Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
Because many biological processes related to the dynamics of infectious diseases are caused by complex interactions between the environment, the host(s) and the agent(s), the necessity to address the methodological implications of this inherent complexity has recently emerged in epidemiology. Most epidemiologists now acknowledge that most human infectious diseases are likely to have complex dynamics. However, this knowledge still percolates with difficulty in their statistical "modus operandi". Indeed, for the study of complex systems, the traditional first-line statistical toolbox of epidemiologists (mainly built around the Generalized Linear Model family), despite its undeniable practicality and robustness, has structural limitations deprecating its usefulness. Three major sources of complexity neglected or not taken into account by this first-line statistical toolbox and having deep statistical implications are the multi-level organization of data, the non-linear relationships between variables and the complex interactions between variables. Three promising candidates to incorporate along with traditional tools for a new first-line statistical toolbox more suitable to apprehend these sources of complexity are the generalized linear mixed models, the generalized additive models, and the structural equation models. The aforementioned methodologies have the advantage to be generalizations of GLM models and are relatively easy to implement. Their assimilation and implementation would thus be greatly facilitated for epidemiologists. More globally, this text underlines that an improved use of other methods as such described here compared to traditional ones has to be performed to better understand the complexity challenging epidemiologists every day. This is particularly true in the field of infectious diseases for which major public health challenges will have to be addressed in the coming decades.
ISSN: 1567-1348

inserm-00872354
http://www.hal.inserm.fr/inserm-00872354
http://www.hal.inserm.fr/inserm-00872354/document
http://www.hal.inserm.fr/inserm-00872354/file/main_document.pdf
DOI : 10.1016/j.meegid.2013.09.006

Éditeur(s) : HAL CCSD Public Library of Science
Résumé : International audience
Understanding the factors underlying the spatio-temporal distribution of infectious diseases provides useful information regarding their prevention and control. Dengue fever spatio-temporal patterns result from complex interactions between the virus, the host, and the vector. These interactions can be influenced by environmental conditions. Our objectives were to analyse dengue fever spatial distribution over New Caledonia during epidemic years, to identify some of the main underlying factors, and to predict the spatial evolution of dengue fever under changing climatic conditions, at the 2100 horizon. We used principal component analysis and support vector machines to analyse and model the influence of climate and socio-economic variables on the mean spatial distribution of 24,272 dengue cases reported from 1995 to 2012 in thirty-three communes of New Caledonia. We then modelled and estimated the future evolution of dengue incidence rates using a regional downscaling of future climate projections. The spatial distribution of dengue fever cases is highly heterogeneous. The variables most associated with this observed heterogeneity are the mean temperature, the mean number of people per premise, and the mean percentage of unemployed people, a variable highly correlated with people's way of life. Rainfall does not seem to play an important role in the spatial distribution of dengue cases during epidemics. By the end of the 21st century, if temperature increases by approximately 3°C, mean incidence rates during epidemics could double. In New Caledonia, a subtropical insular environment, both temperature and socio-economic conditions are influencing the spatial spread of dengue fever. Extension of this study to other countries worldwide should improve the knowledge about climate influence on dengue burden and about the complex interplay between different factors. This study presents a methodology that can be used as a step by step guide to model dengue spatial heterogeneity in other countries.
ISSN: 1935-2727

Droits : http://creativecommons.org/licenses/by/
pasteur-01247037
https://hal-riip.archives-ouvertes.fr/pasteur-01247037
https://hal-riip.archives-ouvertes.fr/pasteur-01247037/document
https://hal-riip.archives-ouvertes.fr/pasteur-01247037/file/journal.pntd.0004211.pdf
DOI : 10.1371/journal.pntd.0004211

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
Nous étudions dans l'anneau Fq[X0, X1] des polynômes à m + 1 variables et à coefficients dans le corps fini à q éléments, l'idéal homogène J engendré par les polynômes homogènes qui s'annulent sur tout l'espace. Cet idéal s'introduit naturellement lors de l'étude des codes de Reed-Muller projectifs ([7], [8]). Nous donnons une résolution libre du quotient Fq[X0, ..., Xm]/J en utilisant le complexe de Eagon et Northcott [4] qui généralise le complexe de Koszul [5]. Ceci permet en particulier de calculer directement les dimensions des composantes homogènes de l'idéal.
ISSN: 0022-4049

hal-00767441
https://hal.univ-antilles.fr/hal-00767441
https://hal.univ-antilles.fr/hal-00767441/document
https://hal.univ-antilles.fr/hal-00767441/file/ccod0010.pdf

