Os campos do Rio Grande do Sul apresentam grande biodiversidade, beleza cênica e um importante recurso para a economia do estado, cuja principal atividade é a pecuária. O manejo inadequado do campo traz conseqüências graves para a diversidade e sustentabilidade desse ecossistema através da perda de espécies promovida pelo pastejo excessivo. Entretanto, algumas espécies desempenham as mesmas funções nos ecossistemas, logo a perda de algumas espécies pode não afetar o seu funcionamento, pela compensação de outras espécies do mesmo grupo funcional. Esta redundância funcional aumentaria a confiabilidade do funcionamento do ecossistema quando perturbado, funcionando como um tipo de "seguro" contra a perda de espécies. A hipótese da redundância funcional foi avaliada em ecossistemas campestres através de um experimento de remoção no qual os tratamentos avaliados consistiram em: 1) remoção de duas espécies graminóides, 2) remoção de duas espécies herbáceas não-graminóides, 3) remoção de uma espécie de graminóide e uma herbácea não-graminóide e 4) sem remoção. As evidências indicam que existe redundância funcional para as espécies de plantas em ecossistemas campestres, pois a remoção de uma espécie de cada grupo funcional causou menos modificações na composição de espécies das comunidades que a remoção de duas espécies do mesmo grupo funcional. Observou-se também que a porcentagem da cobertura das espécies remanescentes tende a compensar a remoção de espécies do mesmo grupo funcional. Entretanto, os tratamentos não tiveram efeito para a biomassa aérea. A avaliação de um banco de dados de um experimento de pastejo realizado anteriormente revelou que a redundância funcional está positivamente relacionada com a resistência da comunidade ao pastejo, enquanto a riqueza de espécies não apresentou relação. Áreas de campo com maiores níveis de redundância funcional são mais resistentes ao pastejo, sofrendo menos modificações na composição de espécies de suas comunidades. Além disso, um manejo adequado não reduz os níveis de redundância funcional dos campos. Áreas de campo dentro de unidades de conservação deveriam ser manejas adequadamente mantendo os níveis de redundância funcional e a estabilidade deste ecossistema. Entretanto, novos estudos são necessários para avaliar a importância da redundância funcional em experimentos de longa duração. / Grasslands from Rio Grande do Sul present great biodiversity, scenic beauty and an important resource to the province economy, which main activity is cattle grazing. Inadequate pasture management brings serious consequences to ecosystem diversity and sustainability by the loss of species promoted by excessive grazing. Although, some species may role the same functions in ecosystems, therefore loss of some species may not affect functioning, due to species compensation within functional groups. This functional redundancy would increase ecosystem reliability when perturbations occur, working as an "insurance" to species loss. Functional redundancy hypothesis was evidenced in grassland plant communities through a removal experiment in which evaluated treatments were: 1) removal of two graminoid species, 2) removal of two forb species, 3) removal of one species of each functional group and 4) no removal (control). Evidences point to functional redundancy for grassland plant species; removal of one species of each functional group caused less harm to community species composition than removal of two species of the same functional group. Furthermore, remaining species percent covers tend to compensate removed species of the same functional group. However, treatments did not effected aboveground biomass. Data set evaluation of a previously conducted cattle-grazing experiment reveled that functional redundancy is positively related to community resistance to cattle grazing, otherwise species richness presented no relation. Grassland areas with higher levels of functional redundancy are more resistant to cattle grazing, suffering less modification in community species composition. Also, adequate management do not reduce functional redundancy on pastures. Grassland areas inside conservation units should be managed adequately maintaining redundancy levels and ecosystem reliability. Although, new studies are required to better evaluate functional redundancy implications in long term experiments.

