreu, sugerindo um importante efeito da história paleoclimática da plataforma continental nos padrões de composição e abundância de espécies da comunidade vegetal das bancadas lateríticas. / The ironstone outcrops called bancadas lateríticas can be considered a peculiar feature of the Central Brazil landscape nearby the cities of Corumbá and Ladário (MS). Although situated in a populated region, near farmlands and cities, the vegetation occurring on these outcrops is completely unknown. In this work we aim to explore the species composition, the floristic relations with other rock outcrops and the community structure of the vegetation thriving on the ironstone surface of the bancadas. The survey was conducted in three outcrops - Banda Alta (19°08´S, 57°34´W, 85 m altitude); São Sebastião do Carandá (19°06´S, 57°31´W, 90 m) and Monjolinho (19°16´S, 57º31´W, 65-150 m). The vegetation found on the rock outcrops occurred as a continuous cover or in groups called soil-islands. In the continuous formation, the vegetation cover was sampled by random Quadrats of 1 m2. A single soil island was taken as a sample unit. The species composition and correspondent cover value was evaluated for each quadrat and soil island. A total of 199 random quadrats and 164 soil islands was used in this work. Along with the abundance of species, the size and other features of each soil islands were included in the survey. The data was explored by the construction of phytosociological tables and statistical classification and ordination techniques. The floristic relations were explored at the level of families and genera by comparisons with published species lists from other outcrops. We detected 188 species belonging to 58 plant families on the three Bancadas outcrops. From this total, 66 species were found in random quadrats and 181 in soil islands. The community structure is primarily influenced by the substrate thickness and the soil island area. Processes involving apparent species competitive exclusion and different responses to water regime can be major sources of the overall spatial disjunction and abundance patterns. The bancadas flora showed similarities with the flora of granite outcrops in the Caatinga formation at the level of genera and families. Climate and other factors can play a role in the floristic disjuntions observed but the rock substrate poses as a dominant factor with the exception of the Bancadas. The departure from the expected similarity with other ironstone outcrops suggests an important effect of Paleoclimatic history of the continental shelf in the patterns of species composition and abundance in the Bancadas plant community.

O Bioma Pampa representa aproximadamente 63% do território do Estado do Rio Grande do Sul - Brasil caracteriza-se pela alta biodiversidade de espécies vegetais e por sua formação predominantemente campestre. Em função da contínua incorporação de espécies exóticas, monoculturas e a práticas por vezes inadequadas de manejo pastoril para produção pecuária, os campos do Bioma Pampa estão rapidamente sendo degradados, fragmentados e descaracterizados. A Biomassa é uma das variáveis biofísicas estratégicas de interesse em estudos de controle, monitoramento e estimativas da vegetação campestre. O objetivo principal deste estudo é contribuir no desenvolvimento de novas estratégias de manejo e monitoramento adequados da vegetação campestre e inferir regressão linear multivariada para estimar a biomassa dos campos nativos a partir de dados remotos e dados de campo. As avaliações foram realizadas em área pertencente à Universidade Federal de Santa Maria, localizada na região central do Rio Grande do Sul. Os manejos pastoris correspondem a duas somas térmicas acumuladas, em dias, de 375 e 750 graus-dias (GD), que determinaram os intervalos entre pastoreio. Os dados remotos utilizados foram oriundos de imagem MSI do Satélite Sentinel-2 e dados de espectrorradiômetro com amplitude de 350-2500 nm. Verificou-se que as duas técnicas apresentaram resultados satisfatórios, em que inferiu-se regressões com r²ajustado = 0.65 para estimar biomassa verde e regressões com r²ajustado = 0.61 para biomassa total e biomassa