Éditeur(s) : Agrocampus-ouest
Résumé : Quantitative evaluations of management measure impact on resource and fleet viability are required to provide diagnostics and feed decision process. Fisheries are complex systems where numerous dynamics interact at various spatial and temporal scales. This complexity and the difficulty to observe marine systems are responsible for the high uncertainty surrounding fishery functioning that makes it hard to identify adequate management measures. The increasing consideration of uncertainties in diagnostics argues in favor of robust management measures rather than optimal management measures. Identifying robust management measures requires the development of models that account for complexity and uncertainty in the description and prediction of management impact as well as the identification of synthetic descriptors (indicators) of the state and dynamic of the system. The aim of this work is to propose generic methodologies for developing such tools. The methodological developments are applied to the pelagic fishery of the bay of Biscay. This fishery mainly focused on anchovy and was highly impacted by fishing ban on this species since 2005. Causes for anchovy drop in biomass either environmental or due to fishing are poorly understood, making management decision much harder in a critical environmental and economic context. The adequacy of traditional management using TAC is questioned and additional spatial management measures are proposed.
Des évaluations quantitatives de l’impact de mesures de gestion sur les ressources et la viabilité des flottilles sont nécessaires pour établir des diagnostics et pour fournir des éléments d’orientation pour la gestion des pêcheries. Les pêcheries sont des systèmes complexes mettant en œuvre de nombreuses dynamiques à des échelles de temps et d’espace très variables et interagissant très fortement entre elles. Cette complexité et la difficulté d’observer les systèmes marins, conduit à une grande incertitude sur leur fonctionnement qui rend difficile la détermination des mesures de gestion adaptées. La reconnaissance de ces incertitudes, plaide de plus en plus en faveur d’une gestion robuste plutôt qu’optimale. Déterminer la robustesse des mesures de gestion nécessite le développement de modèles de description et de prédiction capable d’intégrer la complexité et l’incertitude du système, ainsi que d’outils de description synthétiques (indicateurs) de l’état et de la dynamique de la pêcherie à partir d’observations. L’objectif de ce travail est de proposer des méthodes génériques pour le développement de tels outils. Ces développements méthodologiques sont appliqués à la pêcherie pélagique du golfe de Gascogne. Cette pêcherie ciblant principalement l’anchois a été fortement perturbée par la fermeture de la pêche à cette espèce en 2005. Le déterminisme environnemental et le rôle de la pêche dans le déclin de la population d’anchois sont mal compris ce qui rend difficile la prise de décision dans un contexte économique et écologique critique. La gestion traditionnelle par TAC pourrait s’avérer insuffisante pour permettre l’exploitation durable et des mesures spatialisées sont envisagées en complément.

Droits : 2010 The Author, UEB
http://archimer.ifremer.fr/doc/00025/13630/10712.pdf
http://archimer.ifremer.fr/doc/00025/13630/

Éditeur(s) : HAL CCSD Lavoisier
Résumé : National audience
Dans cette étude, nous présentons, une méthode permettant de découvrir, par le biais de variables agrégées et d’un algorithme génétique, des informations non connues a priori et utiles pour la prédiction de crues. Cette méthode se prête particulièrement bien aux données dans lesquelles des informations à propos d’un même élément sont dispersées à travers plusieurs lignes du jeu de données initial.
In this paper, we present an heuristic based approach for feature selection in the context of flood prediction. Features are complex variables that represent aggregate values. We apply a preprocessing method on data in order to elicit relevant information that could not be easily accessible initially because it is split through several lines of a dataset. A genetic algorithm is used in order to search for the features that may prove the best performances for flood prediction.
ISSN: 1633-1311

