Conselho Nacional de Desenvolvimento Científico e Tecnológico / Populations and species are disappearing due to disturbances in the environment caused by human activities. Given the obvious risk of loss of diversity, it is increasingly necessary to take actions concerning preservation, in which safety features are necessary for measuring the variation of diversity in space and time. The aim of this study was to evaluate the structure and diversity in the arboreal component and natural regeneration in an area of Araucaria forest in the National Forest of São Francisco de Paula, RS. The vegetation sampling was performed by analyzing 180 subunits of 10m x 10m, where all the arboreal individuals and natural regeneration were inventoried. The parameters of frequency and density, dispersion indexes of MacGuinnes, Fracker and Brischle and Payandeh, and different alpha and beta indexes of diversity were calculated. The values of alpha indices were correlated with environmental variables by Canonical Correspondence Analysis (CCA). It was found 102 species, distributed in 66 genera and 36 families. The dispersion indexes showed different patterns for the arboreal component and natural regeneration. The Margalef, Shannon, and the Beta indexes were underestimated, possibly influenced by the size of sample unit. Index Menhinick represented the diversity in a very real form, even in small sampling units. The indexes of Simpson and Mcintosh denote low dominance and the equity indexes showed high uniformity in species. The CCA showed that diversity is strongly related to declive, stoniness and contents of sand and clay, in the arboreal component, and contents of clay, fine sand and silt, in natural regeneration. / Populações e espécies estão desaparecendo devido às perturbações no ambiente por meio de atividades antrópicas. Diante do risco evidente de perda de biodiversidade, deve-se cada vez mais realizar ações de preservação, nas quais são necessárias ferramentas seguras capazes de medir a variação da diversidade no espaço e no tempo. Objetivou-se com esse trabalho avaliar a estrutura e a diversidade no componente arbóreo e na regeneração natural em área de Floresta Ombrófila Mista na Floresta Nacional de São Francisco de Paula, RS. A amostragem da vegetação foi realizada por meio de 180 subunidades de 10m x 10m, onde foram inventariados todos os indivíduos arbóreos e da regeneração natural. Foram calculados os parâmetros de frequência e densidade, os índices de dispersão de MacGuinnes, Fracker e Brischle e Payandeh, e diferentes índices de diversidade alfa e beta. Os valores dos índices alfa foram correlacionados com variáveis ambientais pela Análise de Correspondência Canônica (CCA). Foram encontradas 102 espécies, distribuídas em 66 gêneros e 36 famílias. Os índices de dispersão revelaram diferentes padrões para o componente arbóreo e a regeneração natural. Os índices de Margalef, Shannon e os índices Beta obtiveram valores subestimados possivelmente influenciados pelo tamanho da unidade amostral. O índice de Menhinick representou a diversidade de forma bastante real mesmo em unidades amostrais pequenas. Os índices de Simpson e Mcintosh denotaram baixa dominância e os índices de equidade demonstraram alta uniformidade nas espécies. A CCA mostrou que a diversidade está fortemente relacionada à declividade, pedregosidade e teores de areia grossa e argila no componente arbóreo, e a teores de argila, areia fina e silte na regeneração natural.