senescente. Desta forma, o estudo verificou que é possível minimizar os esforços de campo para auxiliar no monitoramento, organização e conservação dos campos nativos do Bioma Pampa utilizando dados de sensoriamento remoto como ferramenta de manejo buscando a sustentabilidade destes complexos ambientes naturais. / The Pampa Biome represents approximately 63% of the territory in the State of Rio Grande do Sul - Brazil, characterized by the high biodiversity of plant species and its predominantly rural formation. As a result of the continuous incorporation of exotic species, monocultures and excessive cattle production, the Pampa Biome fields are rapidly being degraded, fragmented and decharacterized. Biomass is one of the strategic biophysical variables of interest in studies of prevention, monitoring and estimates of the country vegetation. The main objective of this study is to contribute to the development of new strategies for the proper management and monitoring of the country vegetation and to infer multivariate linear regression to estimate the biomass of the natural grassland in native fields from remote data and field data. The evaluations were carried out in an area belonging to the Federal University of Santa Maria, located in the central region of Rio Grande do Sul. Pastoral management correspond to two accumulated thermal sums in days of 375 and 750 degrees-days (GD), which determined the intervals between grazing. The remote data used came from MSI image of Sentinel-2 Satellite and spectroradiometer data with amplitude of 350-2500 nm. It was verified that the two techniques presented satisfactory results, in which regressions with adjusted r² of 0.65 were estimated to estimate green biomass and regressions with adjusted r²> 0.61 for total biomass and dry biomass. In this way, the study verified that it is possible to minimize the field efforts to assist in the monitoring, organization and conservation of the natural grassland in native fields of the Pampa Biome using remote sensing data as a management tool seeking the sustainability of these complex natural environments.

Submitted by Maicon Juliano Schmidt (maicons) on 2015-03-24T18:49:27Z No. of bitstreams: 1 DissertacaoGianeINiderauer.pdf: 4504038 bytes, checksum: c909b35b56b6c4307e28d588aa4527a0 (MD5) / Made available in DSpace on 2015-03-24T18:49:27Z (GMT). No. of bitstreams: 1 DissertacaoGianeINiderauer.pdf: 4504038 bytes, checksum: c909b35b56b6c4307e28d588aa4527a0 (MD5) Previous issue date: 2012-06-27 / Banco Santander / Banespa / O presente trabalho foi desenvolvido no Centro de Pesquisas e Conservação da Natureza (Prómata), localizado no município de São Francisco de Paula (29º28?S; 50º13?W), a nordeste do Estado do Rio Grande do Sul, em área de Floresta Ombrófila Mista e Mosaicos de Campos, com objetivo de avaliar a existência de sinais históricos do avanço da floresta com base na estrutura etária e no desempenho de crescimento de Araucaria angustifolia, em um gradiente floresta-campo. A Araucaria angustifolia é uma espécie que apresenta anéis de crescimento anuais bem definidos é indicada para estudos dendrocronologicos. Neste sentido, foram selecionados 21 pontos amostrais em uma área aproximada de 25 ha, selecionados aleatoriamente em três níveis de distâncias: pontos em floresta a mais de 50 m de distância da borda mais próxima, (2) pontos em floresta ou campo a até 50 m de distância da borda mais próxima e (3) pontos em campo com mais de 50 m de distância da borda. Para análises dendrocronologicas, em cada ponto foi selecionado um indivíduo por classe de diâmetro à altura do peito para coleta de duas a quatro amostras de lenho com trado de incremento. Em laboratório às amostras foram fixadas em suportes de madeira, polidas com uma série de lixas, e analisadas com auxilio de um estereomicroscópio. Posteriormente, foram realizadas análises de regressão linear e de aleatorização para avaliar como a idade e o crescimento (em diâmetro e altura) dos indivíduos variam em relação à distância da borda floresta-campo. Os resultados mostraram que a estrutura etária e o desempenho de crescimento de A. angustifolia variavam ao longo do gradiente floresta-campo. A