Éditeur(s) : HAL CCSD Elsevier
Résumé : Vegetation phenology is the study of the timing of seasonal events that are considered to be the result of adaptive responses to climate variations on short and long time scales. In the field of remote sensing of vegetation phenology, phenological metrics are derived from time series of optical data. For that purpose, considerable effort has been specifically focused on developing noise reduction and cloud-contaminated data removal techniques to improve the quality of remotely-sensed time series. Comparative studies between time series composed of satellite data acquired under clear and cloudy conditions and from radiometric data obtained with high accuracy from ground-based measurements constitute a direct and effective way to assess the operational use and limitations of remote sensing for predicting the main plant phenological events. In the present paper, we sought to explicitly evaluate the potential use of MODerate resolution Imaging Spectroradiometer (MODIS) remote sensing data for monitoring the seasonal dynamics of different types of vegetation cover that are representative of the major terrestrial biomes, including temperate deciduous forests, evergreen forests, African savannah, and crops. After cloud screening and filtering, we compared the temporal patterns and phenological metrics derived from in situ NDVI time series and from MODIS daily and 16-composite products. We also evaluated the effects of residual noise and the influence of data gaps in MODIS NDVI time series on the identification of the most relevant metrics for vegetation phenology monitoring. The results show that the inflexion points of a model fitted to a MODIS NDVI time series allow accurate estimates of the onset of greenness in the spring and the onset of yellowing in the autumn in deciduous forests (RMSE <= one week). Phenological metrics identical to those provided with the MODIS Global Vegetation Phenology product (MDC12Q2) are less robust to data gaps, and they can be subject to large biases of approximately two weeks or more during the autumn phenological transitions. In the evergreen forests, in situ NDVI time series describe the phenology with high fidelity despite small temporal changes in the canopy foliage. However, MODIS is unable to provide consistent phenological patterns. In crops and savannah, MODIS NDVI time series reproduce the general temporal patterns of phenology, but significant discrepancies appear between MODIS and ground-based NDVI time series during very localized periods of time depending on the weather conditions and spatial heterogeneity within the MODIS pixel. In the rainforest, the temporal pattern exhibited by a MODIS 16-day composite NDVI time series is more likely due to a pattern of noise in the NDVI data structure according to both rainy and dry seasons rather than to phenological changes. More investigations are needed, but in all cases, this result leads us to conclude that MODIS time series in tropical rainforests should be interpreted with great caution. (C) 2013 Elsevier Inc. All rights reserved.
ISSN: 0034-4257

hal-01032407
https://hal.archives-ouvertes.fr/hal-01032407
DOI : 10.1016/j.rse.2013.01.010

Éditeur(s) : HAL CCSD MDPI
Résumé : International audience
Canopy height is a key variable in tropical forest functioning and for regional carbon inventories. We investigate the spatial structure of the canopy height of a tropical forest, its relationship with environmental physical covariates, and the implication for tropical forest height variation mapping. Making use of high-resolution maps of LiDAR-derived Digital Canopy Model (DCM) and environmental covariates from a Digital Elevation Model (DEM) acquired over 30,000 ha of tropical forest in French Guiana, we first show that forest canopy height is spatially correlated up to 2500 m. Forest canopy height is significantly associated with environmental variables, but the degree of correlation varies strongly with pixel resolution. On the whole, bottomland forests generally have lower canopy heights than hillslope or hilltop forests. However, this global picture is very noisy at local scale likely because of the endogenous gap-phase forest dynamic processes. Forest canopy height has been predictively mapped across a pixel resolution going from 6 m to 384 m mimicking a low resolution case of 3 points·km −2. Results of canopy height mapping indicated that the error for spatial model with environment effects decrease from 8.7 m to 0.91 m, depending of the pixel resolution. Results suggest that, outside the calibration plots, the contribution of environment in shaping the global canopy height distribution is quite limited. This prevents accurate canopy height mapping based only on environmental information, and suggests that precise canopy height maps, for local management purposes, can only be obtained with direct LiDAR monitoring.
ISSN: 2072-4292

hal-01590726
https://hal-univ-guyane.archives-ouvertes.fr/hal-01590726
https://hal-univ-guyane.archives-ouvertes.fr/hal-01590726/document
https://hal-univ-guyane.archives-ouvertes.fr/hal-01590726/file/remotesensing-08-00747.pdf
DOI : 10.3390/rs8090747