Empresa Brasileira de Pesquisa Agropecuária / The present work was accomplished in the vicinity of water springs located in remaining portions of phytogeographical formations of the Western part of the Santa Catarina State in Brazil. Its main objective is to study the similarity between the formations and evaluate the relationship between the abundance of species and environmental variables. The database for this study was nine water springs, three from each phytogeographical formation. The sampling was done along two intercepting transects, with 26 plots of 10m x 20m (200m²) per spring. Every arboreal-arbustive species with a diameter at breast height greater than 5.0 cm was analyzed for this study. The environmental (edaphic) variables were obtained from samples of soil collected on each plot. Furthermore, the slope of the soil, sunlight exposure, geographical coordinates and altitude were measured and taken into account. The thesis is divided into five chapters: the first one has as its objective to determine, by means of grouping analysis, the similarity between forest formations of the area of study, indicating that the three formations are floristically different, which brings up the need of separate analysis of each forest formation given the vulnerability of the environments. In the second, third and fourth chapters the abundance of species was linked to the environmental variables (edaphic and topographic) present in each phytogeographical formation. The results unveiled a great variation of the vegetation amongst the springs, presenting particular indicator species for each one. Furthermore, the results made possible to determine the most flexible species, known as generalists, which are capable of developing in different environments. The fifth chapter has as its objective to recognize the functional ecological patterns of the arborescent community of portions of land in the vicinity of water springs of the phytogeographical formations of the Western part of the Santa Catarina State in Brazil. / O trabalho foi realizado no entorno de nascentes em fragmentos das formações fitogeográficas no Oeste do estado de Santa Catarina, tendo como objetivo principal estudar a similaridade entre elas e avaliar a correlação da distribuição da abundância das espécies com variáveis ambientais. A base de dados para o estudo foi nove nascentes, três de cada formação fitogeográfica. As amostragens foram feitas na forma de transecto cruzado, com 26 parcelas de 10 m x 20 m (200 m2) por nascente. Foram avaliadas todas as espécies arbóreo-arbustivas com diâmetro à altura do peito ≥ 5,0 cm. As variáveis ambientais (edáficas) foram obtidas a partir de coletas compostas do solo em cada parcela. Além disso, foram medidos o desnível do solo, a exposição solar, as coordenadas geográficas e a altitude. A tese foi subdividida em cinco capítulos: o primeiro teve por objetivo determinar, por meio de análise de agrupamento, a similaridade entre as formações florestais da região de estudo, indicando que as três formações são floristicamente diferentes, havendo, portanto, a necessidade de analisar cada formação florestal em separado, por se tratar de ambientes altamente vulneráveis. No segundo, terceiro e quarto capítulos, a abundância das espécies foi correlacionada com as variáveis ambientais (edáficas e topográficas) dentro de cada formação fitogeográfica. Os resultados revelaram grande variação da vegetação entre as nascentes, apresentando espécies indicadoras para cada nascente. Além disso, neles buscou-se relacionar as espécies de maior plasticidade, denominadas de generalistas, por possuírem a capacidade de se desenvolver na maioria dos ambientes. O quinto capítulo teve por objetivo reconhecer os padrões ecológicos funcionais da comunidade arborescente de fragmentos no entorno de nascentes das formações fitogeográficas do Oeste do estado de Santa Catarina.

L’écologie de la restauration est une discipline scientifique qui a vu le jour il y a une quarantaine d’années pour tenter de compenser les impacts négatifs du développement des sociétés sur les écosystèmes. Aujourd’hui, suite à des méta-analyses planétaires révélant le succès relatif des opérations de restauration écologique, une nouvelle question émerge : faut-il restaurer activement ou laisser en libre évolution les écosystèmes dégradés ? C’est dans ce contexte que ce projet de thèse a eu lieu avec une démarche qui s’est voulue pluridisciplinaire et a porté sur l’étude de l’impact de carrières sur la steppe méditerranéenne de Crau (Bouches-du-Rhône, France). Deux cas ont été pris en compte, (i) d’anciennes carrières exploitées dans les années 1970 et abandonnées présentant une diversité de modalités d’exploitation ou de réaménagement et, (ii) une carrière encore en cours d’exploitation dont la législation oblige la remise en état. L’objectif est de mieux identifier les éventuels verrous scientifiques en matière de connaissances pour la restauration et mieux définir les attentes sociétales afin de proposer au final une stratégie générale destinée à servir à la gestion future de ces écosystèmes. Les recherches de cette thèse se sont basées sur deux grandes questions, réflexions : (1) Les opérations de restauration