idade das A. angustifolia variou de 12 a 106 anos, decrescendo do interior da floresta para o interior do campo, a posição das árvores explicou 15 % da variação etária (P=0,0823). A idade teve influência no tamanho das árvores, explicando 43 % da variação do diâmetro do tronco (P=0,001) e 12 % da variação da altura viii (P=0,114). Conclui-se que, a estrutura etária e crescimento em altura da A. angustifolia variam no gradiente floresta-campo. / This work was developed at the Centro de Pesquisas e Conservação da Natureza (Pró-Mata), located in the municipality of São Francisco de Paula (29º28’S; 50º13’W), the northeastern of Estado Rio Grande do Sul, in a forest-grassland mosaic habitat. Vê of this study was to evaluate for signs of historical advancement of Forest based on age structure and growth performance of Araucaria angustifolia in a forest-grassland gradient. The Araucaria angustifolia is a species to present annual growth rings well defined is indicated for dendrochronological studies. In this sense, we selected 21 sample points in area of approximately 25 ha, randomly selected from three levels of distances. (1) points in the forest more than 50 m away from the nearest edge, (2) points in the forest or grassland up 50 m away from the nearest adge and (3) points on the grassland with more than 50 m away edge. To dendrochronological analyzes, at each point was selected for individual class diameter at breast height to collect two to four samples of wood with increment borer. Age had influence on tree size, explaining 43% of the variation of stem diameter (P= 0.001) and 12 % of the variation in height (P= 0.114). It was concluded that the structure age and height growth of A. angustifolia vary in forest-grassland gradient. In the laboratorio samples were fixed on a Wood France and polished with a series of sandpapers, and analyzed with help of a stereomicroscope. Later, we performed linear regression analyzes and randomization to assess how the age and growth (height and diameter) individuals vary in distance from the forest-grassland edge. The results showed that the age structure and growth performance of A. angustifolia varied along the gradient forest-grassland. Age of A. angustifolia ranged from 12 to 106 years, decresing from the interior of the forest for the interior grassland, the position of the trees 15 % of variance explained group (P= 0.0823). Age had influence on tree size, explaining 43% of the variation of stem diameter (P= 0.001) and 12 % of the x variation in height (P= 0.114). It was concluded that the structure age and height growth of A. angustifolia vary in forest-grassland gradient.

O objetivo geral desta tese foi examinar diferentes 1 questões associadas ao 2 processo de invasão de plantas para investigar os mecanismos, impactos e 3 medidas de recuperação da comunidade vegetal, abordando diferentes estratégias 4 metodológicas que incluem estudos observacionais, experimentos de remoção e 5 uma revisão metodológica. Os resultados do Capítulo 1 mostraram como as 6 relações entre condições climáticas e estrutura da paisagem podem determinar o 7 grau de invasão de plantas na escala regional. Observou-se que os padrões de 8 invasão dos campos sulinos estão principalmente relacionados com maior 9 densidade de estradas, menor cobertura de campo nativo e com o aumento do 10 déficit hídrico. Além disso, constatou-se que a gramínea Eragrostis plana é a 11 planta invasora mais abundante dos campos sulinos. No entanto, os resultados do 12 experimento do Capítulo 2 mostraram que sua invasão não pode ser explicada 13 pela riqueza de espécies ou composição dos grupos funcionais de plantas da 14 comunidade residente. A invasão de E. plana foi principalmente associada ao 15 distúrbio causado pela remoção de biomassa na comunidade. Neste sentido, as 16 comunidades campestres poderiam ser resistentes à invasão de E. plana até que 17 algum distúrbio aumente sua vulnerabilidade. Por outro lado, através da 18 comparação entre comunidades invadidas, removidas e não-invadidas, os 19 resultados do experimento do Capítulo 3 contribuíram no entendimento do 20 impacto real da invasão de E. plana, em termos de