Éditeur(s) : HAL CCSD
Résumé :   Diplôme : DEA
il s'agit d'un type de produit dont les métadonnées ne correspondent pas aux métadonnées attendues dans les autres types de produit : DISSERTATION
Tropical rainforests are complex ecosystems, rarely investigated from ground-level. The canopy, highest part of the cover, can on the other hand be studied by several remote sensing tools. The aim of this study is to assess in what way the canopy is related to the forest structure below and the forest dynamic. We looked for correlation between the canopy texture (crown size) observed from aerial photographs and various parameters of the forest (spatial and demographic structure, forest dynamic, floristic structure). We worked on the Paracou experimental device, in French Guiana, at two different scales. 1- At a local scale (6.5 ha ; scale of the photographs 1: 4 000), we demarcated six sub-plots with big crowns in the canopy and six with small crowns. 2- At a forest scale (approximately one hundred ha ; scale of the photograph 1: 25 000), we compared 2 forest plots: one with a rough canopy texture and one with a smooth canopy texture. We found that the size of the crowns in the canopy do not express the same characteristics of the fore according to the scale of study : at the local scale, difference in crown size means a difference in basal area but not in density, while at the forest scale it is the opposite. Several hypotheses related to the sylvigenetic cycle, the floristic structure or the nature of the substratum are suggested to explain these results.
Les forêts tropicales humides (FTH) sont des écosystèmes complexes difficiles à étudier depuis le sol. Leur canopée, partie supérieure du couvert, est en revanche abordable par divers outils de télédétection. Cette étude vise à déterminer en quoi la structure de la canopée est le reflet de la structure forestière sous-jacente et de la dynamique spatio-temporelle qui s'y exerce. Nous avons donc recherché des corrélations, entre la texture de canopée (taille des couronnes) vue sur des photographies aériennes, et différents paramètres de la forêt sous-jacente (structure spatiale et démographique, dynamique forestière, structure floristique). Ce travail a été abordé sur le dispositif expérimental forestier de Paracou, en Guyane Française, à 2 échelles différentes. 1- A l'échelle de la parcelle (6,5 ha ; photographies au 1: 4 000), nous avons délimité des placettes à grosses couronnes en canopée et des placettes à petites couronnes. 2- A l'échelle du massif forestier (une centaine d'ha ; photographies IGN au 1: 25 000) nous comparons une parcelle de texture de canopée, grossière et une parcelle de texture fine. Le résultat principal de notre étude est que la taille des couronnes en canopée ne reflète pas les mêmes caractéristiques du couvert selon l'échelle de l’étude : à l'échelle de la parcelle, des différences de taille de couronnes traduisent des différences de surface terrière mais pas de densité, tandis qu'à l'échelle du massif forestier c'est l'inverse. Plusieurs hypothèses, invoquant des différences dans le stade sylvigénétique, la composition floristique ou la nature du substrat des placettes ou parcelles comparées, sont proposées pour expliquer ces résultats.
https://hal.archives-ouvertes.fr/hal-01189180

hal-01189180
https://hal.archives-ouvertes.fr/hal-01189180
PRODINRA : 12184

Éditeur(s) : HAL CCSD Montpellier : CIRAD
Résumé : This article presents the effects of logging on the dynamics of above-ground biomass from the results of the post-logging study within two forests: Cikel in Eastern Para, Brazil and Paracou in French Guiana. The main objective is to compare the impact of commercial logging on the regeneration of the above-ground biomass in these forests whose characteristics differ in terms of structure and growth. In both sites, the intensity of exploitation is a key factor in determining the loss of biomass and the time required for its regeneration. In Paracou, the regeneration of biomass lost during conventional logging of 10 trees per hectare takes 45 years and more than 100 years when operating with higher intensity (21 trees/ha). In Cikei the forest biomass regenerates after 49 years harvesting 6 trees/ha and that takes 87 years after removal of 8 trees/ha. This regeneration needs similar time on both sites but with lower logging intensity at Cikel, in which felled trees are larger with a greater biomass than those of Paracou. This post-logging study has established a direct correlation of the dynamics of the biomass with the initial structure of the forest, as well as with the parameters of forest dynamics: mortality, growth and recruitment. The accumulation of biomass by the tree growth of the two remaining stands is a key parameter for the net carbon storage, while the contribution of recruitment in Paracou becomes significant only after 10 years after felling. Therefore in view to improve the growth of residual trees, it is compulsory to apply adequate silvicultural treatments such as selective thinning or removal of vines. While the two forests are geographically close enough, their regenerative abilities differ and because of the significant difference in size of the trees, the forest could tolerate more intensive harvesting in French Guiana
ISSN: 0006-579X