écologique mises en place permettent-elles de restaurer l’écosystème de référence (= ici l’écosystème préexistant) ? En comparant différents traitements de restauration sur le long terme (transfert de sol selon différentes modalités, dépôts de matériaux anthropogéniques, absence de réhabilitation), nous avons pu montrer que le transfert de sol reste une bonne méthode, surtout lorsque les caractéristiques initiales du sol sont respectées. Néanmoins, il ne compense toujours pas à moyen-terme (35 ans) la destruction de l'écosystème préexistant : le sol et la communauté végétale de la steppe de référence ne sont pas encore complètement rétablis. Une deuxième technique de mélange de sol lorsque le sol originel n’est plus disponible a également été testée, mais elle ne présente pas non plus un succès total de restauration à court terme (3 ans). Une deuxième question s’est donc posée en parallèle: (2) Sans restauration active, quelle est la valeur de la biodiversité générée par les activités humaines ? Et plus globalement, quelles natures voulons-nous ? Les carrières ont détruit l’écosystème steppique qui préexistait mais ont créé également de nouvelles conditions (pédologiques, de nouveaux habitats) qui soutiennent le fonctionnement et la connectivité d’espèces pionnières et abritent une importante biodiversité patrimoniale absente de l’écosystème d’origine. De plus, la comparaison entre les paysages de carrières et l’écosystème de référence auprès de différents acteurs territoriaux et du grand public a montré que les anciennes carrières sont perçues comme étant beaucoup plus naturelles que la steppe de référence et qu’elles sont également associées à une importante biodiversité. Ces résultats pourraient donc réorienter les choix de restauration ou de gestion, afin de choisir entre 1) ce qui est actuellement recommandé (une restauration active appliquée en fin d’exploitation avec l’écosystème historique en référence) et 2) une libre expression de la nature férale encore appelée restauration passive (avec ou pas réaffectation initiale ; i.e. où l’écosystème de référence est différent de l’écosystème préexistant). / Restoration ecology is a scientific discipline that has emerged forty years ago to try to compensate the negative impacts of society development on ecosystems. Today, following global meta-analyzes revealing the relative success of ecological restoration, a new question emerges: should we actively or passively restore degraded ecosystems? It is in this context that this thesis project took place with a multidisciplinary approach and focused on the study of quarrying impacts on the Mediterranean steppe of Crau (Bouches-du-Rhône, France). Two cases were taken into account: (i) old quarries operated in the 1970s and abandoned then, presenting a variety of exploitation types or rehabilitation modalities, and (ii) a quarry still in the process of exploitation, the legislation now requiring repairs. The aim is to better identify possible scientific obstacles in the field of knowledge for ecological restoration and better define societal expectations in order to finally propose a general strategy intended to serve the future management of these ecosystems. The research of this thesis was based on two major questions, reflections: (1) Do ecological restoration actions restore the reference ecosystem (= the pre-existing ecosystem)? By comparing different long-term restoration treatments (various types of soil transfers, anthropogenic deposits, lack of rehabilitation), we were able to show that soil transfer is still the best method, especially when the initial characteristics of the soil are respected. However, it still does not compensate in the medium term (35 years) for the destruction of the pre-existing ecosystem: the soil and the plant community of the reference steppe are not completely restored yet. A second soil mixing technique used when the original soil is no longer available has also been tested, but it is not very successful either in the short-term (3 years). A second question arose in parallel: (2) Without active restoration, what is the value of biodiversity generated by human activities? And more generally, what kind of nature do we want? Quarries have destroyed the pre-existing steppe ecosystem but have also created new conditions (soil, new habitats) that support the functioning and connectivity of pioneer species and shelter important heritage biodiversity that is absent from the pre-existing ecosystem. In addition, the comparison between the quarry landscapes and the reference ecosystem landscapes with different stakeholders and the general public has shown that the old quarries are perceived as being much more natural than the reference steppe and that they are also associated to an important biodiversity. These results could therefore reorient the choice of restoration or management, in order to choose between 1) what is currently recommended (active restoration applied after the end of quarry exploitation with the historical ecosystem as a reference) and 2) the free expression of feral nature also called passive restoration (with or without reclamation, i.e. where the reference ecosystem is different from the pre-existing ecosystem).