La biodiversité est de plus en plus modifiée par les activités humaines, ce qui a conduit à des recherches considérables sur l'effet de la biodiversité sur le fonctionnement des écosystèmes. Ces recherches ont cependant rarement inclu le stress anthropogénique qui cause la perte de biodiversité. Cette thèse analyse l'impact du stress (pesticides et métaux lourds) sur la relation entre la biodiversité et le fonctionnement des écosystèmes à la base de la chaîne alimentaire marine, avec des producteurs marins primaires (diatomées) et leurs principaux consommateurs (copépodes). Le premier résultat est que la pollution chimique altère la régularité plutôt que la richesse des espèces, avec un effet disproportionné sur le fonctionnement si un stress sélectif provoque la domination par des espèces tolérantes avec une faible contribution fonctionnelle. L'effet du stress sur le fonctionnement est prévisible sur la base d'une corrélation des traits biologiques prédisant l'abondance des espèces et des traits prédisant leur effet sur le fonctionnement. L'effet de la biodiversité sur le fonctionnement augmente sous l'effet du stress dû à une augmentation d'interactions facilitatrices entre les espèces (effet de complémentarité). L'effet du stress sur la qualité alimentaire des producteurs primaires réduit le transfert d'énergie au niveau des consommateurs, soulignant la pertinence du changement de biodiversité dans un contexte trophique. La thèse se termine par un cadre synthétique présentant les trois principaux effets (stress sélectif, stress physiologique, complémentarité) déterminant l'effet du stress sur la relation entre la biodiversité et le fonctionnement des écosystèmes. / Biodiversity is increasingly altered by human activities, which has led to considerable research on the consequences of biodiversity loss for ecosystem functioning. The anthropogenic stressors driving the ongoing biodiversity loss are however rarely included into biodiversity-ecosystem functioning experiments. The present thesis analyses the impact of anthropogenic stressors (pesticides, heavy metals) on the biodiversity-ecosystem functioning relation in marine primary producers (diatoms) and consumers (copepods). It was shown that chemical stress alters species evenness rather richness, with a disproportionate effect on functioning if stress causes dominance by tolerant species with a low functional contribution. Stress tolerance and the functional contribution were predictable based on the species’ biological traits, and the effects of stress on ecosystem functioning depended on the correlation of traits predicting species abundance and traits predicting the species’ effect on ecosystem functioning. The biodiversity effect on primary producer biomass production increased under stress due to facilitative interactions (complementarity effect). Stressor-induced biodiversity loss reduced the food quality of primary producers and impaired the energy transfer to the consumer level, highlighting the relevance of stressor-induced biodiversity changers in a trophic context. The thesis ends with a synthetic framework which puts forward the three main effects (selective stress, physiological stress, complementarity) through which anthropogenic stress affects biodiversity-ecosystem functioning relations at the basis of marine food webs.

Os campos do Rio Grande do Sul apresentam grande biodiversidade, beleza cênica e um importante recurso para a economia do estado, cuja principal atividade é a pecuária. O manejo inadequado do campo traz conseqüências graves para a diversidade e sustentabilidade desse ecossistema através da perda de espécies promovida pelo pastejo excessivo. Entretanto, algumas espécies desempenham as mesmas funções nos ecossistemas, logo a perda de algumas espécies pode não afetar o seu funcionamento, pela compensação de outras espécies do mesmo grupo funcional. Esta redundância funcional aumentaria a confiabilidade do funcionamento do ecossistema quando perturbado, funcionando como um tipo de "seguro" contra a perda de espécies. A hipótese da redundância funcional foi avaliada em ecossistemas campestres através de um experimento de remoção no qual os tratamentos avaliados consistiram em: 1) remoção de duas espécies graminóides, 2) remoção de duas espécies herbáceas não-graminóides, 3) remoção de uma espécie de graminóide e uma herbácea não-graminóide e 4) sem remoção. As evidências indicam que existe redundância funcional para as espécies de plantas em ecossistemas campestres, pois a remoção de uma espécie de cada grupo funcional causou menos modificações na composição de espécies das comunidades que a remoção de duas espécies do mesmo grupo funcional. Observou-se também que a porcentagem da cobertura das espécies remanescentes tende a compensar a remoção de espécies do mesmo grupo funcional. Entretanto, os tratamentos não tiveram efeito para a biomassa aérea. A avaliação de um banco de dados de um experimento de pastejo realizado anteriormente revelou que a redundância funcional está positivamente relacionada com a resistência da comunidade ao pastejo, enquanto a riqueza de espécies não apresentou relação. Áreas de campo com maiores níveis de redundância funcional são