hal-01032427
https://hal.archives-ouvertes.fr/hal-01032427

Résumé : L'importance des forêts tropicales dans le cycle du carbone à l'échelle planétaire est majeure, tant en terme de stock qu’en terme de flux de CO2. Plusieurs études mettent en évidence des changements au sein des forêts tropicales au cours des 20 dernières années, notamment des changements de la dynamique forestière et une augmentation de la biomasse aérienne. Les déterminants de ces variations sont aujourd’hui discutés et nous proposons ici d’apporter une contribution à ce débat. Les données utilisées dans ce travail proviennent du dispositif de suivi forestier à long terme de Paracou, Guyane Française, mis en place en 1984 et qui recouvre plus de 120 ha de forêt tropicale humide. Les données météorologiques proviennent de la tour à flux du dispositif, Guyaflux. Les mesures de diamètre proviennent de la base Guyafor pour les données annuelles et bisannuelles, et des mesures de 260 arbres à proximité de la tour à flux pour les mesures diamétriques intra-annuelles.Cette thèse se divise en deux grandes parties. La première concerne l’analyse de la biomasse du dispositif de suivi forestier de Paracou en Guyane et l’implication des changements de structure de cette forêt sur le bilan de carbone. Cette partie est constituée de deux points. (i) Quelles échelles temporelles et spatiales sont pertinentes pour analyser les composantes de structure (biomasse, aire basale et nombre de tiges) et de dynamique (croissance, recrutement et mortalité) des forêts tropicales afin de minimiser les effets d’échantillonnage ? Nous avons établi une méthodologie permettant de relier les intervalles de temps et la surface de mesure aux coefficients de variation de chacune des variables de structure et de dynamique de la forêt. (ii) Quels processus démographiques sont prépondérants dans l’explication des variations de biomasse et comment se redistribue la biomasse accumulée dans le système ? L’augmentation de la biomasse observée sur le dispositif de Paracou serait liée à la rareté des évènements de mortalité des gros arbres qui portent une part très importante de la biomasse.La deuxième partie de la thèse concerne l’analyse de l’effet du climat à moyen terme, inter et intra-annuel, et les effets directs du climat dans les changements de dynamique de la forêt. Cette partie se divise en 3 points. (i) Comment quantifier le stress hydrique en forêt tropicale humide ? Nous avons réalisé un modèle journalier de réserve en eau du sol pour les arbres en forêt tropicale. (ii) Quelles variables sont explicatives de la croissance des arbres en forêt guyanaise ? Nous avons montré que l’eau dans le sol est le facteur le plus explicatif du déterminisme climatique parmi un panel de variables climatiques. (iii) Quels traits fonctionnels sont prédicteurs de la réponse des arbres aux variations climatique ? Nous avons déterminé que des traits spécifiques, densité du bois et la hauteur maximale, ainsi que le diamètre de l'arbre au moment de la mesure, modulent la croissance des arbres en réponse au climat.
At a global scale tropical forest play a major role in term of carbon stock as well as in term of CO2 fluxes. Several studies have highlighted changes in tropical forest functioning during the last 20 years including a faster turnover and an increase of above ground biomass. The drivers of these changes are discussed and throughout this thesis we propose to contribute to this debate. We use the data from the Paracou experimental site in French Guiana established in 1984 on 120 hectares of moist tropical forest. Meteorological data come from the flux tower of the site, Guyaflux. We use annual and bisannual diameter measurements from the Guyafor database, and intra-annual diameter increments from the measurements of 260 trees near the flux tower.This thesis has two main parts. In the first part we present the biomass analysis of the Paracou permanent plots and the impact of structural changes in this forest on the carbon budget. The first part is constituted by two points. (i) Which temporal and spatial scale used to analyze the structure (biomass, basal area and stem density) and dynamics (tree growth, recruitment and mortality)components of tropical forest in order to minimized sample bias ? We establish a simple method to rely measurement interval between census and surface of measurement to the coefficient of variation of forests structure and dynamic components (ii) Which demographic process are involved in the explanation of biomass variation and how the biomass is distributed in the system ? The observed increase of biomass at Paracou could be link to the rarity of big trees mortality events. These big trees represent the larger part of the biomass. In the second part, we present the analysis of intra and inter-annual climate variation effects on forest dynamic changes. This part is divided in two points. (i) How to model drought stress in moist tropical forest ? We built a daily water balance model for tropical trees. (ii) Which climate variables explain the tree growth in guianian forests? We shown that soil water availability is the determinant factor of tree growth among a panel of climate variables. (iii) Which functional traits are involved in the tropical tree growth responses to climate? In this analysis, we determined that wood specific gravity, maximum tree height and tree diameter modulate the tree growth response to climate variations.