Hlogoma Mountain is a small inselberg surrounded by farms and commercial forestry in the Underberg district (KwaZulu-Natal) within the Gs10 Drakensberg Foothill Moist Grassland. As little is known about the vegetation on Hlogoma, a survey was undertaken to classify, map and describe the plant communities occurring on the inselberg. A total of 100 (16 m2) randomly stratified sample plots were placed in homogeneous vegetation units within the 117 ha study area. A TWINSPAN classification, refined by Braun-Blanquet procedures, resulted in the identification of two major communities, five communities, 12 sub-communities and four variants. All communities were described and a vegetation map constructed. Ordinations identified key environmental variables that have an impact on the vegetation at the study site. A plant species checklist was created and analysed for floristic composition; rarity and threatened species; medicinal plants; endemism; phenology; flower colour and elevation range; species richness and plant community biodiversity. A total of 467 species were identified, represented by 271 genera and 87 families. Four Red Data species and a new Aspidoglossum species were discovered. Two near-endemic genera, two endemic species and 88 near-endemic species belonging to the Drakensberg Alpine Centre were found on Hlogoma. This study showed that Hlogoma Mountain is an inselberg with high plant species richness and endemism, and is a refuge of conservation importance for biodiversity. / Environmental Sciences / M. Sc.(Environmental Science)

A vegetation survey was conducted on the newly acquired farm portions of the farm Doornrandje 386 JR, which are being incorporated into Gauteng Department of Agriculture and Rural Development’s protected area expansion. This study provides an ecological basis for establishing an efficient management programme for the area. From a TWINSPAN classification, refined by Braun-Blanquet procedures, five plant communities were identified. A classification and description of the major plant communities is presented. Descriptions of the plant communities include characteristic species, as well as prominent and less conspicuous species of the tree, shrub, herb and grass strata. Floristic analyses of the vegetation and a biodiversity comparison of the different plant communities found on the farm are also undertaken. This study proves that the extended land incorporated into the Reserve contributes to the biological diversity of the study area. / Environmental Sciences / M. Sc. (Environmental Science)

A vegetation survey was conducted on the newly acquired farm portions of the farm Doornrandje 386 JR, which are being incorporated into Gauteng Department of Agriculture and Rural Development’s protected area expansion. This study provides an ecological basis for establishing an efficient management programme for the area. From a TWINSPAN classification, refined by Braun-Blanquet procedures, five plant communities were identified. A classification and description of the major plant communities is presented. Descriptions of the plant communities include characteristic species, as well as prominent and less conspicuous species of the tree, shrub, herb and grass strata. Floristic analyses of the vegetation and a biodiversity comparison of the different plant communities found on the farm are also undertaken. This study proves that the extended land incorporated into the Reserve contributes to the biological diversity of the study area. / Environmental Sciences / M. Sc. (Environmental Science)

Floristic, Ecological and Phytosociological, Study of the Forests of the Mbiye Island at Kisangani, DR Congo The Mbiye Island is situated on the Congo River, in the Eastern part of Kisangani. It is located upstream of the Wagenia Falls, between latitude 0º31’ North and longitude 25º11’ East, with 376 m of altitude. It adjoins the town of Kisangani, and it is 14 km long and 4 km wide. All around Kisangani, the Mbiye Island is the only ecosystem which has a dense forest that is relatively well preserved. This Island has an area of 1,400 ha, and it comprises three types of forest: dry land forest, periodically flooded forest and swampy forest. The main purpose of this study has been to make a qualitative and quantitative analysis of the physiognomy and structure of these three types of forest. I relied on two methods in data collecting: the Phytosociological plotting method with a sampling equal to 100 individuals for the higher strata and 100 m² grass strata, associated with dbh measure for trees whose diameter is ≥ 10 cm. In total, out of the 80 plottings, 67 multi strata and structurals each of which has been decomposed in 4 sub-plottings fit together. Based on the presence-absence criterion, the other 13 multi strata and structurals have been made into Chablis, and all the data have been globally analyzed. An inventory of 33991 individuals has been made in this study. After their identification, it has been found out that there 470 species belonging to 297 types and 90 families. Rubiaceae, Caesalpiniaceae and euphorbiaceae families have been found out to be the most important in Genus and species. In the raw spectra of biological types, the phanerophyts (83.0%) are the most predominant, and within which the mesophanerophyts (trees 10 to 30 m tall in their old age) are the most predominant (39.0%). The Phytogeographical spectra have more omni-guineo-congolese (35%) elements; and amongst the diaspore types, the sarcochores (69%) are the most predominant. 