hal-01184510
https://hal.archives-ouvertes.fr/hal-01184510

Résumé : In the booming context of shrimp culture in French Caledonia, some research actions need to be considered in order (i) to make sure this production will not impact the quality and biodiversity of the coastal environment it is contingent upon (ii) for the field to highlight, based on objective bases, its good "environmental image", for an efficient commercialization of its products. In the framework of the MOM project, a number of actions have been carried out: Action 1: A bibliographical synthesis of the impact of shrimp culture on coastal environment has been conducted, taking into account the works carried out throughout the world and in French Caledonia in order to draw the main issues the Territory will have to deal with. It seems necessary to start with an assessment of the future and the impact of the effluents on the coastal environment. Action 2: With this aim in mind, one or more indicators, among the lagoon's environmental variables, of the influence of the farms' rejects on the coastal environment, have been searched for. The first results are presented in this report. Action 3: In parallel, a long-term program taking into account the different aspects of the problem in its entirety and complexity has been defined in relation with our partners in French Caledonia. With this aim in mind, a symposium entitled "Ecosystems and shrimp culture", organised by the Ifremer in French Caledonia, with the support of the IRD, was held in Nouméa from June 22nd to June 24th, 2005. This symposium enabled the drawing out of the research perspective on the benthos, on the functioning of the lagoon ecosystem, and on the evolution of the mangrove swamps subjected to organic matter inflows from farms.
Dans un contexte d'essor de la filière crevette de Nouvelle Calédonie, des actions de recherche doivent être envisagées (i) pour assurer que cette production n'hypothéquera pas la qualité de l'environnement littoral et sa biodiversité dont elle dépend et (ii) pour que la filière puisse mettre en avant, sur des bases objectives, son image «environnementalement correcte» garante de la bonne commercialisation de son produit. Dans le cadre de ce projet MOM, plusieurs actions ont été menées : Action 1 : Une synthèse bibliographique de l'impact de l'aquaculture de crevettes sur l'environnement littoral a été réalisée en prenant en compte les travaux conduits à travers le monde et en Nouvelle-Calédonie afin de dégager les principales problématiques concernant le Territoire. Il apparaît nécessaire comme première priorité d'être capable d'évaluer le devenir et l'impact des effluents sur l'environnement littoral. Action 2 : A ce titre, un travail de recherche d'un ou de plusieurs indicateurs, parmi les variables environnementales du lagon, de l'influence des rejets aquacoles sur l'environnement littoral a été mené. Les premiers résultats sont présentés dans ce rapport. Action 3 : Parallèlement, il a été défini en relation avec nos partenaires présents en Nouvelle-Calédonie un programme pluriannuel prenant en compte les différents aspects du problème dans sa globalité et sa complexité. Pour cela, un séminaire intitulé « Écosystèmes et crevetticulture » organisé par l'Ifremer en Nouvelle-Calédonie avec le soutien de l'IRD s'est tenu à Nouméa du 22 au 24 juin 2005. Ce séminaire a permis d'élaborer des perspectives de recherche sur le benthos, sur le fonctionnement de l'écosystème lagon et sur l'évolution des mangroves soumis à des apports de matière organique issus des élevages.

Droits : info:eu-repo/semantics/openAccess
http://archimer.ifremer.fr/doc/2006/rapport-6593.pdf
http://archimer.ifremer.fr/doc/00000/6593/

Éditeur(s) : HAL CCSD Springer Verlag
Résumé : International audience
Meconium is the earliest stool of newborns, and is a complex matrix that reflects the degree of exposure of the fetus to xenobiotics. To investigate fetal exposure to volatile organic compounds, an analytical method was developed to identify and quantify BTEX (benzene, toluene, ethylbenzene, and o,m,p-xylene) and two chlorinated solvents (trichloroethylene and tetrachloroethylene) in meconium. Headspace-solid-phase microextraction coupled with gas chromatography-mass spectrometry was selected because it is simple, sensitive, can be automated, and requires no extensive sample preparation. Several extraction variables were optimized (fiber type, incubation time, temperature of fiber, and use of salt). Because meconium is a complex matrix, quantification by SPME was considered carefully because of potential interference, for example competitive adsorption. Calibration in water was compared with calibration in meconium using external and internal methods (with isotope-labeled compounds). In meconium, limits of quantification were determined to be in the range 0.064-0.096 ng g(-1) for the investigated compounds. All target compounds were determined in "real-case" meconium samples.
ISSN: 1618-2642

hal-01064967
https://hal-univ-rennes1.archives-ouvertes.fr/hal-01064967
DOI : 10.1007/s00216-014-7836-2
PUBMED : 24838489