http://www.theses.fr/2011AGUY0481/document

Éditeur(s) : HAL CCSD
Résumé :   Diffusion du document : INRA Antilles-Guyane, SDAR/ERIST, Domaine Duclos, F97170 Petitt-Bourg (Guadeloupe) Diplôme : DEA
il s'agit d'un type de produit dont les métadonnées ne correspondent pas aux métadonnées attendues dans les autres types de produit : DISSERTATION
ln French Guiana, the mangrove covers almost ail the coastline. It results original because of a dynamic influenced by the system of the amazon muds' dispersion. The coastline's variations are important and the mangrove growing there constitutes various facies. To understand the structure and dynamic of this forest, a test-site has been selected at the Kaw river's estuary. Field measurements of structure parameters permited to identify 3 major types of mangrove forests with different developement stages. Relations between the measured parameters show the facies'dynamic with a self-thinning phnomenon then a gap regeneration. Correlations between structure parameters and corresponding spectral parameters extracted from an IKONOS image have been tested. The use of texture indexes are discussed. Then, a classification of the image reveals the spatial distribution of the facies Parallely, the constitution of a temporal series from 1950 to 2001, base on various remote sensed datas, shows the evolution of the coastline and the associated mangrove. During this period, succeed accretion phases (1950-1981; 1991-2001) and erosion phase (1981-1991). A theorical ages map of the forest stands are also established. Relating with the facies map, it explains the complex evolution of limited Rhizophora zones. A comparison with Sinnamary test-site is made and futurs researches considered.
En Guyane française, la mangrove couvre l'essentiel du littoral. Elle tire son originalité d'une dynamique influencée par le système de dispersion des boues amazoniennes. Les variations de la ligne de côte sont alors importantes et les mangroves s'y développant constituent des faciès variés. Pour mieux comprendre la structure et la dynamique de ces mangroves, un site atelier a été choisi au niveau de l'estuaire de la rivière de Kaw. Les mesures sur le terrain de paramètres de structure des peuplements permettent d'identifier 3 grands types de forêt avec différents stades de développement. Des relations entre les différents paramètres mesurés mettent en évidence la dynamique de ces faciès, basée sur un phénomène d'auto-éclaircie puis une régénération par chablis. Un essai de corrélation entre paramètres de structure et paramètres spectraux correspondant sur une image IKONOS a été faite. L'apport des indices de texture est discuté. Une classification de l'image montre la distribution spatiale des faciès. Parallèlement, la reconstitution d'une série temporelle de 1950 à 2001 à partir de données de télédétection variées permet de suivre l'évolution de la côte et de la mangrove associée. Pendant cette période se succèdent phases d'accrétion (1950-1981 ; 1991-2001) et phase d'érosion (1981-1991). Une carte des âges théoriques des peuplements est également établie. Mise en relation avec la carte des faciès, elle explique notamment l'évolution complexe des zones limitées de mangrove à Rhizophora. Une comparaison avec le site de Sinnamary est faite et de futures recherches envisagées.
https://hal.archives-ouvertes.fr/hal-01189238

hal-01189238
https://hal.archives-ouvertes.fr/hal-01189238
PRODINRA : 18075

Éditeur(s) : HAL CCSD
Résumé : Diffusion du document : INRA Documentation de Centre, Domaine de Duclos, 97170 Petitt-Bourg (Guadeloupe, France) Diplôme : DEA
il s'agit d'un type de produit dont les métadonnées ne correspondent pas aux métadonnées attendues dans les autres types de produit : DISSERTATION
Basralocus (Dicorynia guianensis Amshoff, Caesalpiniaceae), first timber exploited species in French Guyana, shows a spatial aggregative distribution. In order to characterize Basralocus clump organisation, many approaches were used : the study of different forest variables (total height, diameter at breast height, .... j, a spatial analysis studied by population splitting in 3 classes according to their diameters at breast height and an architectural analysis. To evaluate the pertinence of different approach, 2 clump structures, located on control and logged plots,are compared. The trees of the logged clump have a smaller height and diameter. a slender habit, a taller crown and a better growth than the control. These 2 clumps stand out by a different number of mature trees and different types of spatial distribution of juvenile trees around them are possible. The opening of the environment by forestry provokes a Jump rejuvenation by the presence of numerous unbranched young trees. The architectural analysis helps to detect a large expression variability in growth strategies which can be reflected by different architectural vigours. The combination between forest variables, spatial and architectural analysis results could suggest a dynamic evolution of these 2 clumps.
L'Angélique (Dicorynia guianensis Amshoff, Caesalpiniaceae), première espèce exploitée Pour le bois d'oeuvre en Guyane française, a une répartition spatiale de type agrégatif. Afin de caractériser la structure d’un agrégat d'Angélique, plusieurs approches sont utilisées: une étude des différents paramètres dendrométriques (hauteur totale, diamètre à 1.30 m....j. une analyse spatiale réalisée en découpant la population des individus en 3 classes de diamètre bien distinctes et une analyse architecturale. Afin de valider la pertinence de ces approches, 2 agrégats, l'un en parcelle témoin et l'autre en parcelle exploitée pour le bois d'oeuvre, sont comparés. Ainsi, les arbres de l'agrégat exploité ont une hauteur et un diamètre plus petit, un port plus élancé, un houppier plus grand et une meilleure croissance. Ces 2 agrégats se distinguent par un nombre variable de semenciers et différents types de répartition spatiale des juvéniles autour de ceux-ci sont possibles. L'ouverture du milieu par l'exploitation forestière provoque un rajeunissement de l'agrégat souligné par la présence de nombreux individus jeunes non ramifiés. L'analyse architecturale permet de déceler une grande variabilité d'expression dans le développement qui se traduit par différentes vigueurs architecturales. En combinant les résultats des analyses dendrométrique, spatiale et architecturale, une évolution dynamique de la structure de ces 2 agrégats est proposée.
https://hal.archives-ouvertes.fr/hal-01189192

