No description available.

A wildfire of variable intensity occurred in mid-July of 1988 in the Shenandoah National Park and adjacent private lands. This study was established to measure post-fire forest soil and vegetation characteristics in both the mixed pine and mixed oak forest types occurring on Dovel Mountain in the Shenandoah National Park, Virginia. Vegetation and soil results represent one (1989) and two (1990) growing seasons following fire occurrence in the mixed pine forest type, whereas only second year results are reported for the mixed oak forest type. Forest floor and mineral soil parameters differed among burn levels and followed similar trends for both forest types. Forest floor depth and weight were significantly less on low and high burn areas as compared to unburned areas. Low intensity fires removed the Oi-Oe layer while leaving the Oa layer relatively intact. Conversely, high intensity fifes resulted in the consumption of the entire forest floor. Total carbon, nitrogen, and nutrient differences between low and unburned areas for the entire forest floor (Oi-Oe + Oa) were negligible. However, loss of these constituents were much greater following high intensity fifes. Differences in the surface 10 cm of mineral soil were also detected following following high intensity fifes. Differences in the sulface 10 cm of mineral soil were also detected following low and high intensity fires. Mineral soil acidity was lower while exchangeable cation concentrations were higher in burned areas compared to unburned areas. Total carbon and nitrogen levels were higher in low burn areas, whereas total carbon and nitrogen levels were lower on high bum areas. Inorganic nitrogen levels were higher in both low and high burn areas compared to unburned areas. / Master of Science

Text in Afrikaans, abstract in Afrikaans and English / Lineere plantegroeipatrone bestaande uit afwisselende grasryke mikrotrog- en boomryke- kruinbane wat parallel met die helling georienteer is, kom voor op die voetbang van die Baviaansberg, in die Warmbadgebied, Noordelike Provinsie. Die hoofdoelwitte van die studie was om uit 'n landskapekologiese benadering die verskynsel te probeer verklaar, om die vertikale en horisontale variasie van eienskappe te kwantifiseer en om die impak van die mens daarop te benadruk. Die differensiasie in die Tukulu- en Etoshagrondvorms in onderskeidelik mikrotroe en -kruine is die gevolg van grondprosesse soos illuviasie van klei, loging en ferrolise.Granulometriese analise dui daarop dat die moedermateriaal van gronde van een bron afkomstig is. Kolluviasie en afsetting van moedermateriaal op die voethang was waarskynlik die beginpunt van landskapontwikkeling waar verdeelde oppervlakvloei van water oenskynlik die mikrostrukture geinisieer het. Na aanleiding hiervan het die ontwikkeling van grondverskille en waterregimeverskille tot plantegroeiverskille gelei. Faktore wat die patroon geaksentueer het is termietaktiwiteite, brand en beweiding. / A linear vegetation pattern consists of alternating grassland micro troughs and woody micro crests lanes which were orientated parallel to the slope occurred on the foot slope of the Baviaansberg, in the Warmbaths area, Northern Province. From a landscape ecological approach the principal aims of the study was to explain the phenomenon, to quantify the vertical and horizontal variation of features and to accent the impact of man on it. The differentiation in Tukulu and Etosha forms in micro troughs and micro crests respectively were the effect of several soil processes. Granulometric analyses showed that the parent material of soil was derived from one source. Colluviation and deposit of parent material was probably the starting-point of landscape development where the divided surface flow of water apparently initiated the micro structures. As a result of this the development of soil and water regime differences lead to vegetation differences. / Geography / M. Sc. (Geografie)