Éditeur(s) : HAL CCSD
Résumé : National audience
Afin de prendre en compte les inégalités sociales de santé, il est fréquent d'utiliser des indices socio-économique qui synthétisent différents aspects du statut socio-économique (SES) à l'échelle de l'individu, du voisinage ou de la région. De nombreux indices existent déjà mais nombre d'entre eux utilisent un faible nombre de variables, des méthodes d'agrégation simples et/ou sélectionnent les variables uniquement depuis la littérature. Nous avons développé une nouvelle procédure de création d'indices SES visant à sélectionner à partir d'un grand nombre de variables celles qui composeront l'indice et à les agréger à l'aide de techniques d'analyses de données. Cette procédure est plus complexe à mettre en œuvre pour des non statisticiens que d'autres approches existantes, c'est pourquoi nous l'avons implémentée dans un package R nommé SesIndexCreatoR. Ce package vise à donner des outils aussi simples d'emploi que possible pour effectuer la procédure de création et exploiter ses résultats, tout en conservant les différentes possibilités et la flexibilité qu'elle offre.
Deuxièmes rencontres R
Lyon, France

Droits : info:eu-repo/semantics/OpenAccess
hal-01433087
https://hal.archives-ouvertes.fr/hal-01433087
https://hal.archives-ouvertes.fr/hal-01433087/document
https://hal.archives-ouvertes.fr/hal-01433087/file/SesIndexCreatoR_20130328.pdf

Éditeur(s) : HAL CCSD Lavoisier
Résumé : International audience
In this paper, we present an heuristic based approach for feature selection in the context of flood prediction. Features are complex variables that represent aggregate values. We apply a preprocessing method on data in order to elicit relevant information that could not be easily accessible initially because it is split through several lines of a dataset. A genetic algorithm is used in order to search for the features that may prove the best performances for flood prediction.
ISSN: 1633-1311

hal-00840738
https://hal.archives-ouvertes.fr/hal-00840738
DOI : 10.3166/isi.16.3.113-139

Éditeur(s) : ESA Publication
Résumé : Satellite remote sensing has emerged as an essential and necessary observing system to acquire global information about the state of the ocean. Complemented with in situ observing networks, the ultimate goals are to be able to make accurate estimates of selected key sets of geophysical variables, with the intention of either making operational predictions across time and spatial boundaries, or advancing fundamental knowledge through development of empirical relationships and theoretical models. For satellite oceanography, improvements are then constantly being sought in our understandings of the geophysical processes, the sensor physics, the electromagnetic and microwave properties and interactions at the complex air-sea interface. Challenges appear as unlimited as the variety of sea surface dynamics and boundary layer meteorological conditions with their broad range of spatial and temporal scales across the globe. To face these challenges, numerous efforts took places over the passed decade to build an ever-increasing quality, quantity, duration and integration of ocean observations. In parallel, simulation capabilities largely improved. All these efforts are then all critically calling for improved methodologies to better structure the wealth of information that is made readily accessible. This latter aspect is a very demanding new component for future multidisciplinary scientific research. Major innovations to consolidate sensor data repositories, to automate tailored queries, to extract, reveal and quantify relationships will then closely associate computer science developments and applied statistics with comprehensive theoretical and experimental thematic studies.
Proceedings of OceanObs'09: Sustained Ocean Observations and Information for Society (ESA Publication), 2010

Droits : info:eu-repo/semantics/openAccess
http://archimer.ifremer.fr/doc/00029/14046/11241.pdf
DOI:10.5270/OceanObs09.pp.12
http://archimer.ifremer.fr/doc/00029/14046/