mais resistentes ao pastejo, sofrendo menos modificações na composição de espécies de suas comunidades. Além disso, um manejo adequado não reduz os níveis de redundância funcional dos campos. Áreas de campo dentro de unidades de conservação deveriam ser manejas adequadamente mantendo os níveis de redundância funcional e a estabilidade deste ecossistema. Entretanto, novos estudos são necessários para avaliar a importância da redundância funcional em experimentos de longa duração. / Grasslands from Rio Grande do Sul present great biodiversity, scenic beauty and an important resource to the province economy, which main activity is cattle grazing. Inadequate pasture management brings serious consequences to ecosystem diversity and sustainability by the loss of species promoted by excessive grazing. Although, some species may role the same functions in ecosystems, therefore loss of some species may not affect functioning, due to species compensation within functional groups. This functional redundancy would increase ecosystem reliability when perturbations occur, working as an "insurance" to species loss. Functional redundancy hypothesis was evidenced in grassland plant communities through a removal experiment in which evaluated treatments were: 1) removal of two graminoid species, 2) removal of two forb species, 3) removal of one species of each functional group and 4) no removal (control). Evidences point to functional redundancy for grassland plant species; removal of one species of each functional group caused less harm to community species composition than removal of two species of the same functional group. Furthermore, remaining species percent covers tend to compensate removed species of the same functional group. However, treatments did not effected aboveground biomass. Data set evaluation of a previously conducted cattle-grazing experiment reveled that functional redundancy is positively related to community resistance to cattle grazing, otherwise species richness presented no relation. Grassland areas with higher levels of functional redundancy are more resistant to cattle grazing, suffering less modification in community species composition. Also, adequate management do not reduce functional redundancy on pastures. Grassland areas inside conservation units should be managed adequately maintaining redundancy levels and ecosystem reliability. Although, new studies are required to better evaluate functional redundancy implications in long term experiments.

Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2016-05-31T00:33:23Z No. of bitstreams: 1 GuilhermeGerhardtMazzochini_TESE.pdf: 3651293 bytes, checksum: b4d523ac7c4eb9bd986c602bece5e003 (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2016-06-03T00:00:09Z (GMT) No. of bitstreams: 1 GuilhermeGerhardtMazzochini_TESE.pdf: 3651293 bytes, checksum: b4d523ac7c4eb9bd986c602bece5e003 (MD5) / Made available in DSpace on 2016-06-03T00:00:09Z (GMT). No. of bitstreams: 1 GuilhermeGerhardtMazzochini_TESE.pdf: 3651293 bytes, checksum: b4d523ac7c4eb9bd986c602bece5e003 (MD5) Previous issue date: 2014-12-05 / O estudo dos efeitos que a diversidade de esp?cies pode causar nos processos ecossist?micos tem crescido vertiginosamente nas ?ltimas duas d?cadas. Diversos trabalhos experimentais realizados no mundo todo t?m demonstrado que uma maior diversidade de plantas contribui para o aumento da produtividade de ecossistemas terrestres. Al?m disso, esse efeito pode influenciar processos em diversos n?veis tr?ficos, contribuindo assim para a estabilidade dos processos ecossist?micos a longo prazo. Paralelamente com os estudos do efeito da diversidade, muita aten??o tem sido dada para desvendar o papel das caracter?sticas funcionais das esp?cies no funcionamento dos ecossistemas. Isto porque as caracter?sticas funcionais das esp?cies t?m se mostrado importantes "pe?as" no entendimento dos efeitos que esp?cies individuais podem exercer nos ecossistemas e suas respostas ao ambiente. Nesta tese de doutorado eu explorei algumas lacunas de conhecimento dentro dessa ?rea em crescente desenvolvimento conhecida na literatura ecol?gica como "biodiversidade e funcionamento dos ecossistemas". No primeiro cap?tulo, eu busquei evid?ncias para mecanismos que podem explicar a rela??o positiva entre diversidade e funcionamento com foco em cinco mecanismos relacionados ?s intera??es entre plantas, tendo como par?metro de funcionamento a produtividade prim?ria. No segundo cap?tulo, eu utilizei t?cnicas para a estimativa de padr?es de diversidade em escalas biogeogr?ficas e bases de dados de sat?lites com longa dura??o para desvendar se a biodiversidade em escalas macroecol?gicas promove a estabilidade da produtividade dos ambientes terrestres no semi?rido brasileiro. Por fim, o objetivo do terceiro cap?tulo foi entender como a perda da cobertura vegetal origin?ria do uso da terra por comunidades tradicionais no semi?rido brasileiro influenciaria os processos de intera??es entre plantas e o papel das caracter?sticas funcionais das esp?cies nessas intera??es. Acredito que a contribui??o individual de cada cap?tulo preenche lacunas de conhecimento importantes dessa ?rea da Ecologia que ainda se encontra em expans?o.