Ainda que o conhecimento sobre a flora e a ecologia do cerrado tenha sido consideravelmente ampliado nas últimas décadas, persistem dificuldades relacionadas com a caracterização estrutural das fitofisionomias e pouco se sabe sobre as transformações fisionômicas que ocorrem nesta vegetação ao longo do tempo em áreas protegidas. Adicionalmente, mediante as mudanças climáticas, surgiu a demanda de quantificação de fitomassa e estoque de carbono em formações vegetais, entre os quais o cerrado. Os objetivos deste estudo foram: a) verificar se a classificação fisionômica de três fisionomias de cerrado reflete diferenças florísticas e estruturais; b) caracterizar a dinâmica espaço-temporal das fisionomias de cerrado; c) quantificar a fitomassa dessa vegetação e sua contribuição para estoque de carbono; d) avaliar a aplicabilidade de dados de sensoriamento remoto para estimar a fitomassa do cerrado. A pesquisa foi desenvolvida na Estação Ecológica de Assis (EEcA) localizada no estado de São Paulo. O cerrado típico, cerrado denso e cerradão foram caracterizados florística e estruturalmente e comparados para verificar se podem ser considerados distintos. Foram alocadas 30 parcelas de 20 x 50 m, sendo 10 parcelas para cada um dos tipos fisionômicos. Os indivíduos de espécies lenhosas com DAP \'> OU =\' 5 cm foram identificados e medidos. As fisionomias mostraram-se estruturalmente distintas, em classes de densidade, área basal e altura média das árvores e o melhor descritor para classificá-las, por ser pouco variável com o critério de inclusão, é a área basal (\'M POT.2\'/ha). Floristicamente, há diferenças apenas entre o cerradão e as fisionomias savânicas. Analisou-se a dinâmica espaço-temporal das fisionomias de cerrado, ao longo de 44 anos, com base em aerofotos (1962, 1984 e 1994) e imagens QuickBird (2006). Após a criação da unidade de conservação e devido à suspensão das atividades antrópicas (fogo e agropecuária), tem ocorrido um adensamento da vegetação, em que áreas de campo foram ocupadas por fisionomias de maior fitomassa, o cerradão correspondendo, em 2006, 91,43% da EEcA. Analisou-se, em particular, a oscilação na área ocupada por uma espécie invasora de samambaia (Pteridium arachnoideum). As imagens de 1994 e 2006 mostram que espécies arbóreas estão aumentando em densidade e cobertura em meio às manchas de samambaias e, em campo, constatou-se que a fitomassa das samambaias está diminuindo consideravelmente sob as copas das árvores. A fitomassa do cerrado stricto sensu e cerradão da EEcA foi estimada por meio de equações alométricas. Utilizaram-se parcelas de 20 x 40 m, sendo 20 parcelas para cada fitofisionomia. Utilizou-se regressão robusta com reamostragem Bootstrap para explorar as relações entre a fitomassa aérea do cerrado e as imagens do QuickBird e TM/Landsat, índices espectrais de vegetação (IV), componentes principais (CP), modelo linear de mistura espectral (MLME). Na EEcA, os valores médios de fitomassa aérea (23,22 Mg/ha) e total (28,88 Mg/ha) do cerrado stricto sensu foram próximos aos descritos na literatura para o cerrado do Brasil Central e os valores médios de fitomassa aérea (98,18 Mg/ha) e total (118,36 Mg/ha) do cerradão aproximaramse aos descritos para florestas estacionais. As bandas espectrais dos sensores QuickBird e TM apresentaram correlações fracas a moderadas com a fitomassa aérea de cerrado. As transformações espectrais (IV e CP) melhoram, em geral, a predição da fitomassa aérea de cerrado, contudo as correlações se mantiveram entre fracas e moderadas. / Even though knowledge on the ecology and flora of cerrado vegetation has considerably improved in recent years, gaps are still remaining on structural differentiation of the cerrado physiognomies and few is known about cerrado vegetation dynamics after protection from human pressure. In addition, before the climate changes, phytomass and carbon storage quantification has been a new challenge for different vegetation types, including the cerrado physiognomies. The present study was carried out with the aim of a) to characterize three cerrado physiognomies to verify if they are structurally and floristically distinct; b) to analyze the vegetation dynamics in time and space, to verify if the vegetation is undergoing a sucessional process, whose structural