58 endemic species (12.3%) of the central forest sector have been found out at the Mbiye Island. After the classification and ordination of the 80 plottings by Twinspan and DCA, three Phytosociological groupings have been found out in the dry land forest: (1) grouping of Funtumia elastica and Albizia ealaensis, (2) grouping of Nesogordonia leplaei and Antiaris toxicaria, (3) prouping of Olyra latifolia and Campylospermum elongatum; two others for floodable forests (1) grouping of Rinorea oblongifolia and Byrsocarpus coccineus, (2) grouping of Pachystela seretii, Dracaena kindtiana and Culcasia yangambiensis, and three others for swampy forests: (1) grouping of Strombosia grandifolia, Calamus deerratus, Culcasia angolensis and Elaeis guineensis, (2) grouping of Cleistopholis patens and Pterygota bequaertii, (3) grouping of Diospyros bipendensis, Strychnos icaja and Palisota barteri and then two groupings for the Chablis plottings (1) grouping of Palisota schweinfurthii, Eremospatha haullevilleana, Thomandersia hensii and Pycnocoma insularis and (2) grouping of Nephrolepis biserrata and Scaphopetalum thonneri. The coelocaryon botryoides has shown a uniform spatial distribution in the in the forest periodically flooded, whereas Gilbertiodendron dewevrei has shown an aggregated distribution. Comparison of richness in different taxa within the sampling has given high values as far as the numbers of species is concerned and genus of the same family in the phytosociological sampling. Out of all the species listed, 189 have been considered to be common species with regard to the three forest types, other 144 belonged to two types. 90 species have been listed only on solid land, 27 on periodically flooded forest 20 on swampy forest. Euphorbiaceae, Rubiaceae and Caesalpiniaceae families were the best represented in species and genus. With regard to richness in each forest type, the forest of solid soil was the only one to have high values both in species<p>and genus, and the swampy forest was less represented. With reference to richness according to strata, the herbaceous and under-shrub stratum (H) was the richest whereas the stratum of dominant trees (E+A1) was the poorest. Mbiye island forest is the nearest one of Kisangani for which it plays the most important role with regards to supplies in different forest products and non-ligneous forest products. In total, 12.3% endemic species of Central forest sector have been listed there, but due to frequent use by man this time, all these endemic species may disappear from this island. It is therefore urgent that particular attention be paid on it. Actions of conservation and lasting exploitation must be started in order to block their extinction. Key words: DRCongo, Flora, Ecology, Phytosociology, insularity, flooding, spatial distribution, biodiversity, synusie, Mbiye Island, Kisangani./Etude floristique, écologique et phytosociologique des forêts de l'île Mbiye, Kisangani, RDCongo. L'île Mbiye est une île du fleuve Congo située dans la partie Est de la ville de Kisangani, en amont des chutes Wagenia, entre 0° 31' de latitude Nord et 25° 11' de longitude Est, avec une altitude de 376 m. Elle jouxte la ville de Kisangani et sa longueur maximale est de 14 km sur 4 km de largeur. Elle est le seul écosystème à proximité de la ville, possédant encore une forêt dense relativement bien conservée et formant un îlot de 1400 ha comprenant trois types forestiers :la forêt de terre ferme, la forêt périodiquement inondée et la forêt marécageuse. L’objectif principal de cette étude a été de faire une analyse quantitative et qualitative de la physionomie et la structure de ces trois types forestiers. Deux méthodes ont permis la récolte de données ;la méthode de relevés phytosociologiques à effort d’échantillonnage égal à 100 individus pour les strates supérieures et 100 m² pour la strate herbacée, associée à celle de mesure de D130, pour les arbres à diamètre ≥ 10 cm. Au total, 80 relevés dont 67 multistrates et structurels, décomposés chacun en 4 sous-relevés emboîtés, ont été mis en place. 