Os campos do Rio Grande do Sul apresentam grande biodiversidade, beleza cênica e um importante recurso para a economia do estado, cuja principal atividade é a pecuária. O manejo inadequado do campo traz conseqüências graves para a diversidade e sustentabilidade desse ecossistema através da perda de espécies promovida pelo pastejo excessivo. Entretanto, algumas espécies desempenham as mesmas funções nos ecossistemas, logo a perda de algumas espécies pode não afetar o seu funcionamento, pela compensação de outras espécies do mesmo grupo funcional. Esta redundância funcional aumentaria a confiabilidade do funcionamento do ecossistema quando perturbado, funcionando como um tipo de "seguro" contra a perda de espécies. A hipótese da redundância funcional foi avaliada em ecossistemas campestres através de um experimento de remoção no qual os tratamentos avaliados consistiram em: 1) remoção de duas espécies graminóides, 2) remoção de duas espécies herbáceas não-graminóides, 3) remoção de uma espécie de graminóide e uma herbácea não-graminóide e 4) sem remoção. As evidências indicam que existe redundância funcional para as espécies de plantas em ecossistemas campestres, pois a remoção de uma espécie de cada grupo funcional causou menos modificações na composição de espécies das comunidades que a remoção de duas espécies do mesmo grupo funcional. Observou-se também que a porcentagem da cobertura das espécies remanescentes tende a compensar a remoção de espécies do mesmo grupo funcional. Entretanto, os tratamentos não tiveram efeito para a biomassa aérea. A avaliação de um banco de dados de um experimento de pastejo realizado anteriormente revelou que a redundância funcional está positivamente relacionada com a resistência da comunidade ao pastejo, enquanto a riqueza de espécies não apresentou relação. Áreas de campo com maiores níveis de redundância funcional são mais resistentes ao pastejo, sofrendo menos modificações na composição de espécies de suas comunidades. Além disso, um manejo adequado não reduz os níveis de redundância funcional dos campos. Áreas de campo dentro de unidades de conservação deveriam ser manejas adequadamente mantendo os níveis de redundância funcional e a estabilidade deste ecossistema. Entretanto, novos estudos são necessários para avaliar a importância da redundância funcional em experimentos de longa duração. / Grasslands from Rio Grande do Sul present great biodiversity, scenic beauty and an important resource to the province economy, which main activity is cattle grazing. Inadequate pasture management brings serious consequences to ecosystem diversity and sustainability by the loss of species promoted by excessive grazing. Although, some species may role the same functions in ecosystems, therefore loss of some species may not affect functioning, due to species compensation within functional groups. This functional redundancy would increase ecosystem reliability when perturbations occur, working as an "insurance" to species loss. Functional redundancy hypothesis was evidenced in grassland plant communities through a removal experiment in which evaluated treatments were: 1) removal of two graminoid species, 2) removal of two forb species, 3) removal of one species of each functional group and 4) no removal (control). Evidences point to functional redundancy for grassland plant species; removal of one species of each functional group caused less harm to community species composition than removal of two species of the same functional group. Furthermore, remaining species percent covers tend to compensate removed species of the same functional group. However, treatments did not effected aboveground biomass. Data set evaluation of a previously conducted cattle-grazing experiment reveled that functional redundancy is positively related to community resistance to cattle grazing, otherwise species richness presented no relation. Grassland areas with higher levels of functional redundancy are more resistant to cattle grazing, suffering less modification in community species composition. Also, adequate management do not reduce functional redundancy on pastures. Grassland areas inside conservation units should be managed adequately maintaining redundancy levels and ecosystem reliability. Although, new studies are required to better evaluate functional redundancy implications in long term experiments.