climax will be a forest physiognomy; c) to quantify phytomass in different physiognomies and their contribution to carbon stock; and d) to assess the application of remote sensing techniques to estimate the cerrado phytomass in large scale. This study was carried out at Assis Ecological Station (EEcA), located in the southwestern São Paulo state, Brazil. This protected area preserves one of the few brazilian cerrado (savanna) biome remnants in the State. Three distinct physiognomic types of cerrado (typical cerrado, dense cerrado and woodland cerrado) were floristically and structurally characterized and submitted to comparative analyses to verify if they can, or not, be considered as separate vegetation types within the cerrado gradient. Thirty permanent plots (20 x 50 m each) were set, ten in every cerrado type, and all woody individuals with DBH \'> OR =\' 5 cm were measured and identified. The three types of cerrado under comparison are structurally distinct in terms of density, medium height and basal area (\'M POT.2/ha), the last being considered as the best and more precise descriptor to classify the physiognomies of the cerrado vegetation. The woodland cerrado is also distinct by its flora, but the two open physiognomies (dense cerrado and cerrado stricto sensu) are floristically very similar. The dynamics of the vegetation types along 44 years in the studied area was analyzed by using aerial photographs (1962, 1984 and 1994) and QuickBird images (2006). After protection from human pressures (fire and agriculture), the woody vegetation density and phytomass has continuously increased, with open physiognomies tending to disappear and woodland cerrado replacing them. Surprisingly, the area covered by the invasive fern Pteridium arachnoideum has also decreased, replaced by sparse or clustered trees. The cerrado phytomass was estimated by allometric equations. Forty plots (20 x 40 m each) were used, twenty in every cerrado type - cerrado stricto sensu and woodland cerrado. Robust Regression and Bootstrap methods were used to explore relationships between aboveground cerrado phytomass and remote sensing data of QuickBird and Landsat Thematic Mapper (TM) sensors (spectral bands, vegetation index - VI, principal components - PC, linear spectral mixture model LSMM). Valors of medium phytomass obtained for the cerrado stricto sensu at Assis Ecological Station were close to those described in the literature for Central Brazilian cerrado. The valors of medium phytomass of the woodland cerrado were close to those described for seasonal forests. Spectral bands of QuickBird and TM sensors presented weak to moderate correlations with the aboveground cerrado phytomass. In general, spectral transformations (VI and PC) improved the prediction of the cerrado phytomass, however the correlation remained from weak to moderate.

Extensas áreas de campo do sudoeste do Rio Grande do Sul, no Bioma Pampa, ocorrem sobre solos arenosos suscetíveis ao processo de arenização. Nesses campos predominam condições de déficit hídrico, irradiação intensa, temperaturas extremas, baixa fertilidade do solo e uso pecuário. O estudo teve como objetivos, conhecer a diversidade florística dos campos com arenização; analisar os principais tipos de estratégias adaptativas apresentadas pelas espécies e as proporções em que essas adaptações ocorrem nas proximidades do areal, em um campo pastejado e em um campo excluído de pastejo; avaliar as mudanças nas proporções de área ocupada por solo exposto e por vegetação, considerando os fatores tempo (um ano) e cobertura vegetal (alta, intermediária e baixa). Áreas de campos nativos com arenização dos municípios de Alegrete, Manoel Viana e São Francisco de Assis foram percorridas para o levantamento florístico. O levantamento quantitativo foi realizado em três áreas de campos sob diferentes manejos (pastejo intenso, pastejo moderado e exclusão de pastejo) em novembro de 2007 e dezembro de 2008. Foram 36 unidades amostrais permanentes no campo excluído e 24 em cada campo pastejado. Para cada espécie amostrada foi registrada a cobertura absoluta, as estratégias adaptativas apresentadas e o respectivo hábito. Foram registradas 343 espécies pertencentes a 52 famílias, com duas novas citações para o Rio Grande do Sul: Eragrostis articulata e Eragrostis leucosticta e uma para o Brasil: Croton lorentzii. 