Orientador: Ana Maria Goulart de Azevedo Tozzi / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Biologia / Made available in DSpace on 2018-08-15T18:24:29Z (GMT). No. of bitstreams: 1 Silva_EdsonDiasda_D.pdf: 12200976 bytes, checksum: 79cc0fa88c35507c64b4c435c1d6f438 (MD5) Previous issue date: 2010 / Resumo: Considerada a quinta área mais ameaçada e rica em espécies no mundo, a Floresta Atlântica está atualmente reduzida a pequenos fragmentos florestais, que representam apenas 7,6% da formação original. Inventários florísticos e estudos fitossociológicos realizados em várias regiões da Floresta Atlântica já estão oferecendo evidências da importância da família Leguminosae na composição e estrutura desse bioma. O presente estudo traz o inventário das espécies de Leguminosae que ocorrem em uma área de Floresta Ombrófila Densa do litoral norte do estado de São Paulo, entre Picinguaba e Santa Virgínia em altitudes que variam desde a restinga, próximo do nível do mar, até a vegetação do topo dos morros do Cuscuzeiro, Seringa e Corcovado, entre 1000 e 1200 m altitude. O levantamento florístico envolveu coletas de plantas arbóreas, arbustivas, herbáceas e lianas e foi realizado mensalmente entre 2006 e 2009 percorrendo trilhas nas diferentes fitofisionomias. A família Leguminosae está bem representada na Floresta Ombrófila Densa dos Núcleos Picinguaba e Santa Virgínia (140 espécies), sendo 18 Caesalpinioideae, 42 Mimosoideae e 80 Papilionoideae. Sessenta e uma espécies possuem hábito arbóreo, 56 são subarbustos e arbustos e 23 são lianas. É apresentada uma lista de gêneros e espécies com suas respectivas descrições, ilustrações, informações sobre hábitat, distribuição geográfica, períodos de floração e frutificação, além de chaves para a identificação específica, discussão sobre semelhanças fenotípicas e delimitação taxonômica. A análise de agrupamento realizada para 13 faixas altimétricas distintas revelou a existência de pelo menos três conjuntos ao longo do gradiente: um formado pelas espécies que ocorrem de 0-50 m, correspondente a Floresta de Restinga, outro ocorrendo de 800 a 1200 m, na faixa mais alta da Floresta Ombrófila Densa Montana e outro de 50 a 500 m, que abrange a Floresta Ombrófila Densa de Terras Baixas + Floresta Ombrófila Densa Submontana. A Floresta de Restinga apresentou a menor similaridade com as demais cotas altitudinais considerando todas as espécies levantadas ou apenas as subarbustivas e arbustivas. As árvores indicam a Floresta Ombrófila Densa Montana como a formação com a menor similaridade com as demais faixas altitudinais. A Floresta Ombrófila Densa de Terras Baixas e a Floresta Ombrófila Densa Submontana apresentaram maior similaridade, indicando haver alto compartilhamento de espécies entre essas duas fitofisionomias. As espécies da família Leguminosae, apesar de ocorrerem de forma bastante expressiva até os 1000 m, mostram-se mais adaptadas a ocupar áreas de menor altitude, entre 0 e 500 m, apresentando comparativamente menor riqueza específica nas áreas de maior altitude (1100 a 1200 m). Vinte e seis espécies de Leguminosae encontradas na área de estudo são endêmicas da Floresta Atlântica. / Abstract: Considered the fifth most endangered and richest in species area in the world, the Atlantic Forest is presently reduced to small forest fragments, which represent just 7,6% of the original formation. Floristic inventories and phytosociological studies accomplished in several regions of the Atlantic Forest are already offering evidences of the importance of the Leguminosae family in the composition and structure of this biome. This study brings the inventory of the Leguminosae species which occur in an area of Ombrophilous Dense Forest from the north coast of the state of São Paulo, in Picinguaba and Santa Virgínia in altitudes that vary from