13 autres ont été faits dans les chablis avec l’appréciation basée sur le critère de présence-absence et toutes les données ont été analysées globalement. A l’issue de cette étude, 33991 individus ont été recensés. Après leur identification, au total 470 espèces appartenant à 297 genres et 90 familles ont été obtenues. Les familles des Rubiaceae, des Caesalpiniaceae et des Euphorbiaceae sont les plus importantes, tant en genres qu’en espèces. Dans les spectres bruts de types biologiques, les phanérophytes (83,0 %) sont les plus abondants et au sein desquels les mésophanérophytes (arbres de 10 à 30 m au stade adulte) sont les plus dominants (39,0 %). Les spectres phytogéographiques ont montré une prédominance d’éléments omni-guinéo-congolais (35 %) alors que parmi les types de diaspores, les sarcochores (69 %) sont les plus dominants. 58 espèces endémiques du Secteur forestier central (soit 12,3 %) ont été recensées à l’île Mbiye. Après classification et ordination de ces 80 relevés par Twinspan et DCA, trois groupements phytosociologiques ont été obtenus dans la forêt de terre ferme: (1) groupement à Funtumia elastica et Albizia ealaensis, (2) groupement à Nesogordonia leplaei et Antiaris toxicaria, (3) groupement à Olyra latifolia et Campylospermum elongatum ;deux autres pour les forêts inondables (1) groupement à Rinorea oblongifolia et Byrsocarpus coccineus, (2) groupement à Pachystela seretii, Dracaena kindtiana et Culcasia yangambiensis ;trois autres encore pour les forêts marécageuses :(1) groupement à Strombosia grandifolia, Calamus deerratus, Culcasia angolensis et Elaeis guineensis, (2) groupement à Cleistopholis patens et Pterygota bequaertii, (3) groupement à Diospyros bipendensis, Strychnos icaja et Palisota barteri, et enfin, deux groupements pour les relevés de chablis (1) groupement à Palisota schweinfurthii, Eremospatha haullevilleana, Thomandersia hensii et Pycnocoma insularum et (2) groupement à Nephrolepis biserrata et Scaphopetalum thonneri. L’espèce Coelocaryon botryoides a montré une répartition spatiale uniforme dans la forêt périodiquement inondée alors que Gilbertiodendron dewevrei y a montré une répartition agrégée. La comparaison de la richesse en différents taxa au sein de relevés et de placettes a donné de valeurs supérieures en nombre d’espèces, de genres et même de familles, dans les relevés phytociologiques. Sur le total des espèces recensées, 189 ont été considérées comme espèces communes aux trois types forestiers, 144 autres appartenaient aux deux types. 90 espèces<p>n’ont été recensées qu’en forêt de terre ferme, 27 en forêt périodiquement inondée et 20 dans la forêt marécageuse. Les familles des Euphorbiaceae, des Rubiaceae et celle des Caesalpiniaceae ont été les mieux représentées en espèces et en genres. En ce qui concerne la richesse dans chaque type forestier, la forêt de terre ferme était la seule à avoir des valeurs supérieures tant en espèces qu’en genres et la forêt marécageuse était la moins représentée. Quant à la richesse spécifique selon les strates, la strate herbacée et sous-arbustive (H) était la plus riche alors que la strate des arbres dominants (E+A1) était la plus pauvre. La forêt de l’île Mbiye est la seule forêt proche de la ville de Kisangani pour laquelle elle joue un rôle de premier plan en ce qui concerne le ravitaillement en différents produits forestiers et forestiers non ligneux. Au total, 12,3% d’espèces endémiques du Secteur forestier Central y ont été recensées, mais vue l’action anthropique qu’elle connaît ce dernier temps, toutes ces espèces endémiques risquent de disparaître de cette île. Il est alors urgent qu’une attention particulière, soit tournée vers elle. Des actions de conservation et d’exploitation durable doivent être amorcées en vue contrecarrer l’extinction des celles-ci. Mots clés :RDCongo, flore, écologie, phytosociologie, insularité, inondation, répartition spatiale, biodiversité, synusie, île Mbiye, Kisangani. / Doctorat en Sciences / info:eu-repo/semantics/nonPublished

I. An original methodological discussion is proposed on the problem of the typology of tropical rain forest’s plant communities, based on the study of forest types across gradients of continentality and elevation, within Atlantic central Africa. These investigations were based on the statement that the main problems in forest typology are related to the non-zonal or zonal character of the different vegetation types and to non considering the relations and differences between forest strata. II. Field data consisted in phytosociological homogeneous sample plots localized within different recognized phytogeographical entities, in a region of tropical Africa where these entities are known to be well conserved. A total of 37 such plots were inventoried in the region extending from the littoral forests of Ndoté, Equatorial Guinea, which are wet evergreen forests, to the continental forests of the Dja, Cameroon, known as evergreen seasonal forests. The studied region also included the oriental Atlantic forests of Equatorial Guinea, known as moist evergreen forests or caesalp forests. In various parts