hal-01189192
https://hal.archives-ouvertes.fr/hal-01189192
PRODINRA : 13161

Éditeur(s) : HAL CCSD Public Library of Science
Résumé : We examined tree-soil habitat associations in lowland forest communities at Paracou, French Guiana. We analyzed a large dataset assembling six permanent plots totaling 37.5 ha, in which extensive LIDAR-derived topographical data and soil chemical and physical data have been integrated with precise botanical determinations. Map of relative elevation from the nearest stream summarized both soil fertility and hydromorphic characteristics, with seasonally inundated bottomlands having higher soil phosphate content and base saturation, and plateaus having higher soil carbon, nitrogen and aluminum contents. We employed a statistical test of correlations between tree species density and environmental maps, by generating Monte Carlo simulations of random raster images that preserve autocorrelation of the original maps. Nearly three fourths of the 94 taxa with at least one stem per ha showed a significant correlation between tree density and relative elevation, revealing contrasted species-habitat associations in term of abundance, with seasonally inundated bottomlands (24.5% of species) and well-drained plateaus (48.9% of species). We also observed species preferences for environments with or without steep slopes (13.8% and 10.6%, respectively). We observed that closely-related species were frequently associated with different soil habitats in this region (70% of the 14 genera with congeneric species that have a significant association test) suggesting species-habitat associations have arisen multiple times in this tree community. We also tested if species with similar habitat preferences shared functional strategies. We found that seasonally inundated forest specialists tended to have smaller stature (maximum diameter) than species found on plateaus. Our results underline the importance of tree-soil habitat associations in structuring diverse communities at fine spatial scales and suggest that additional studies are needed to disentangle community assembly mechanisms related to dispersal limitation, biotic interactions and environmental filtering from species-habitat associations. Moreover, they provide a framework to generalize across tropical forest sites.
ISSN: 1932-6203

hal-01260289
https://hal.archives-ouvertes.fr/hal-01260289
PRODINRA : 342930
DOI : 10.1371/journal.pone.0141488

Éditeur(s) : HAL CCSD
Résumé :  
For a comprehensive evaluation of human influences on the genetic system, the simulation model EGO-GENE was developed. The model combines population genetic and population dynamic processes with forest growth models. The dynamics of genetic structures (single locus and multilocus genotypes up to 200 loci) can be simulated. Spatial and temporal genetic dynamics are included in the model. Overlapping or separated generations can be created and different modes of mating systems can be implemented. The model can be run with empirical and fictitious input data. The simulated genetic structures can be directly analysed with established genetic analysis programmes. An example demonstrates the use of the model far validation of interactions among components of the mating system. Another example demonstrates the applicability of the model in decision-making for seed sampling strategies. The genetic differences between the adult trees and the harvested seeds and the danger of genetic drift regarding to different seed sampling strategies are investigated by use of the genetic distance and the reproductively effective population size.
Silvae genetica

hal-01032083
https://hal.archives-ouvertes.fr/hal-01032083

Éditeur(s) : HAL CCSD Nature Publishing Group
Résumé : Tropical forests store large amounts of biomass despite they generally grow in nutrient-poor soils, suggesting that the role of soil characteristics in the structure and dynamics of tropical forests is complex. We used data for >34 000 trees from several permanent plots in French Guiana to investigate if soil characteristics could predict the structure (tree diameter, density and aboveground biomass), and dynamics (growth, mortality, aboveground wood productivity) of nutrient-poor tropical forests. Most variables did not covary with site-level changes in soil nutrient content, indicating that nutrient-cycling mechanisms other than the direct absorption from soil (e.g. the nutrient uptake from litter, the resorption, or the storage of nutrients in the biomass), may strongly control forest structure and dynamics. Ecosystem-level adaptations to low soil nutrient availability and long-term low levels of disturbance may help to account for the lower productivity and higher accumulation of biomass in nutrient-poor forests compared to nutrient-richer forests.
ISSN: 2045-2322

hal-01606170
https://hal.archives-ouvertes.fr/hal-01606170
DOI : 10.1038/srep45017
PRODINRA : 389876

Éditeur(s) : HAL CCSD Springer Verlag
Résumé : Geomorphic landscape features have been suggested as indicators of forest diversity. However, their explanatory power has not yet been explicitly tested at a regional scale in tropical rainforest. We used forest inventories conducted according to a stratified sampling design (3,132 plots in 111 transects at 33 sites) and holistic multi-scale geomorphological mapping derived from a Shuttle Radar Topography Mission digital elevation model to describe and explain spatial patterns in floristic composition across French Guiana (80,000 km2). We measured and identified 123,906 trees with DBH C20 cm and used constrained and unconstrained ordinations to analyze variations in the abundance of 221 taxa and 51 families. Variance partitioning and variograms were used to detect spatial patterns in species composition, compare the explanatory power of spatial and environmentalfactors, and select the variables that best explain forest composition. Strong floristic patterns corresponded to a major latitudinal gradient and significant sub-regional floristic structure. Geomorphological landscapes shaped by historic climate fluctuations and major geological events successfully captured these patterns and explained the variation in abundance of 80 taxa, corresponding to 65 % of the inventoried trees. Our findings suggest that long-term forest dynamics are under substantial ‘‘geomorphographic control’’. A geomorphological perspective on landscapes that incorporates current and past environmental filters and historical biogeographical processes could thus be used more systematically in tropical regions for regional planning and forest conservation
ISSN: 0960-3115

hal-01204224
https://hal.archives-ouvertes.fr/hal-01204224
DOI : 10.1007/s10531-014-0854-8
PRODINRA : 295767