the Restinga Forest, close to the sea level, to the vegetation at the top of the Cuscuzeiro, Seringa e Corcovado mountains, between 1,000 at 1,200 meters of altitude. The floristic survey involved the collecting of herbs, shrubs, trees and lianas and was done monthly between 2006 and 2009 going through range in the different phytophysiognomies. The Leguminosae family is well represented in the Ombrophilous Dense Forest of Picinguaba and Santa Virgínia nucleus (140 species), being 18 Caesalpinioideae, 42 Mimosoideae and 80 Papilionoideae. Sixty-one species have arboreal habit, 56 are shrubs and 23 are lianas. A list is presented of the genera and species with their corresponding descriptions, illustrations, habit information, geographical distribution, flowering and fruiting periods and also keys to the specific identification, discussion on phenotipic similarities and taxonomic delimitation. The clustering analysis performed to 13 different altimetric range revealed the existence of at least three groups along the gradient: one formed by the species which occur from 0-50 m, corresponding to the Restinga Forest, another occurring from 800 to 1,200 m, on the highest level of Montane Ombrophilous Dense Forest and another from 50 to 500 m which comprises the Lowland Ombrophilous Dense Forest and the Submontane Ombrophilous Dense Forest. The Restinga Forest presented the smallest similarity to the other altitudinal quotas considering all the studied species or just the shrubs. The trees indicate the Montane Ombrophilous Dense Forest as the formation with the smallest similarity to the other altimetric range. The Lowland Ombrophilous Dense Forest and the Submontane Ombrophilous Dense Forest presented the greatest similarity, showing the existence of a great species share between these two phytophysiognomies. The Leguminosae family species seem most adapted to occupy lower altitude areas, between 0 and 500 m, comparatively presenting a smaller richness in highest altitude areas (1,100 to 1,200 m). Twenty-six Leguminosae species found in the studied area are endemic to the Atlantic Forest. / Doutorado / Doutor em Biologia Vegetal

Orientador: Simone Aparecida Vieira / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Biologia / Made available in DSpace on 2018-08-25T17:16:26Z (GMT). No. of bitstreams: 1 QuimbayoGuzman_LuisCarlos_M.pdf: 9203667 bytes, checksum: c85d0f23d04d1a3983335f8084f81a62 (MD5) Previous issue date: 2014 / Resumo: A madeira morta (MM) é um importante compartimento na dinâmica do Carbono nas florestas tropicais. Neste trabalho foi quantificado o estoque e a produção de MM em áreas de Floresta Ombrófila Densa (FOD) no litoral norte do estado de São Paulo, Brasil, ao longo de um gradiente altitudinal (10 a 1.066 m), onde a temperatura média anual diminui 0,6°C a cada 100 m de altitude. Os objetivos deste trabalho foram: (1) identificar o efeito das condições ambientais no estoque e na produção de MM ao longo do gradiente e (2) verificar o efeito a longo prazo (após 25 anos) da exploração seletiva de madeira no estoque e na produção de MM. O estudo foi desenvolvido em 16 parcelas de 1 ha, distribuídas nas formações vegetais: Restinga (10 m), Terras Baixas (10-100 m), Submontana (100-500 m) e Montana (500-1.066 m). A produção e o estoque de MM caída (Ø ? 2 cm) foram quantificados através do método de linhas de interceptação (anos 2012 e 2013). O estoque de MM em pé foi quantificado por meio de inventário florestal considerando as árvores mortas em pé (DAP ? 