of this continentality gradient, some plots were localized within climax non-zonal formations, namely the submontane rain forests. The emphasis was put on the vegetation of the Monte Alén National Park. The sampling methodology was willing to be as "complete ", including all strata, "quantitative ", enumerating all individuals, and "representative ", within each stratum, as possible. These multi-layers plots were realised using nested sub-plots, with a sampling size of 100 individuals for every ligneous stratum recognized (dominant trees, dominated trees and shrubs) and a sampling size of 200m² for the herbaceous and suffrutex stratum. Forest types were defined independently for each stratum and the differences were analysed. A method was proposed for the simultaneous analysis of all floristic data, converting and standardizing the values from ligneous strata, on the one hand, and from understorey strata, on the other hand. III. Ten forest types were described using IndVal and discussed in the general context of the guineo-congolian region, from a syntaxonomic view point (agglomerative classification) and from a phytogeographical view point (divisive classification). Homologies between these two approaches are described. The proposed phytogeographical system is based on an "open " conception of hierarchical classifications, combining advantages of agglomerative and divisive classifications. In concrete terms, the non-zonal criteria, for example the submontane variants, are categorised separately and in analogy with the zonal criteria, related to the usual phytochoria. Analysis of ecological relationships for the 10 communities showed that the main variables related to the floristic variability in our mainland rain forests are elevation, rainfall, hygrometry (estimated using bryophytes cover levels) and distance to the ocean. The two extremes on the vertical microclimatic gradient, dominant trees stratum and herbaceous stratum, give similar typologies, however canonical analysis showed that for the herbaceous layer, non-zonal variables (hygrometry and elevation) were gaining more importance when the influence of the two zonal variables was attenuated. In every case, spatial autocorrelation was less important than the environment in explaining floristic variability but its role increased in the spatial arrangement of understorey species, whose dispersal capacity is generally lower than canopy trees. The phytosociological, phytogeographical and ecological description of forest types is accompanied by a physiognomical description using biological types spectrum, as well as architectural models, leaf sizes, etc. With regard to diversity, we have demonstrated that species richness was higher from upper to lower strata because of the accumulation in lower strata of species from various strata. On the other hand, the proper stratum diversity, i.e. the structural set, decreased from dominant trees to shrubs. The proper diversity of the herb layer showed relatively high figures mainly due to the higher individual density in relation to the existence of microstrata. Within the 37 sample plots, 1,050 taxa have been identified to species or morpho-species levels, for a total of 25,750 individuals. These taxa represent 442 genus among 104 families. The richest forest type is found on the foothills of the Niefang range, on the windward side. This forest type is also characterised by a high number of oligotypic genus and by species belonging to functional types indicators of glacial refuges. These functional types are defined on the basis of the dispersal capacity and on kind of stand needed for effective germination. We formulated the hypothesis that this kind of "foothills refuge ", characterised by his zonal nature, could have been one of the rare refuges for species from mainland rain forests, while montane and fluvial refuges would mainly have preserved species from non-zonal forest types: (sub)montane and riverine. Based on indicator species of submontane forests, a potential distribution map of this forest type has been realised at the Atlantic central African scale. More than 400 submontane forest localities have been mapped. These forests begin at 400m of altitude near the ocean, and progressively at higher altitude for increasing distance to the ocean. Many lowland localities also comprised submontane species, which could indicate the existence of ecological transgressions. These transgressions would allow migratory tracks for submontane species between isolated mountain ranges, not only during glacial periods, through heights at the northern and southern borders of the congo basin, but also contemporarily through the lowland riverine forest network, in the centre of this basin. Finally, a special attention has been attributed to littoral forests and to some cases of choroecological transgressions, coupled to the ecological equalization phenomenon.