redução da riqueza e cobertura 21 de plantas nativas na comunidade. No entanto, embora os métodos de remoção 22 utilizados reduziram a cobertura da invasora, não foram suficientes para conseguir 23 sua erradicação local. Além disso, após três anos de remoção de E. plana, as comunidades se tornaram distintas às invadidas mas não 1 foram semelhantes às 2 comunidades não invadias, o qual poderia indicar que outras medidas de 3 restauração são ainda necessárias. Embora os experimentos de remoção de 4 espécies sejam úteis para investigar questões associadas ao processo de 5 invasão, existem limitações importantes a considerar, como foi evidenciado nos 6 Capítulos 2 e 3. Neste sentido, oferecemos a revisão bibliográfica sistemática do 7 Capítulo 4, onde se discute o potencial dos métodos de remoção utilizados para 8 estudar a resistência e a recuperação da comunidade à invasão, apontando 9 algumas limitações. Como resultado, o Capítulo 4 mostrou que a maioria dos 10 trabalhos não propõem controles adequados nos experimentos, o que pode dar 11 lugar a confundimento de efeitos. Desta forma, foram desenvolvidas algumas 12 sugestões para serem consideradas nos experimentos de remoção de espécies, 13 com o objetivo de continuar avançando nesta temática. As informações geradas 14 nesta tese podem contribuir para o entendimento do processo de invasão de 15 plantas nos ecossistemas campestres, com vistas ao manejo, à conservação e à 16 restauração das comunidades invadidas, adquirindo um senso crítico no 17 planejamento de desenhos experimentas. / The general aim of this thesis was to investigate different 1 issues associated 2 with plant invasion process to understand the mechanisms, impacts and 3 community recovery, by employing different methodological strategies such as 4 observational studies, removal experiments and a literature review. The results 5 from Chapter 1 showed how the interactions between climate and landscape 6 structure can determine the level of invasion of South Brazilian grasslands, 7 highlighting that invasion is mainly related to high road density, less native 8 grassland cover and increased aridity. Furthermore, it was confirmed that 9 Eragrostis plana is the most important invasive species in the South Brazilian 10 grasslands. However, the results from the experiment of Chapter 2 showed that its 11 invasion could not be explained by the species richness or functional group 12 composition in the community. Eragrostis plana invasion was associated with the 13 disturbance effect caused by the amount of removed biomass. Thus, grassland 14 communities may be resistant to E. plana invasion until some disturbance 15 increases their vulnerability. Moreover, by comparing invaded, removed and non16 invaded communities, the experimental results of Chapter 3 highlighted the 17 ecological impact of E. plana invasion, in terms of richness reduction and native 18 species cover. However, although removals methods reduced the cover of the 19 invasive species, they were not enough to locally extinct it. Moreover, after three 20 years of the invasive removal, communities became different from invaded ones 21 but not resembling non-invaded references, which suggest that community 22 recovery may require restoration strategies. Although removal experiments have 23 been useful to investigate certain issues associated with invasion process, there are important limitations to consider, as was shown in Chapter 1 2 and 3. For this 2 purpose, our systematic review presented in Chapter 4 discussed the potential of 3 removal methods for assessing community resistance and recovery from invasion. 4 There, we showed that most of the studies did not use adequate controls in 5 removal experiments, which can lead to confounding effects. Thus, we developed 6 suggestions to be considered in experimental designs to advance the 7 methodological technique of removals. The information generated in this thesis can 8 contribute to the understanding of plant invasion process in South Brazilian 9 grasslands and, consequently, aid to management, conservation and restoration of 10 invaded communities by acquiring a critical sense in experimental designs.