4,8 cm) da base de dados do projeto BIOTA Gradiente Funcional/FAPESP (anos 2006, 2008 e 2012). A MM caída foi classificada em 5 categorias, de acordo com seu grau de decomposição e as árvores mortas em pé em 4 categorias. A produção de MM caída foi de 4,21±0,5 Mg ha-1 ano-1 (± erro padrão). Não foi encontrada relação entre produção e altitude. Não houve diferença na produção entre os períodos de maior (4,44±0,7 Mg ha-1 ano-1) e menor (3,99±0,6 Mg ha-1 ano-1) intensidade de chuva (teste t pareado duas caudas; GL=15; t=0,53; p=0,6), nem entre locais explorados e não explorados. O tempo de residência estimado da MM caída foi de 1,5 anos para a madeira fina (Ø de 2 a 10 cm) e de 6 anos para a madeira grossa (Ø > 10 cm). A taxa de decomposição estimada foi 0,67 ano-1 para madeira fina e 0,18 ano-1 para madeira grossa (18% da massa de MM grossa é decomposta por ano). O estoque total de MM (caída + em pé) foi 16,4±1,3 Mg ha-1, 81% corresponde a MM caída (13,3±1,2 Mg ha-1), enquanto só 19% corresponde a MM em pé (3,1±0,3 Mg ha-1). O estoque de MM caída não diferiu entre os anos 2012 (13,3±1,2 Mg ha-1) e 2013 (13,1±1,4 Mg ha-1) (teste t pareado duas caudas; GL=15; t=0,15; p=0,88). A MM equivale de 4% a 10% da biomassa viva acima do solo. O estoque médio de Carbono na MM foi de 7,5±0,6 Mg C ha-1. No ano 2013, o estoque de MM aumentou com a altitude (regressão linear; R2=0,3; p=0,02). O estoque de MM foi 7,1±0,9 Mg ha-1 na formação Restinga; e aumentou gradativamente até 22,1±1,4 Mg ha-1 na formação Montana, a 1.050 m de altitude, o que confirmou a hipótese de que as condições climáticas têm influência no estoque de MM, encontrando-se mais MM nos locais mais frios e menos chuvosos. Não houve diferença no estoque de MM entre locais explorados e não explorados. Conclui-se que a relação positiva entre altitude e estoque de MM é determinada principalmente pela diminuição da temperatura do solo com a altitude / Abstract: Deadwood (DW) is an important component of carbon dynamics in tropical forests. In this work, DW stocks and production were quantified along an altitudinal gradient (10 to 1066 m) in an Ombrophilous Dense Atlantic Forest. Along the gradient, the mean annual temperature decreases 0.6°C with every increase of 100 m in altitude. The objectives were: (1) to identify the effect of environmental conditions on DW stocks and production along the gradient, and (2) to verify the long term effect (after 25 years) of selective wood harvesting on DW stocks and production. The study was carried out in 16 plots (1 ha each), distributed in four vegetal formations: Restinga (10 m), Low lands (10-100 m), Submontane (100-500 m) and Montane (500-1066 m). Fallen DW stocks and production (Ø ? 2 cm) were quantified using line intercept sampling (years 2012 and 2013). Standing DW stocks were quantified via forest inventory, considering all standing dead trees (DBH ? 4.8 cm) from BIOTA Gradiente Funcional/FAPESP project database (years 2006, 2008 and 2012). Fallen DW was classified into five classes according to its decomposition state, and standing DW into four classes. Fallen DW production was 4.21±0.5 Mg ha-1 yr-1 (± standard error). No relationship between production and altitude was found. There was no difference in DW production between periods of high (4.44±0.7 Mg ha-1 yr-1) and low (3.99±0.6 Mg ha-1 yr-1) rain intensities (two tailed paired t-test; DF=15; t=0.53; p=0.6), or between harvested and pristine areas. Estimated residence time was 1.5 years for fine fallen DW (Ø between 2 and 10 cm) and 6 years for coarse fallen DW (Ø > 10 cm). Estimated decay rate was 0.67 yr-1 for fine and 0.18 yr-1 for coarse fallen DW (18% of coarse fallen DW is decomposed per year). Total DW stocks (fallen + standing) were 16.4 ±1.3 Mg ha-1, 81% corresponds to fallen DW (13.3±1.2 Mg ha-1), while only 19% corresponds to standing DW (3.1±0.3 Mg ha-1). Fallen DW stocks didn¿t change between years 2012 (13.3±1.2 Mg ha-1) and 2013 (13.1±1.4 Mg ha-1) (two tailed paired t-test; DF=15; t=0.15; p=0.88). DW equals 4% to 10% of aboveground live biomass stocks. Mean carbon stocks in DW were 7.5±0.6 Mg C ha-1. In 2013, DW