I. An original methodological discussion is proposed on the problem of the typology of tropical rain forest’s plant communities, based on the study of forest types across gradients of continentality and elevation, within Atlantic central Africa. These investigations were based on the statement that the main problems in forest typology are related to the non-zonal or zonal character of the different vegetation types and to non considering the relations and differences between forest strata.<p><p>II. Field data consisted in phytosociological homogeneous sample plots localized within different recognized phytogeographical entities, in a region of tropical Africa where these entities are known to be well conserved. A total of 37 such plots were inventoried in the region extending from the littoral forests of Ndoté, Equatorial Guinea, which are wet evergreen forests, to the continental forests of the Dja, Cameroon, known as evergreen seasonal forests. The studied region also included the oriental Atlantic forests of Equatorial Guinea, known as moist evergreen forests or caesalp forests. In various parts of this continentality gradient, some plots were localized within climax non-zonal formations, namely the submontane rain forests. The emphasis was put on the vegetation of the Monte Alén National Park.<p><p>The sampling methodology was willing to be as "complete ", including all strata, "quantitative ", enumerating all individuals, and "representative ", within each stratum, as possible. These multi-layers plots were realised using nested sub-plots, with a sampling size of 100 individuals for every ligneous stratum recognized (dominant trees, dominated trees and shrubs) and a sampling size of 200m² for the herbaceous and suffrutex stratum.<p><p>Forest types were defined independently for each stratum and the differences were analysed. A method was proposed for the simultaneous analysis of all floristic data, converting and standardizing the values from ligneous strata, on the one hand, and from understorey strata, on the other hand.<p><p>III. Ten forest types were described using IndVal and discussed in the general context of the guineo-congolian region, from a syntaxonomic view point (agglomerative classification) and from a phytogeographical view point (divisive classification). Homologies between these two approaches are described. The proposed phytogeographical system is based on an "open " conception of hierarchical classifications, combining advantages of agglomerative and divisive classifications. In concrete terms, the non-zonal criteria, for example the submontane variants, are categorised separately and in analogy with the zonal criteria, related to the usual phytochoria.<p><p>Analysis of ecological relationships for the 10 communities showed that the main variables related to the floristic variability in our mainland rain forests are elevation, rainfall, hygrometry (estimated using bryophytes cover levels) and distance to the ocean. The two extremes on the vertical microclimatic gradient, dominant trees stratum and herbaceous stratum, give similar typologies, however canonical analysis showed that for the herbaceous layer, non-zonal variables (hygrometry and elevation) were gaining more importance when the influence of the two zonal variables was attenuated. In every case, spatial autocorrelation was less important than the environment in explaining floristic variability but its role increased in the spatial arrangement of understorey species, whose dispersal capacity is generally lower than canopy trees. The phytosociological, phytogeographical and ecological description of forest types is accompanied by a physiognomical description using biological types spectrum, as well as architectural models, leaf sizes, etc.<p><p>With regard to diversity, we have demonstrated that species richness was higher from upper to lower strata because of the accumulation in lower strata of species from various strata. On the other hand, the proper stratum diversity, i.e. the structural set, decreased from dominant trees to shrubs. The proper diversity of the herb layer showed relatively high figures mainly due to the higher individual density in relation to the existence of microstrata. Within the 37 sample plots, 1,050 taxa have been identified to species or morpho-species levels, for a total of 25,750 individuals. These taxa represent 442 genus among 104 families. The richest forest type is found on the foothills of the Niefang range, on the windward side. This forest type is also characterised by a high number of oligotypic genus and by species belonging to functional types indicators of glacial refuges. These functional types are defined on the basis of the dispersal capacity and on kind of stand needed for effective germination. We formulated the hypothesis that this kind of "foothills refuge ", characterised by his zonal nature, could have been one of the rare refuges for species from mainland rain forests, while montane and fluvial refuges would mainly have preserved species from non-zonal forest types: (sub)montane and riverine.<p><p>Based on indicator species of submontane forests, a potential distribution map of this forest type has been realised at the Atlantic central African scale. More than 400 submontane forest localities have been mapped. These forests begin at 400m of altitude near the ocean, and progressively at higher altitude for increasing distance to the ocean. Many lowland localities also comprised submontane species, which could indicate the existence of ecological transgressions. These transgressions would allow migratory tracks for submontane species between isolated mountain ranges, not only during glacial periods, through heights at the northern and southern borders of the congo basin, but also contemporarily through the lowland riverine forest network, in the centre of this basin. Finally, a special attention has been attributed to littoral forests and to some cases of choroecological transgressions, coupled to the ecological equalization phenomenon.<p> / Doctorat en sciences agronomiques et ingénierie biologique / info:eu-repo/semantics/nonPublished