89,5% do total das espécies apresentaram uma ou mais adaptações como resposta aos fatores de estresse. A presença de caracteres adaptativos no campo pastejado e no campo sem pastejo parece caracterizar as espécies dos campos de solos arenosos, cujas condições climáticas e edáficas são comuns. Os três campos diferem em composição e abundância das espécies. O campo com pastejo intermediário apresentou características de ambos os outros, indicando que há influência do pastejo na composição de espécies e na fisionomia dos campos. A dinâmica da vegetação nos campos pastejados foi influenciada tanto pelo tempo quanto pela cobertura vegetal. No campo excluído ocorreu variação na composição e abundância de espécies apenas conforme o gradiente de cobertura. O processo de arenização avança numa escala maior nas áreas pastejadas que no campo excluído, pois neste as áreas de maior cobertura vegetal tiveram menor influência do processo de arenização. Fatores como chuva e vento também contribuem para o aumento das áreas arenizadas ao longo do tempo, especialmente nos locais de menor cobertura vegetal. Para retardar o processo de arenização desses campos são sugeridas medidas como a retirada do gado das áreas atingidas e a utilização de materiais que impessam o deslocamento do solo, associadas à utilização de espécies nativas com potencial de resistência ao processo. / Extensive grassland areas in southwestern Rio grande do Sul, belonging to the Pampa Biome, are found in sandy soils that are susceptible to sandy patch process. In these grasslands, the prevailing conditions are water shortage, excessive irradiation, extreme temperatures, low fertility and cattle grazing. The objectives of this study were: to know the floristic diversity at grassland undergoing sandy patch process; to evaluate changes in exposed and plant-covered soil, considering time (one year) and vegetal cover (high, intermediary and low) as determining factors. Native grassland areas, located at Alegrete, Manoel Viana and São Francisco de Assis municipalities, were surveyed for the construction of a floristic list. The quantitative survey was carried out at three grassland areas under different managements (heavy-grazing, moderately-grazing and grazing-excluded) in November 2007 and December 2008, using 36 permanent sampling units at the excluded grassland area and 24 at each grazed grassland area. For each sampled species we registered absolute cover, adaptative strategies and habit. We registered 343 species, pertaining to 52 families, with two new citations for Rio Grande do Sul (Eragrostis articulata and Eragrostis leucosticte) and one for Brazil (Croton lorentzii). Among these species, 89.5% showed one or more adaptations as response to the stress-driving factors. The presence of adaptative characters at grazed grasslands and grazing-excluded grasslands seems to characterize sandy-soil grasslands species, since they share common climatic and edaphic conditions. The three grassland areas differ in species composition and abundance. The moderately-grazed area showed characteristics common to both other areas, indicating the influence of grazing on grassland formation physiognomy and species composition. Vegetation dynamics in grazed grasslands was influenced by time and vegetal cover. At the excluded grassland, variation in species composition and abundance occurred only according to the cover gradient. The sandy patch process advances faster in grazed areas than in grazing-excluded areas, due to higher soil cover values in the latter. Elements such as rain and wind also contribute to the enlargement of sandy patches over time, especially at sites presenting low soil cover values. In order to hamper the sandy patch process at these grasslands, we suggest measures such as cattle removal and use of soil-fixing equipments, associated to the use of native plant species that potentially resist the process.