stocks increased with altitude (linear regression; R2=0.3; p=0.02). DW stocks were 7.1±0.9 Mg ha-1 in the Restinga formation, and gradually increased to 22.1±1.4 Mg ha-1 in the Montane formation, which confirmed the initial hypothesis of climatic conditions influencing DW stocks, given that colder and less rainy areas presented more DW. There was no difference in DW stocks between harvested and pristine areas. It was concluded that the positive relationship between DW stocks and altitude is mainly determined by decrease in soil temperature with altitude / Mestrado / Ecologia / Mestre em Ecologia

Cette thèse a pour objectif d’établir une vision globale et continue mer-terre-mer d’une barrière littorale paraglaciaire, en combinant une approche morphologique et architecturale. L’originalité de l’isthme de Miquelon-Langlade (NO Atlantique) entre les deux îles éponymes réside en sa diversité et complexité de systèmes (flèches, cordons littoraux, lagune) qui s’emboitent formant une barrière composite évoluant le long de deux façades ouvertes sur l’océan. Ce travail a été rendu possible grâce à l’association de données topographiques (GPS-RTK), de géophysique marine (sismique HR) et terrestre (géo-radar), sédimentaires (carottes, échantillons de surface) ainsi que des datations par OSL. L’étude de cette barrière a permis 1) d’identifier l’ensemble des dépôts d’une séquence paraglaciaire (dépôts de chute, bas niveau, augmentation et haut niveau marin); 2) de pointer l’importance de la géométrie des îles, du substratum rocheux ainsi que de la topographie héritée dans la formation de la barrière; 3) de définir l’architecture interne de beach ridges mixte sable-galet; et 4) de pointer le potentiel de ces systèmes en tant marqueur du paléo-niveau marin tout en proposant la première courbe de tendance de l’évolution du RSL pour l’Archipel de Saint-Pierre-et-Miquelon sur les derniers 3000 ans (+1mm/an). L’association de l’ensemble de ces informations a permis de proposer un schéma de mise en place de la barrière, depuis le retrait de la calotte (13 700 ans) jusqu’à sa forme actuelle, évoluant dans un contexte d’augmentation du RSL et de fort apport sédimentaire, tout en proposant une vision de détail de la formation de la plaine de beach ridges au cours des derniers 3000 ans. / The purpose of this PhD thesis was to define a complete and continuous sea-land-sea view of a paraglacial coastal barrier system, combining morphological and architectural approaches. Particularity of the Miquelon-Langlade isthmus (NO Atlantic) lies in its diversity and complexity of systems (spits, beach ridges, and lagoon) that fit together to form a composite barrier prograding along two open coasts. Investigations of this barrier included topographic (RTK-GPS), ground-penetrating radar (GPR) seismic, sedimentary (cores, augers, samples) and chronologic (OSL) data collections. This study of this paraglacial composite barrier make possible 1) to characterize all deposits of the paraglacial sequence (FSST, LST, TST, HST); 2) to emphasize the primary influence of islands geometry, bedrock and inherited topography on the coastal progradational pattern; 3) to provide a better understanding of the internal architecture of depositional processes and driving mechanisms of mixed sand-and-gravel beach ridges; 4) to examine the potential of these mixed beach-ridge systems to record and preserve paleo-sea-level information, to produce the first sea level trends for the Saint-Pierre-et-Miquelon archipelago over the last 3000 years. The combination of all these data provide a model of development of the barrier since the ice sheet retreat (13 700 yrs) to its present form, in a context of RSL rise and high local glaciogenic sediment supply, and a detailed view of the formation of the beach ridge plain over the last 3000 yrs.