O pastejo é considerado um fator determinante para a manutenção e conservação da vegetação campestre (campos) no sul do Brasil, sendo considerado o responsável por construir a heterogeneidade estrutural, a composição de espécies e a diversidade dos campos. Porém o manejo do pastejo é fundamental para sucesso da atividade pecuária, uma vez que longos períodos de pastejo excessivo e intenso ao invés de manter áreas campestres pode degradá-las, diminuindo a diversidade e produtividade, tornando os campos suscetíveis a invasão por espécies exóticas. Para o sucesso da relação construída entre o manejador, os animais e a vegetação, as estratégias de manejo que possibilitem uma pecuária rentável, mas que busque conservar a vegetação campestre, são fundamentais. O objetivo desta tese foi verificar os efeitos do pastejo sobre a vegetação campestre, tendo como ferramenta principal para isso, diferentes manejos. Os efeitos do pastejo foram avaliados através de índices de diversidade taxonômica e funcional, e da variação de atributos funcionais de gramíneas, analisando especificamente: (1) como estratégias de manejo contínuo e rotativo influenciam a dinâmica da vegetação campestre (biomassa, diversidade, riqueza e equitabilidade); (2) a influência do manejo do pastejo na composição e diversidade funcional de comunidades campestres descritas por gramíneas, e os efeitos sobre a variação intra- e interespecífica de atributos foliares (SLA e LDMC) das espécies de gramíneas; (3) o efeito de diferentes intensidades de pastejo (ofertas de forragem) nas diversidades alfa e beta das comunidades de plantas. Foram verificadas mudanças na diversidade, equitabilidade, riqueza de espécies, biomassa condicionadas pelo pastejo rotativo, que se mostrou benéfico para vegetação campestre. O pastejo rotativo proporcionou também um incremento na diversidade funcional, condicionado pela variação intra- e interespecífica nos atributos das gramíneas. Diferentes intensidades de pastejo promoveram diferenças de alfa e beta diversidade sob diferentes intensidades de pastejo. O pastejo demonstrou exercer papel determinante em toda a dinâmica campestre, tendo efeitos consideráveis em todos os níveis de avaliação, desde a riqueza de espécies até na diversidade funcional, ficando evidente que a decisão do manejo é fundamental tanto para a produtividade quanto para a manutenção da vegetação campestre. / Grazing is considered a determining factor for the maintenance and conservation of grasslands in southern Brazil and is considered one of the main driver of the structural heterogeneity, species composition and diversity of the grasslands. But the management of grazing is key to the success of livestock farming, since long periods of excessive and intense grazing instead of maintaining grassland, can degrade them, reducing the diversity and productivity, leading the grassland susceptible to invasion of exotic species. For the success of the relationship built between managers, animals and vegetation, a strategic management that allows an efficient farming together with the conservation of grassland vegetation is essential. The objective of this thesis was to investigate the effects of grazing on the grassland, based on different managements of grazing. The effects of grazing were evaluated by taxonomic and functional diversity indices, and the variation of functional traits of grasses, analyzing specifically: (1) how continuous and rotational grazing management influence the dynamics of grassland vegetation (biomass, diversity, richness and evenness); (2) the influence of grazing management on the functional composition and functional diversity in grassland communities described by grasses, and its effects on the intra- and interspecific variation of leaf traits (SLA and LDMC) of grass species; (3) the effect of different grazing intensities (forage dry matter offer) in the alpha and beta diversity of plant communities. Changes in diversity, evenness, species richness, and biomass were verified, which were conditioned by the rotational grazing that proved beneficial to grassland. The rotational grazing provided also an increase in functional diversity, due to intra and interspecific variation of the grasses traits. Different grazing intensities promoted differences in alpha and beta diversity under different grazing intensities. Grazing demonstrated to have a decisive role in the whole dynamics of grassland communities, with considerable effects on all levels here assessed, from the species richness to the functional diversity, becoming apparent that the management decision is very important for both the productivity and the maintenance of grassland structure and diversity.