Submitted by Marlene Santos (marlene.bc.ufg@gmail.com) on 2016-07-27T21:10:07Z No. of bitstreams: 3 Tese - Diogo Borges Provete - 2016.pdf: 3048056 bytes, checksum: 5e3c0b282d1a0790baa37204e8059ed1 (MD5) Ata de defesa - Diogo Borges Provete.PDF: 971791 bytes, checksum: d78fa4a86345ee06bcf4e49553a98347 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2016-07-28T11:01:30Z (GMT) No. of bitstreams: 3 Tese - Diogo Borges Provete - 2016.pdf: 3048056 bytes, checksum: 5e3c0b282d1a0790baa37204e8059ed1 (MD5) Ata de defesa - Diogo Borges Provete.PDF: 971791 bytes, checksum: d78fa4a86345ee06bcf4e49553a98347 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2016-07-28T11:01:30Z (GMT). No. of bitstreams: 3 Tese - Diogo Borges Provete - 2016.pdf: 3048056 bytes, checksum: 5e3c0b282d1a0790baa37204e8059ed1 (MD5) Ata de defesa - Diogo Borges Provete.PDF: 971791 bytes, checksum: d78fa4a86345ee06bcf4e49553a98347 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2015-03-30 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Phylogenetic information has increasingly been included in studies of local communities and also at broad spatial scales. Despite recent criticisms in the last four years, phylogenetic relationships may still provide insights into the organization and assembly of ecological communities. The objectives of this study were 1) to review the history of the use of phylogenetic information, as well as criticisms and perspectives of its use in community ecology; 2) understand how the size and shape of phylogenetic trees and the phylogenetic structure of metacomunidaes affect the amount of variation accounted for by a eigenvector-based method used to describe the phylogenetic composition of metacomunidaes (PCPS); 3) to test the effect of diversity of evolutionary history (MNTD and MPD) and species richness as predictors of three variables of freshwater ecosystem functioning (productivity, respiration, and decomposition); and finally 4) to test how environmental gradients, especially pond canopy cover, influence the phylogenetic structure of an anuran metacommunity from southeastern Brazil. I found that the structure of metacommunities had greater impact on eigenvalues of PCPS than tree shape metrics, such as symmetry and stemminess. In addition, decomposition and respiration were best predicted by MNTD as a linear function, while productivity was affected by the quadratic term of MNTD. Finally, pond canopy cover and floating vegetation strongly affected the phylogenetic structure of the anruan metacommunity, influencing lineage sorting. These findings 1) can help users interpret the results of PCPS; 2) provide better understand of the effect of species loss in multitrophic, freshwater ecosystems; and 3) improve our knowledge about the effect of canopy cover on the lineage composition in anuran metacomunities. / Informações filogenéticas tem sido cada vez mais incluídas em estudos de comunidades locais e em larga escala espacial. Apesar das críticas dos últimos quatro anos, relações de parentesco ainda podem fornecer idéias sobre a organização e montagem de comunidades ecológicas. Os objetivos deste estudo foram 1) revisar o histórico do uso de informações filogenéticas, bem como críticas e perspectivas quanto ao seu uso em ecologia de comunidades; 2) entender como o formato e tamanho de árvores filogenéticas e a estrutura filogenética da metacomunidade afetam a quantidade de variação capturada por um método baseado em autovetores usado para descrever a composição filogenética de metacomunidaes (PCPS); 3) testar o efeito da diversidade de história evolutiva (MNTD e MPD) e da riqueza de espécies como preditores de três métricas de funcionamento de ecossistemas aquáticos (produtividade, respiração, e decomposição); e por fim 4) testar como gradientes ambientais, especialmente cobertura de dossel, influenciam a estrutura filogenética de uma metacomunidade de anuros no sudeste do Brasil. Encontrei que a estrutura da metacomunidade teve maior impacto nos autovalores do PCPS do que métricas de formato de árvore, tais como simetria e enraizamento (stemminess). Além disso a decomposição e respiração foram melhor preditas por MNTD, enquanto produtividade foi melhor predita pelo termo quadrático desta medida. Por fim, a cobertura de dossel e vegetação flutuantes afetaram fortemente a estrutura filogenética da metacomunidade de anuros, influenciando a distribuição de linhagens. Estes resultados 1) podem auxiliar na interpretação dos resultados do PCPS por usuários; 2) permitem compreender melhor o efeito da perda de espécies em ecossistemas aquáticos multitróficos; e 3) ampliam o conhecimento sobre o efeito da cobertura de dossel sobre a composição de linhagens em metacomunidades de anuros.