Dans le contexte d'augmentation de la pression démographique et des risques liés aux changements climatiques, la question de l'érosion est de plus en plus cruciale. La gestion actuelle des milieux des montagnes méditerranéennes, particulièrement affectés par ces phénomènes, repose sur une bonne connaissance de leurs sensibilités à l'érosion et de leurs trajectoires à long terme. Une approche rétrospective des trajectoires des environnements a été menée par l'étude de trois archives sédimentaires lacustres prélevées dans les Alpes du Sud (lacs Petit, Vens et Allos). L'étude multidisciplinaire (sédimentologie, géochimie, palynologie) a permis de caractériser les dynamiques érosives et végétales depuis 14000 ans. À Allos et Vens, le début de l'Holocène est marqué par la maturation des écosystèmes, l'altération chimique des sols et la progression altitudinale des forêts. Ces dynamiques environnementales conduisent à la stabilisation progressive des écosystèmes, suivie d'un optimum bio-pédologique, d'une ouverture des paysages et d'une déstabilisation des sols respectivement datées de 12000-8000 cal. BP, 8000-6000 cal. BP, 6000-1900 cal. BP et de 1900 cal.BP à aujourd'hui. Cette dernière période de déstabilisation des sols intervient plus tôt au Lac Petit, à 4200 cal. BP.Les dénominateurs communs au déclenchement de l'érosion sont à chaque fois une période de précipitations fréquentes sur un milieu fragilisé : l'impact des sociétés a provoqué un abaissement du seuil de sensibilité des milieux aux perturbations, particulièrement à l'agressivité des précipitations. / Within the context of increasing demographic pressures and hazards related to climate change, the problems posed by landscape erosion have become particularly crucial. The current management of Mediterranean mountain environments, which are highly exposed to erosion hazards, needs to be supported by a thorough understanding of their susceptibility to these hazards and their long-term trajectories.A retrospective analysis of environmental trajectories has been conducted from the study of three lacustrine sedimentary archives in the Southern Alps (Lakes Petit, Vens and Allos). This multi-disciplinary study, based on sedimentology, geochemistry and palynology, has enabled characterization of the dynamics of erosion and changes in plant cover over the last 14,000 years.In Lakes Allos and Vens, the start of the Holocene is marked by the maturing of ecosystems, by chemical weathering of soils and by forest encroachment at higher altitudes. These environmental changes resulted in progressive ecosystem stabilization, followed by a bio-pedological optimum, and subsequently by more open landscapes, and then soil destabilization. These four successive phases have ben dated, respectively, at 12,000-8000 cal. BP, 8000-6000 cal. BP, 6000-1900 cal. BP, and 1900 cal. BP to Present. The last of these phases occurred earlier, at 4200 cal. BP, in Lake Petit.The triggering of soil erosion has systematically been hinged on periods of heavy precipitation affecting landscapes rendered vulnerable, by human societies, to the effects of such perturbations.

No description available.

Western juniper has rapidly expanded into sagebrush steppe communities in the Intermountain West during the past 120 years. This expansion has occurred across a wide range of soil types and topographic positions. These plant communities, however, are typically treated in current peer-reviewed literature generically. The focus of this research is to investigate watershed level response to Western juniper encroachment at multiple topographic positions. Data collected from plots used to measure vegetation, soil moisture, and infiltration rates show that intercanopy sites within encroached Western juniper communities generally exhibit a significant decrease in intercanopy plant density and cover, decreased infiltration rates, increased water sediment content, and lower soil moisture content. High-resolution remotely sensed imagery and Geographic Information Systems were used with these plot level measurements to characterize and model the landscape-scale response for both biotic and abiotic components of a Western juniper encroached ecosystem. These data and their analyses included an inventory of plant density, plant cover, bare ground, gap distance and cover, a plant community classification of intercanopy patches and juniper canopy cover, soil moisture estimation, solar insulation prediction, slope and aspect. From these data, models were built that accurately predicted shrub density and shrub cover throughout the watershed study area, differentiated by aspect. We propose a new model of process-based plant community dynamics associated with current state-and-transition theory. This model is developed from field measurements and spatially explicit information that characterize the relationship between the matrix mountain big