Éditeur(s) : HAL CCSD Wiley
Résumé : 1. Resource control over abundance, structure and functional diversity of soil microbial communities is a key determinant of soil processes and related ecosystem functioning. Copiotrophic organisms tend to be found in environments which are rich in nutrients, particularly carbon, in contrast to oligotrophs, which survive in much lower carbon concentrations. We hypothesized that microbial biomass, activity and community structure in nutrient-poor soils of an Amazonian rain forest are limited by multiple elements in interaction. We tested this hypothesis with a fertilization experiment by adding C (as cellulose), N (as urea) and P (as phosphate) in all possible combinations to a total of 40 plots of an undisturbed tropical forest in French Guiana. After 2years of fertilization, we measured a 47% higher biomass, a 21% increase in substrate-induced respiration rate and a 5-fold higher rate of decomposition of cellulose paper discs of soil microbial communities that grew in P-fertilized plots compared to plots without P fertilization. These responses were amplified with a simultaneous C fertilization suggesting P and C colimitation of soil micro-organisms at our study site. Moreover, P fertilization modified microbial community structure (PLFAs) to a more copiotrophic bacterial community indicated by a significant decrease in the Gram-positive:Gram-negative ratio. The Fungi:Bacteria ratio increased in N fertilized plots, suggesting that fungi are relatively more limited by N than bacteria. Changes in microbial community structure did not affect rates of general processes such as glucose mineralization and cellulose paper decomposition. In contrast, community level physiological profiles under P fertilization combined with either C or N fertilization or both differed strongly from all other treatments, indicating functionally different microbial communities. While P appears to be the most critical from the three major elements we manipulated, the strongest effects were observed in combination with either supplementary C or N addition in support of multiple element control on soil microbial functioning and community structure. We conclude that the soil microbial community in the studied tropical rain forest and the processes it drives is finely tuned by the relative availability in C, N and P. Any shifts in the relative abundance of these key elements may affect spatial and temporal heterogeneity in microbial community structure, their associated functions and the dynamics of C and nutrients in tropical ecosystems.
ISSN: 0269-8463

hal-01204229
https://hal.archives-ouvertes.fr/hal-01204229
DOI : 10.1111/1365-2435.12329
PRODINRA : 311389

Éditeur(s) : HAL CCSD Wiley
Résumé : International audience
Questions: Several studies have described aggregated spatial patterns in tropical tree species. This study investigates aggregate dynamics of Vouacapoua americana (Aublet), a climax species whose spatial pattern is not simply related to light and soil conditions or to its short seed dispersal range. Location: Two rain forest sites: Nouragues and Paracou, in the Guiana Shield. Methods: We described the spatial pattern of tree locations and spatial autocorrelation of tree diameters, using statistics derived from Ripley's K. We particularly used methods to define analysis subplots according to local density or local mean diameter. We investigated relationships between spatial distributions of adults and saplings. Results: At both sites, populations of Vouacapoua demonstrated several nested levels of aggregation. Tree diameters were spatially autocorrelated, revealing the existence of clusters with similar diameters. In the largest aggregates, tree diameters declined from the centre to the edge. Regeneration was aggregated and occurred mainly at cluster edges and around rare isolated trees, and sapling densities and basal area of adults were negatively correlated. We show that long-distance dispersal events are rare. Conclusions: Environmental factors and seed dispersal only explain part of the observed spatial patterns. We provide two main hypotheses about Vouacapoua population dynamics. First, the lack of regeneration in aggregate centres results in the ageing of existing aggregates. We suggest that this lack of recruitment close to mature trees is due to a Janzen-Connell effect. However, aggregates can continue to grow along colonization fronts. Second, long-distance dispersal events allow the formation of new clusters and play a crucial role in the colonization process. We investigate aggregate dynamics of Vouacapoua americana (Aublet) whose spatial pattern is not simply related to environmental conditions or to its seed dispersal. Regeneration does not occur in centers of aggregate of adults. We suggest that rare long-distance dispersal events and density-dependence predation of seeds and seedlings play a crucial role in formation of new clusters and structuration of larger aggregates.
ISSN: 1100-9233

hal-01502642
https://hal-agroparistech.archives-ouvertes.fr/hal-01502642
DOI : 10.1111/jvs.12080

Éditeur(s) : HAL CCSD Springer Verlag/EDP Sciences
Résumé : International audience
Background : This study examines the radial trend in wood stiffness of tropical rainforest trees. The objective was to determine if the type of growing environment (exposed plantation or dense primary forest) would have an effect on this radial trend. Methods : The axial elastic modulus of wood samples, representing a pith to bark cross-section, of six trees from several French Guianese species (two of Eperua falcata, one of Eperua grandiflora, two of Carapa procera and one of Symphonia gloubulifera) was measured using a dynamic "forced vibration" method. Results : Primary forest trees were observed to have a decrease in wood stiffness from pith to bark, whereas plantation trees, from the same genus or species, displayed a corresponding increase in wood stiffness. Juvenile wood stiffness appears to vary depending on the environment in which the tree had grown. Conclusion : We suggest that the growth strategy of primary forest trees is to produce wood resistant to selfbuckling so that the height of the canopy may be obtained with the maximum of efficiency. In contrast, the growth strategy of the trees growing in an exposed plantation is to produce low-stiffness wood, important to provide flexibility in wind. Further experiments to study the behaviour of more species, with more individuals per species, growing across a range of physical environments, are required.
ISSN: 1286-4560

hal-00644335
https://hal.archives-ouvertes.fr/hal-00644335
https://hal.archives-ouvertes.fr/hal-00644335/document
https://hal.archives-ouvertes.fr/hal-00644335/file/McLean_al.pdf
DOI : 10.1007/s13595-011-0085-z