Cette thèse étudie par trois modèles et une expérience les effets des interactions entre le réseau vert (basé sur la production primaire) et brun (basé sur la décomposition) sur le fonctionnement des écosystèmes. Trois interactions entre les deux réseaux sont étudiées: 1) les nutriments recyclés par tous les organismes couplent les réseaux verts et bruns par une relation de mutualisme/compétition entre les producteurs primaires et les décomposeurs; 2) les consommateurs généralistes se nourrissent de proies des deux réseaux trophiques et 3) les connexions spatiales par les consommateurs généralistes mobiles en haut des réseaux trophiques et les flux de nutriments et de détritus en bas. Les résultats montrent que ces interactions peuvent affecter le fonctionnement des écosystèmes y compris les cascades trophiques, la stabilité et l'importance relative de la chaîne verte et brune. Ces résultats conduisent à (1) de nouvelles perspectives sur la modélisation du fonctionnement des réseaux trophiques, (2) des prédictions originales, (3) des nouvelles interprétations de résultats empiriques, (4) des idées sur des différences fondamentales de fonctionnement entre types d’écosystèmes, (5) des prédictions sur la réponses des écosystèmes aux changements globaux. / This thesis studies through three models and an experiment the important effects of the interactions between the green (based on primary production) and brown (based on decomposition) food webs on ecosystem functioning. Three interactions between the two food webs are studied: 1) nutrients recycled from all organisms in the food web couple the green and brown food webs through mutualistic/competitive interaction between primary producers and decomposers; 2) generalist consumers feed on prey from both food webs; and 3) the spatial connections between the two food webs through mobile generalist consumers at the top and through nutrient and detritus fluxes at the bottom of the food webs. The results show that these interactions may affect the ecosystems functioning including trophic cascades, stability and the relative importance of green and brown food chains. These results lead to new (1) insights on the way to model food webs, (2) predictions on food web functioning, (3) interpretation of empirical results, (4) ideas to compare the functioning of different ecosystem types and (5) predictions on food web responses to global changes.

Root processes – decomposition, production, and mortality – are intrinsic parts of ecosystem carbon cycling and yet have been understudied in the context of biodiversity-ecosystem functioning relationships. In a long-term grassland diversity experiment (the Jena Experiment), I investigated (1) whether and how plant diversity affected root decomposition and (2) how plant diversity enhanced standing root biomass and influenced root turnover. Plant diversity may affect root decomposition via three non-mutually exclusive pathways: (1) root litter quality, (2) soil biota, and/or (3) soil abiotic conditions. In Chapter 3, via three decomposition experiments, I demonstrate that plant species richness negatively affects root decomposition via the root litter quality pathway and the soil environment pathway (including soil biotic and abiotic conditions). The presence of grasses negatively affects root decomposition while the presence of legumes positively affects root decomposition. In Chapter 4, I further explored the pathways driving the relationship between plant diversity and root decomposition using structural equation modeling. The final structural equation model suggests that root chemistry is a universal pathway for explaining the effects of plant diversity on root decomposition and that Oribatida are also involved in root decomposition. Most importantly, I directly show that different measures of plant diversity influence root decomposition via different pathways. In Chapter 5, I shift my focus to root production and mortality which collectively determine standing root biomass. I monitored in situ root dynamics biweekly for one growing season using minirhizotrons. I show that plant species richness consistently enhances standing root length throughout the observational period. However, plant species richness did not affect rates of root length production, mortality, or turnover. Only root lifespan increased with increasing plant species richness. The lack of significant diversity effect on root length-based measures may be due to the age of the studied communities. In mature grasslands, plant species richness may only have a minor effect on root turnover and one growing season may be too short to detect such a small effect. The results of this dissertation collectively provide new insights into the relationship between plant species richness and soil carbon stock in mature grasslands. Based on the new paradigm of soil carbon formation, plant species richness may enhance soil carbon stock through a greater input of partially decomposed root residuals (reduced root decomposition) and a higher input of root exudates (increased standing root biomass).:ACKNOWLEDGEMENTS .......................................................................................III 1 GENERAL INTRODUCTION ............................................................................... 1 1.1 Biodiversity-Ecosystem Functioning Research in the Context of Global Biodiversity Loss ……........................................................................................... 1 1.2 Biodiversity-Ecosystem Functioning Relationships .............................................................................................................................. 2 1.3 Effects of Plant Diversity on Decomposition ................................................... 6 1.4 Effects of Plant Diversity on Root Turnover .................................................. 10 1.5 Scope and Contents of this Dissertation ...................................................... 