O manejo de campo nativo apresenta o desafio de desenvolver estratégias de intensificação sustentável para superar o dilema produçãoconservação. Embora a intensificação da produção através da agregação de insumos (fertilizantes e/ou sementes) seja possível, também pode causar mudanças negativas na sustentabilidade. Apesar das quantificações preliminares, existe uma lacuna no conhecimento acerca de como e quando são alteradas a diversidade, a estabilidade e a resiliência em processos de intensificação. Neste sentido, esta tese se concentra em identificar, através de dois experimentos: i) os principais fatores de controle da estabilidade temporal da produtividade de pastagens enriquecidas com nitrogênio (N), fósforo (P) e água; e ii) determinar os efeitos da fertilização fosfatada na diversidade de espécies e no estado da vegetação em campos nativos sobressemeados com leguminosas exóticas. O primeiro experimento apresenta o efeito de quatro anos de adição de N, P e água na produtividade e estabilidade temporal da produção. A adição de N e N + P diminuiu a estabilidade, processo associado à redução da constância da produção das espécies mais dominantes, enquanto que as adições de P e água incrementaram a estabilidade devido ao aumento da assincronia das espécies. Estas duas formas de controle da estabilidade são alternativas e a expressão está regulada pela limitação de N. Contrariamente ao proposto na teoria predominante “diversidade-estabilidade”, os resultados deste trabalho indicam que a estabilidade da produção depende principalmente do comportamento das espécies dominantes e não da diversidade de espécies. O segundo experimento expõe o efeito de mais de 10 anos de fertilização fosfatada na diversidade de espécies de um campo nativo sobressemeado com Trifolium repens e Lotus corniculatus. As respostas em riqueza e diversidade de espécies se relacionaram negativamente com a concentração de P do solo. Foram identificados limiares de concentração de P a partir dos quais se alcançaram as perdas máximas de diversidade e também foi desenvolvido um modelo de estados e transições da vegetação. Em síntese, esta tese contribui para ampliar o conhecimento do impacto de práticas de melhoramento em atributos de sustentabilidade (produtividade, diversidade, estabilidade, e resiliência) dos pastos nativos do bioma Pampa. A continuação destes esforços permitirá avanços na definição dos limites da intensificação baseados em critérios ecológicos, e assim contribuirá com o desafio de desenvolver estratégias de manejo sustentável. / Native grassland management faces the challenge to develop sustainable intensification strategies to overcome the dilemma between production and conservation. Although the intensification of the production through addition of inputs (fertilizers and/or seeds) to be possible, it can potentially induces negative changes in the sustainability of the ecosystems. There is a gap in the knowledge about how and when diversity, stability and resilience are altered in the intensification process. The main body of this thesis consist of two experiments focused on: i) identifying the main factors that control the temporal stability of productivity of a native grassland enriched with nitrogen (N), phosphorus (P) and water; and ii) determining the effects of phosphorus fertilization in the diversity and the state of the vegetation of a native grassland overseeded with exotic legumes. The first experiment analyzed the effect of four years of N, P and water addition on the productivity and the temporal stability of a native grassland production. The single addition of N or combined with P decreased the temporal stability by reducing the constancy of the production of the most dominant species, whereas the additions of P and water increased the stability by increasing species asynchrony. Both controls of temporal stability appear to be alternative mechanisms, and the specific expression of each one would be regulated by the limitation of N. In the second experiment, the long term effects of two P fertilization doses on species richness and diversity of a native grassland overseeded with Trifolium repens and Lotus corniculatus were examinated. Extractable soil P was negatively related with species richness and diversity. Additionally, P concentration thresholds related to the maximum loss of species richness and diversity were identified, and a model of the states and transitions of the vegetation was developed. In summary, this thesis contributes to the understanding of the impact of intensification practices on sustainability attributes (productivity, diversity, stability and resilience) of native grasslands. The continuation of this research will focus on a detailed definition of the thresholds of the intensification based on ecological criteria, and thus contribute to the challenge to develop sustainable systems.