Éditeur(s) : HAL CCSD Elsevier
Résumé :  
The within-population spatial structure of genetic diversity is shaped by demographic processes, including historical accidents such as forest perturbations. Information drawn from the analysis of the spatial distribution of genetic diversity is therefore inherently linked to demographic-historical processes that ultimately determine the fate of populations. All adult trees and saplings in a 1.4-ha plot within a mixed Norway spruce (Picea abies [L.] Karst) stand were characterised by means of chloroplast (paternally inherited) markers, and a large sub-sample of these were genotyped at mitochondrial (maternally inherited) molecular markers. These data were used to analyse the spatial distribution of genetic variation and to compare the patterns corresponding to the two marker types. The plot presented non-homogeneous local stem density in the younger cohorts, and the indirect effect of this source of variation on the spatial genetic structure was investigated. Results suggest that (i) spatially limited seed dispersal induced patchiness in genotype distribution, while pollen flow had a homogenising effect; (ii) deviations from random spatial structure were stronger in the demographically most stable portions of the stand, while they were weaker where sudden bursts of regeneration occurred; (iii) spatially overlapping adult and sapling cohorts displayed the same spatial genetic structure (stronger on stable areas, weaker in portions of the stand undergoing events of intense regeneration), which was substantiated by the influence of local demographic processes. Regeneration dynamics as modulated by demography thus influences the distribution of genetic diversity within the stand both in the younger life stages and in the adult population.
ISSN: 0378-1127

hal-01032063
https://hal.archives-ouvertes.fr/hal-01032063
DOI : 10.1016/j.foreco.2008.03.023

Résumé : La distribution actuelle des espèces d’arbres des forêts tropicales humides amazoniennes est le résultat de processus dynamiques sur lesquels les perturbations passées, d’origine climatique ou anthropique, peuvent avoir jouées un rôle majeur. Des outils génétiques actuels associés à des méthodes d’analyses puissantes permettent d’inférer et de dater l’empreinte génomique des évènements démographiques subits par les populations naturelles. Le but de cette thèse est de contribuer à une vision d’ensemble de l’évolution démographique de la forêt tropicale humide du plateau des Guyanes pendant le Quaternaire. J’ai d’abord étudié les propriétés des microsatellites (répétitions simple de séquences, SSR) et les limites de leurs applications en tant que marqueurs moléculaires en phylogéographie. Plusieurs allèles SSR contenu dans diverses populations de trois genres d’arbres différents ont été séquencés et le polymorphisme, porté par chaque partie des fragments ADN contenant des SSR, a été étudié. J’ai observé que la quantité de variation génétique était aussi importante dans les régions encadrant le microsatellite que dans le SSR lui-même. Ces régions ont contribué significativement à l’information phylogénétique et ont parfois été la principale source de différenciation entre les individus et les populations. Cette expérience m’a permis de choisir la meilleure stratégie à adopter pour traiter les données de variation génétique des SSR dans des modèles démographiques. J’ai ensuite utilisé ces marqueurs SSR et des méthodes ABC pour étudier l’impact local des occupations humaines précolombiennes sur les forêts en Guyane française. Les changements démographiques subis par des populations de quatre espèces d’arbres ont été comparés entre cinq sites occupés et quatre sites non perturbés. Les perturbations naturelles et induites par l’homme ont contribué à la mise en place de la diversité observée actuellement, comme le suggère des signatures d’effet fondateur détectées sur les sites perturbés ou non. Néanmoins, la taille des populations fondatrices semble être proportionnelle à l’impact humain estimé, indiquant que les perturbations anthropique ont induit un remplacement des populations plus important que les processus écologiques naturels. Pour terminer, j’ai mené une étude de démographie comparative d’espèces d’arbres tropicaux à l’échelle régionale de la Guyane française. Cinq scénarios démographiques ont été testés pour neuf populations de huit espèces en utilisant des microsatellites nucléaires et des séquences chloroplastiques. Les changements démographiques les plus récents détectés convergent vers une signature générale d’expansion de la forêt humide depuis probablement la dernière glaciation. Cela peut suggérer, soit que la composition des forêts du plateau des Guyanes était différente, ou que le couvert forestier était moins étendu. Ce résultat constitue le premier indice de changement passé, probablement d’origine climatique, subi par les communautés forestières du plateau des Guyanes ; qui n’avait jusque-là pas pu être testé avec des méthodes standards en raison d’un manque de fossiles. En conclusion, j’ai recueilli des preuves génétiques que les perturbations, à différents échelles de temps, ont marqué la structure et la composition génétique des forêts du plateau des Guyanes. L’application de ces méthodes à d’autres régions Néotropicales pourrait contribuer, à l’avenir, à reconstruire l’histoire écologique de la forêt amazonienne de façon détaillée.
The current tree species distribution in Amazonian tropical rainforests is the outcome of dynamic processes in which past perturbations, climatic or anthropogenic, may have played a major role. Currently available genetic tools associated with powerful analytical methods allow inferring and dating the genomic imprint of demographic events undergone by natural populations. The aim of my thesis is to contribute to an overview of the demographic evolution of the tropical rainforest in the Guiana shield during the Quaternary. I have first studied the properties of microsatellites or Simple sequence repeat (SSR) and the limits in their applications as molecular markers in phylogeography. Several SSR alleles in multiple populations of three divergent tree genera have been sequenced and the sources of polymorphisms, carried by each portion of SSR-containing DNA amplicons, have been broken apart. I have observed that the amount of variation was as large in flanking regions as in the SSR itself. The former contributed significantly to the phylogenetic information and sometimes was the main source of differentiation among individuals and populations. This experiment allowed me to choose the best strategy to treat SSR variation in demographic modeling. I have subsequently used SSR markers and ABC methods to study the impact of pre-Columbian human settlements at the local level on rainforests in French Guiana. Demographic changes undergone by populations of four tree species were compared between five settled and four undisturbed sites. Both natural turnover and human-induced disturbances appeared as shaping currently observed genetic diversity, with signatures of founder effect both at disturbed and undisturbed sites. Nevertheless, the size of founding populations appeared to be proportional to the estimated human impact, suggesting that human-made disturbance caused deeper population turnover than background ecological processes. Finally, I have conducted a comparative demography study of tropical tree species at the regional level across French Guiana. Five demographic scenarios have been tested for nine populations from eight species using both nuclear microsatellites and chloroplastic sequences. The most recent demographic signals detected converge on a global signature of expansion of the rainforest since probably the last glaciation. This finding can suggest either that the composition of the Guiana shield rainforests was different, or that its global extent was smaller, during earlier epochs of the Pleistocene than today. This result is the first firm indication of changes undergone by the forest communities of the Guiana shield in the distant climatic past – something which could not be tested with standard methods due to almost complete lack of fossil evidence. In conclusion, I have gathered genetic evidence that disturbances, both in historical and geological times, have marked the structure and composition of forests in the Guiana shield.The extensive application of these methods to other regions of the Neotropics maycontribute, in the future, to the detailed reconstruction of the ecological history ofAmazonian forests.