Les connaissances sur la morphologie de la Terre sont essentielles à la compréhension d’une variété de processus géomorphologiques et hydrologiques. Des avancées récentes dans le domaine de la télédétection ont significativement fait progresser notre habilité à se représenter la surface de la Terre. Parmi celles-ci, les données d’élévation LiDAR permettent la production de modèles numériques d’altitude (MNA) à haute résolution sur de grands territoires. Le LiDAR est une avancée technologique majeure permettant aux scientifiques de visualiser en détail la morphologie de la Terre et de représenter des reliefs peu prononcés, et ce, même sous la canopée des arbres. Une telle avancée technologique appelle au développement de nouvelles approches innovantes afin d’en réaliser le potentiel scientifique. Dans ce contexte, le présent travail vise à développer deux approches de cartographie numérique utilisant des données d’élévation LiDAR et servant à l’évaluation de la composition du sous-sol. La première approche à être développée utilise la localisation de crêtes de plage identifiées sur des MNA LiDAR afin de modéliser l’étendue maximale de la mer de Champlain, une large paléo-mer régionalement importante. Cette approche nous a permis de cartographier avec précision les 65 000 km2 autrefois inondés par la mer. Ce modèle sert à l’évaluation de la distribution des sédiments marins et littoraux dans les basses-terres du Saint-Laurent. La seconde approche utilise la relation entre des échantillons de matériel parental des sols (MPS) et des attributs topographiques dérivés de données LiDAR afin de cartographier à haute résolution et à une échelle régionale le MPS sur le Bouclier canadien. Pour ce faire, nous utilisons une approche novatrice combinant l’analyse d’image orientée-objet (AIOO) avec une classification par arbre décisionnel. Cette approche nous a permis de produire une carte du MPS à haute résolution sur plus de 185 km2 dans un environnement hétérogène de post-glaciation. Les connaissances issues de la production de ces deux modèles ont permis de conceptualiser la composition du sous-sol dans les régions limitrophes entre les basses-terres du Saint-Laurent et le Bouclier canadien. Ce modèle fournit aux chercheurs et aux gestionnaires de ressources des connaissances détaillées sur la géomorphologie de cette région et contribue à l’amélioration de notre capacité à saisir les services écosystémiques et à prédire les aléas environnementaux liés aux processus du sous-sol. / Knowledge of the earth’s morphology is essential to the understanding of many geomorphic and hydrologic processes. Recent advancements in the field of remote sensing have significantly improved our ability to assess the earth’s surface. From these, LiDAR elevation data permits the production of high-resolution digital elevation models (DEMs) over large areas. LiDAR is a major technological advance as it allows geoscientists to visualize the earth’s morphology in high detail, even allowing us to resolve low-relief landforms in forested areas where the surface is obstructed by vegetation cover. Such a technological advance calls for the development of new and novel approaches to realize the scientific potential of this new spatial data. In this context, the present work aims to develop two digital mapping approaches that use LiDAR elevation data for assessing the earth’s subsurface composition. The first approach to be developed uses the location of low-relief beach ridges observed on LiDAR-derived DEMs to map the extent of a large and regionally important paleo-sea, the Champlain Sea. This approach allowed us to accurately map the 65,000 km2 area once inundated by sea water. The model serves to the assessment of the distribution of marine and littoral sediments in the St. Lawrence Lowlands. The second approach uses the relationship between field-acquired samples of soil parent material (SPM) and LiDAR-derived topographic attributes to map SPM at high-resolution and at a regional scale on the Canadian Shield. To do so, we used a novel approach that combined object-based image analysis (OBIA) with a classification tree algorithm. This approach allowed us to produce a fine-resolution 185 km2 map of SPM in a heterogeneous post-glaciation Precambrian Shield setting. The knowledge obtained from producing these two models allowed us to conceptualize the subsurface composition at the limit between the St. Lawrence Lowlands and the Canadian Shield. This insight provides researchers and resource managers with a more detailed understanding of the geomorphology of this area and contributes to improve our capacity to grasp ecosystem services and predict environmental hazards related to subsurface processes.