14 2 GENERAL METHODS ......................................................................................17 2.1 The Jena Experiment ................................................................................... 17 2.2 Root Decomposition Experiments ................................................................ 18 2.3 Applying Structural Equation Modeling to Understand the Diversity-Decomposition Relationships ............................................................................. 22 2.4 In-situ Monitoring of Root Turnover Using Minirhizotrons ............................. 25 3 PLANT SPECIES RICHNESS NEGATIVELY AFFECTS ROOT DECOMPOSITION IN GRASSLANDS ................................................................................................27 4 ROOT CHEMISTRY AND SOIL FAUNA, BUT NOT SOIL ABIOTIC CONDITIONS EXPLAIN THE EFFECTS OF PLANT DIVERSITY ON ROOT DECOMPOSITION..44 5 THE EFFECTS OF PLANT SPECIES RICHNESS ON ROOT TURNOVER IN EXPERIMENTAL GRASSLANDS...........................................................................74 6 GENERAL DISCUSSION..................................................................................105 6.1 Summary of Contents in this Dissertation.................................................... 105 6.2 Advancing Our Understanding of Root Decomposition in BEF Research.... 106 6.3 Time Matters for the Diversity–Root Turnover Relationship......................... 112 6.4 Outlook........................................................................................................ 117 REFERENCES...................................................................................................120 SUMMARY .........................................................................................................137 ZUSAMMENFASSUNG .......................................................................................142 LIST OF PUBLICATIONS....................................................................................148 AUTHOR CONTRIBUTION STATEMENTS..........................................................150 DECLARATION OF INDEPENDENT WORK........................................................159 CURRICULUM VITAE .........................................................................................160

Rising temperature caused by global warming alters physiology, phenology and/or distribution in a wide array of plant and animal species, which has dramatic knock-on effects at different levels of organisation. This study investigates the individual and interactive effects of temperature (18ᵒC, 22ᵒC and 30ᵒC) and additions of the macroalga Gracilaria gracilis (high and low) on the performance of the seagrass Zostera capensis, which occurs in Langebaan Lagoon, South Africa over a seven-week period. Results from the laboratory experiment revealed that G. gracilis did not significantly affect the performance of Z. capensis although temperature did result in greater leaf width, fouling and senescent biomass, as well as marginally greater leaf area and lower below-ground biomass at 30ᵒC. Increasing temperature also increased G. gracilis biomass, percent cover and fouling by microalgae. In addition, there was no interaction between temperature and the additions of Gracilaria. The overall findings of this study indicate that Z. capensis abundance is likely to decrease while G. gracilis will conversely increase in abundance in response to warming. Changes in abundance of those two ecosystem engineers highlight the possibility of a phase shift from a seagrass- to macroalgal-dominated state in Langebaan Lagoon.

Invasions from coho salmon were first reported in the Cape Horn Biosphere Reserve (CHBR) in 2019 which is the most southern distribution registered to date. The CHBR is known for its high number of endemic species and unique biodiversity, such as the native fishes Galaxias maculatus and Aplochiton taeniatus. There are now three invasive salmonid species in the rivers of CHBR and are a potential threat to the native fish taxa. Stable isotope and gut content analysis were used to understand resource utilization by both native galaxiid and invasive salmonid taxa, as well as aquatic macroinvertebrates and riparian spiders. The natural laboratory study approach applied to this research, allowed for comparisons of differences within streams that contain conditions in which fish do not occur naturally, to sites in which high densities of invasive salmonid exist. Analysis of the trophic niche and diet in this study showed the importance of marine resource use by the native galaxiid and coho salmon juveniles supported with elevated δ15N and δ34S ratios. Diet analysis also confirmed there was the highest similarity between the coho salmon juveniles and the native fish. Altered behavior and habitat use was shown through the isotope and diet analysis for the galaxiid in snow melt streams which could be due to the high density of invasive salmonids in these streams. The invasive salmonids were found to impact aquatic macroinvertebrate populations, specifically larger bodied Trichoptera abundance. Aquatic insect emergence was negatively correlated to salmonid densities. The aquatic insect emergence revealed alterations with significantly higher biomass of aquatic insect emergence in upstream sites without invasive fish. Lastly, aquatic insect predator (Hydrobiosidae: Rheochorema sp.) exhibited a suppressed trophic position in rivers with invasive salmonids. An unexpected finding from the study was the refugium stream habitat conditions that has been shown to be important for conserving the native fish in the CHBR. We conclude that the impacts from invasive species are significant enough to disrupt lower trophic levels especially aquatic insects. The altered aquatic emergence and insect behavior have the potential to disrupt food webs and ecosystem functioning in the southernmost protected ecoregion in the world.