http://www.theses.fr/2012AGUY0568/document

Éditeur(s) : HAL CCSD Springer Verlag
Résumé : International audience
Heartwood/sapwood profile and the tradeoff between trunk and crown increment in a natural forest: the case study of a tropical tree (Dicorynia guianensis Amsh., Fabaceae) Key message Sapwood area and the radial growth rate of the trunk follow the same pattern at breast height, with an initial increase and subsequent constant value, resulting from the increasing growth allocation toward the crown rather than tree decline. Heartwood area and heartwood volume in the trunk increase more rapidly after this shift occurs. Abstract Sapwood (SW) and heartwood (HW) are two functionally distinct classifications of wood in perennial stems for which quantities can vary greatly in tropical trees. Numerous positive correlations have been found between the radial growth rate (RGR) and SW quantity; however, variations in the SW/HW quantities have not been studied in light of the ontogenetic variation of RGR. Wood core sampling, intensive measurements of tree structure (number of branches, stem volumes), and radial growth monitoring were performed on an abundant and highly exploited tree species in French Guiana (Dicorynia guianensis) to investigate the relationship between RGR, SW/HW quantity, tree structure, and their variations on the course of a tree's ontogeny. SW area and RGR followed the same pattern of variation throughout tree development, both increasing first and reaching a steady state after 50 cm DBH (diameter at breast height). After this value, we observed a strong increase in both the HW area and HW volume increment, concomitant with a more rapid increase in crown volume. The stabilization of RGR for trees with DBH >50 cm was related not to a tree's decline but rather to an increasing wood allocation to the crown, confirming that RGR at breast height is a poor indicator of whole-tree growth for bigger individuals. We also confirmed that HW formation is an ontogenetic process managing SW quantity that is continuously and increasingly produced within the crown as the tree grows. This study highlights the effect of growth-mediated ontogenetic changes on the localization of water and carbohydrate storage within a tree, resulting from SW and HW dynamics throughout tree ontogeny.
ISSN: 0931-1890

hal-01654003
https://hal.archives-ouvertes.fr/hal-01654003
https://hal.archives-ouvertes.fr/hal-01654003/document
https://hal.archives-ouvertes.fr/hal-01654003/file/Lehnebach%20et%20al.%E2%80%94Trees%202016-%20Own%20version.pdf
DOI : 10.1007/s00468-016-1473-7