sagebrush plant community and intercanopy plant community patterns occurring within a Western juniper dominated woodland at a landscape scale. Model parameters (states, transitions, and thresholds) are developed based on differences in shrub density and cover, steady-state infiltration rates, water sediment content, and percent bare ground in response to juniper competition and topographic position. Results from both analysis of variance and multivariate hierarchical cluster analysis indicate that states, transitions, and thresholds can be accurately predicted for intercanopy areas occurring within the study area. In theory, this model and the GIS-based layers produced from this research can be used together to predict states, transitions, and thresholds for any location within the extent of the study area. This is a valuable tool for assessing sites at risk and those that have exceeded the ability to self-repair. / Graduation date: 2005

Until the 1930s, flows of the Colorado River maintained approximately 781,060 hectares of wetlands in its delta. These wetlands provided important feeding and nesting grounds for resident and migratory birds as well as spawning and protection habitat for many fish and other invertebrate species. However, the Delta's wetlands started to disappear as water was used for agricultural and urban uses in the United States and Mexico. The 1944 United States-Mexico water treaty, which allocates 1.8 million m��/year to Mexico, did not define a minimum flow to maintain the Delta's ecosystems. The resulting degraded Delta lead to the perception in the 1980s that the Delta was a dead ecosystem. This study investigates whether this "dead Delta" perception is valid. Its central hypothesis is that regenerated vegetation in riparian and flood plain zones is associated with surplus river flows during the 1990s. A vegetation analysis, using satellite imagery and field methods, shows that native trees have regenerated during the last 20 years, and now account for 23% of vegetation in a 100 km, non-perennial, stretch of river below the United States-Mexico border. A spatial trend analysis using multi-temporal data on percent vegetation cover indicates that there are 6,320 hectares that show a significant increasing trend (p-value<0.05) in vegetation cover, with the Delta's riparian zone having at least 18% of its area showing this trend. The study estimates that once in four years February to April flow of 300 million m�� (at 80-120 m��/s) is sufficient to germinate and establish new cohorts of native trees, and highlights the need for smaller but more periodic flows in order to maintain wetland areas. It is concluded that there is clear evidence of the resilience of the Delta's ecosystems and that the "dead Delta" perception is no longer valid. There exist critical habitat in the Delta that needs to be protected, while there also exist short and long term opportunities to ecologically enhance and expand current habitat. Hydrological and ecological studies are needed to estimate specific water requirements for these areas in order to efficiently target them for immediate and long term conservation actions. / Graduation date: 2003

Untersuchungen zur Rolle natürlicher Störungen in der Vegetation bzw. in Ökosystemen zeigen, dass natürliche Störungen ein essentielles und intrinsisches Element in Ökosystemen darstellen, substanziell zur Vitalität und strukturellen Diversität der Ökosysteme beitragen und Stoffkreisläufe sowohl auf dem lokalen als auch auf dem globalen Niveau beeinflussen. Feuer als Grasland-, Busch- oder Waldbrand ist ein besonderes Störungsagens, da es sowohl durch biotische als auch abiotische Umweltfaktoren verursacht wird. Es beeinflusst biogeochemische Kreisläufe und spielt für die chemische Zusammensetzung der Atmosphäre durch Freisetzung klimarelevanter Spurengase und Aerosole aus der Verbrennung von Biomasse eine bedeutende Rolle. Dies wird auch durch die Emission von ca. 3.9 Gt Kohlenstoff pro Jahr unterstrichen, was einen großen Anteil am globalen Gesamtaufkommen ausmacht.<br /> <br /> Ein kombiniertes Modell, das die Effekte und Rückkopplungen zwischen Feuer und Vegetation beschreibt, wurde erforderlich, als Änderungen in den Feuerregimes als Folge von Änderungen in der Landnutzung und dem Landmanagement festgestellt wurden. Diese Notwendigkeit wurde noch durch die Erkenntnis unterstrichen, daß die Menge verbrennender Biomasse als ein bedeutender Kohlenstoffluß sowohl die chemische Zusammensetzung der Atmosphäre und das Klima, aber auch die Vegetationsdynamik selbst beeinflusst. Die bereits existierenden Modellansätze reichen hier jedoch nicht aus, um entsprechende Untersuchungen durchzuführen. Als eine Schlussfolgerung daraus wurde eine optimale Menge von Faktoren gefunden, die das Auftreten und die Ausbreitung des Feuers, sowie deren ökosystemare Effekte ausreichend beschreiben. Ein solches Modell sollte die Merkmale beobachteter Feuerregime simulieren können und Analysen der Interaktionen zwischen Feuer und Vegetationsdynamik unterstützen, um auch Ursachen für bestimmte Änderungen in den Feuerregimes herausfinden zu können. Insbesondere die dynamischen Verknüpfungen zwischen Vegetation, Klima und Feuerprozessen sind von Bedeutung, um dynamische Rückkopplungen und Effekte einzelner, veränderter Umweltfaktoren zu analysieren. Dadurch ergab sich die Notwendigkeit, neue Feuermodelle zu entwickeln, die die genannten Untersuchungen erlauben und das Verständnis der Rolle des Feuer in der globalen Ökologie verbessern.