As mudanças climáticas globais, sem dúvida, vêm impactanto muitas culturas de plantas agrícolas e pastagens, afetando potencialmente o crescimento e a produtividade das plantas. Os efeitos significativos podem ser causados pelo aumento das concentrações de CO2 e temperatura no quadro da mudança global. As plantas C3 em termos de fotossíntese e produção de biomassa geralmente respondem positivamente em elevado CO2 atmosférico do que as plantas C4. No entanto, os efeitos do elevado CO2 e do aquecimento podem ser mais complexos do que previsto anteriormente, devido às interações desses fatores. O objetivo deste estudo foi avaliar os efeitos do elevado CO2 (600 ppm) e aquecimento (2 °C de aumento sobre a temperatura ambiente) sobre a fisiologia e bioquímica da forrageira tropical C4 Panicum maximum Jacq. cultivada em um Trop-T-Face (Tropical-Temperature free-air controlled enhancement and free-air carbon dioxide enrichment) sem limitação de água e nutrientes. Foi estimado que o elevado CO2 quando combinado com o aquecimento (eC+eT) aumentou a massa seca do colmo (93%), a biomassa seca total (52%), a área foliar específica (AFE) (12%) e reduziu a razão folha/colmo (30%). Aos 30 dias de experimento, os açúcares solúveis totais das folhas aumentaram com o elevado CO2 (eC) e aquecimento (eT) e depois de 80 dias os açúcares redutores e amido reduziram em eC+eT. Em relação aos carboidratos estruturais, o teor de celulose nas folhas não sofreu variação significativa entre os tratamentos, no entanto, o teor de lignina aumentou em 100% sob eC. Além do incremento em biomassa as plantas foram capazes de evitar fotoinibição do aparato fotossintético e peroxidação lipídica. No tratamento eT os níveis de clorofila aumentaram em 12% e os níveis de -tocoferol reduziram em média de 55%. Os níveis de zeaxantina e luteína, o ciclo de xantofila (VAZ) e o estado de-epoxidação (DPS) do ciclo da xantofila foram alterados pelo elevado CO2 e aquecimento somente depois de 60 dias. Conclui-se que o elevado CO2 e o aquecimento apesar de promoverem uma redução no mecanismo de fotoproteção, a eficiência fotoquímica do fotossistema (PSII) não é danificada se as plantas são cultivadas em condições de disponibilidade hídrica e sem limitação nutricional. No entanto, apesar do aumento em biomassa com elevado CO2 combinado com aquecimento, o elevado CO2 parece resultar em diminuição da qualidade da forragem. / Global climate change undoubtedly impact many agricultural plant crops and grasslands, potentially affecting plant growth and yield. Significant effects may be caused by increased temperature and CO2 concentrations in the frame of global change. It is generally thought that C3 plants may respond positively to atmospheric CO2 enrichment than C4 plants in terms of photosynthesis and biomass accumulation. However, the effects of elevated CO2 and warming may be more complex than previously anticipated due to interactions between these factors. The aim of this study was to evaluate the effects of elevated CO2 (600 ppm) and warming (2 °C increase above ambient temperature) over physiology and biochemistry of the tropical C4 forage Panicum maximum Jacq. grown in a Trop-T-FACE (Tropical-Temperature free-air controlled enhancement and free-air carbon dioxide enrichment) facility under well-watered conditions without nutrient limitation. The combination of elevated CO2 and warming (eC+eT) increased stem dry weight (93%), total dry biomass (52%), specific leaf area (SLA) and decreased the ratio leaves/stem (30%). After 30 days of experiment, total soluble sugar of leaves increased under elevated CO2 (eC) and warming (eT) and carbohydrate sugars and starch decreased in elevated CO2 combined with warming, after 80 days. Regarding structural carbohydrates, cellulose content in leaves had no significant variation between treatments, however, lignin content increased 100% under eC. Furthermore, beyond increase in biomass, plants were able to avoid photoinhibition of the photosynthetic apparatus and lipid peroxidation in response to elevated CO2 and warming. In treatment eT chlorophyll levels increased by 12% and -tocopherol levels decreased by 55%. The de-epoxidation state (DPS) of the xanthophyll cycle, an indicator of excess energy dissipation as heat, xanthophyll cycle pool (VAZ) and both zeaxanthin and lutein levels were changed by elevated CO2 and warming only after 60 days. We conclude that the elevated CO2 and warming while promoting a reduction in photoprotection mechanism, the photochemical efficiency of photosystem (PSII) is not damaged if plants are grown under well-watered and adequate nutritional conditions. However, despite growth increase with elevated CO2 combined warming, elevated CO2 appears to result in decreased quality of forage.