<br /> <br /> Als Schlussfolgerung der Dissertation wird festgestellt, dass Feuchtebedingungen, ihre Andauer über die Zeit (Länge der Feuersaison) und die Streumenge die wichtigsten Komponenten darstellen, die die Verteilung der Feuerregime global beschreiben. Werden Zeitreihen einzelner Regionen simuliert, sollten besondere Entzündungsquellen, brandkritische Klimabedingungen und die Bestandesstruktur als zusätzliche Determinanten berücksichtigt werden. Die Bestandesstruktur verändert das Niveau des Auftretens und der Ausbreitung von Feuer, beeinflusst jedoch weniger dessen interannuelle Variabilität. Das es wichtig ist, die vollständige Wirkungskette wichtiger Feuerprozesse und deren Verknüpfungen mit der Vegetationsdynamik zu berücksichtigen, wird besonders unter Klimaänderungsbedingungen deutlich. Eine länger werdende, vom Klima abhängige Feuersaison bedeutet nicht automatisch eine im gleichen Maße anwachsende Menge verbrannter Biomasse. Sie kann durch Änderungen in der Produktivität der Vegetation gepuffert oder beschleunigt werden. Sowohl durch Änderungen der Bestandesstruktur als auch durch eine erhöhte Produktivität der Vegetation können Änderungen der Feuereigenschaften noch weiter intensiviert werden und zu noch höheren, feuerbezogenen Emissionen führen. / Studies of the role of disturbance in vegetation or ecosystems showed that disturbances are an essential and intrinsic element of ecosystems that contribute substantially to ecosystem health, to structural diversity of ecosystems and to nutrient cycling at the local as well as global level. Fire as a grassland, bush or forest fire is a special disturbance agent, since it is caused by biotic as well abiotic environmental factors. Fire affects biogeochemical cycles and plays an important role in atmospheric chemistry by releasing climate-sensitive trace gases and aerosols, and thus in the global carbon cycle by releasing approximately 3.9 Gt C p.a. through biomass burning. <br /> <br /> A combined model to describe effects and feedbacks between fire and vegetation became relevant as changes in fire regimes due to land use and land management were observed and the global dimension of biomass burnt as an important carbon flux to the atmosphere, its influence on atmospheric chemistry and climate as well as vegetation dynamics were emphasized. The existing modelling approaches would not allow these investigations. As a consequence, an optimal set of variables that best describes fire occurrence, fire spread and its effects in ecosystems had to be defined, which can simulate observed fire regimes and help to analyse interactions between fire and vegetation dynamics as well as to allude to the reasons behind changing fire regimes. Especially, dynamic links between vegetation, climate and fire processes are required to analyse dynamic feedbacks and effects of changes of single environmental factors. This led us to the point, where new fire models had to be developed that would allow the investigations, mentioned above, and could help to improve our understanding of the role of fire in global ecology. <br /> <br /> In conclusion of the thesis, one can state that moisture conditions, its persistence over time and fuel load are the important components that describe global fire pattern. If time series of a particular region are to be reproduced, specific ignition sources, fire-critical climate conditions and vegetation composition become additional determinants. Vegetation composition changes the level of fire occurrence and spread, but has limited impact on the inter-annual variability of fire. The importance to consider the full range of major fire processes and links to vegetation dynamics become apparent under climate change conditions. Increases in climate-dependent length of fire season does not automatically imply increases in biomass burnt, it can be buffered or accelerated by changes in vegetation productivity. Changes in vegetation composition as well as enhanced vegetation productivity can intensify changes in fire and lead to even more fire-related emissions. <br><br> ---<br> Anmerkung:<br> Die Autorin ist Trägerin des von der Mathematisch-Naturwissenschaftlichen Fakultät der Universität Potsdam vergebenen Michelson-Preises für die beste Promotion des Jahres 2002/2003.