Résumé : Cette thèse à pour objet l'étude théorique et numérique de solutions dans les équations aux dérivées partielles non linéaires de la physique, en particulier en dynamique des fluides. La présence de discontinuités dans les solutions de ces équations complique la compréhension mathématique des phénomènes mis enjeu et leur traitement numérique, notamment en vue de simulations informatiques . Nous étudions ces équations par une méthode de régularisation dans un espace fonctionnel approprié. Lorsque des schémas numériques construits par des méthodes différentes conduisent à des résultats identiques, ceci jusque dans leurs moindres détails, il semble alors naturel de s'interroger dans quelle mesure ces suites de solutions numériques constituent une approximation d'une solution des équations étudiées. Nous construisons des suites de solutions approchées à partir d'un schéma numérique original,stable et suffisamment simple pour démontrer que ses suites constituent une méthode asymptotique de Maslov au sens des distributions en dimension trois d'espèce. La technique de régularisation employée consiste à étendre les variables réelles du problème ne des variables complexes, ce qui nous permet de construire des familles de solutions particulières que l'on ramène au cas réel en faisant tendre un petit paramètre vers O. Les solutions physiques recherchées apparaissent alors comme valeurs au bord de fonction holomorphes. Nous illustrons les résultats obtenus par des applications en cosmologie dans les cadres Newtoniens et relativistes pour des systèmes sans pression, puis avec pression et auto-gravitation, ainsi que pour le système des gaz parfaits.
This thesis is devoted to the theoretical and numerical study of singular solutions appearing in nonlinear partial differential complicates the mathematical understanding of the phenomena under concem as well as their numerical treatment, in particular in view of computation. These equations are studied by a regularization method in an appropriate functional space. When completely different numerical methods give the same results up to the smallest details one can reasonably expect that these numerical results suggest the existence of a mathematical solution of theses equations. We construct sequences of approximate solutions from an original numerical scheme, which is stable and simple enough to prove that these sequences constitute a Maslov asymptotic method in three space dimension. The regularization technique in use consits in extending the real variables of the problem into complex ones, which perrnits to construct families of particular equations that we bring back to the real case by letting a small paramater tend to zero. The expected physical solutions appear as boundary values of holomorphie functions . Illustrations are given by applications to cosmology in the Newtorian and re1ativistic settings for pressure1ess fluid dynamics, then in presence of self-gravitation and pressure as weil as for the systemof ideal gases

http://www.theses.fr/2011AGUY0478/document

Éditeur(s) : HAL CCSD Taylor & Francis (Routledge)
Résumé : International audience
The influences of task difficulty (index difficulty: 2-4), input device of different length, range of motion and mode of resistance (joystick or rotorcraft stick), and directions of movement (leftward rightward) on motor patterns in a realistic control situation were examined with a multilevel analysis (joint kinematics and muscular variables, and global task performance). Eight subjects controlled the displacements of a virtual object during a slalom task characterized by a realistic inertial model. Pilots adapted the endpoint kinematic organization to increasing accuracy constraints to preserve task success whatever the device and the direction. However, the rotorcraft stick manipulation remains highly complex in comparison to the joystick due to poorer proprioceptive information, higher inertial constraints, and an asymmetrical muscle control.
ISSN: 0022-2895

hal-01442056
https://hal.archives-ouvertes.fr/hal-01442056
https://hal.archives-ouvertes.fr/hal-01442056/document
https://hal.archives-ouvertes.fr/hal-01442056/file/Fuente%20-%20J%20Motor%20Behav%20-%202013.pdf
DOI : 10.1080/00222895.2013.837424

Éditeur(s) : HAL CCSD springer
Résumé : International audience
In this paper, we address the problem of flood prediction in complex situations. We present an original solution in order to achieve the main goals of accuracy, flexibility and readability. We propose the SM2D modular data driven approach that provides predictive models for each sub-process of a global hydrological process. We show that this solution improves the predictive accuracy regarding a global approach. The originality of our proposition is threefold: (1) the predictive model is defined as a set of aggregate variables that act as classifiers, (2) an evolutionary technique is implemented to find best juries of such classifiers and (3) the flood process complexity problem is addressed by searching for sub-models on sub-processes identified partly by spatial criteria. The solution has proved to perform well on flash flood phenomena of tropical areas known to be hardly predictable. It was indeed successfully applied on a real caribbean river dataset after both preprocessing and preliminary analysis steps presented in the paper.
Lecture notes in computer science
Singapour, Singapore

hal-00878780
https://hal.archives-ouvertes.fr/hal-00878780
DOI : 10.1007/978-3-642-40897-7_13

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
Nous étudions dans l'anneau Fq[Xo, X1, ..., Xm] des polynômes à m + 1 variables et à coefficients dans le corps fini à q éléments, l'idéal homogène J engendré par les polynômes homogènes qui s'annulent sur tout l'espace. Cet idéal s'introduit naturellement lors de l'étude des codes de Reed-Muller projectifs ([7], [8]). Nous donnons une résolution libre du quotient Fq [Xo, ..., Xm]/J en utilisant le complexe de Eagon et Northcott [4] qui généralise le complexe de Koszul [5]. Ceci permet en particulier de calculer directement les dimensions des composantes homogènes de l'idéal.
ISSN: 0022-4049

hal-00767463
https://hal.univ-antilles.fr/hal-00767463
https://hal.univ-antilles.fr/hal-00767463/document
https://hal.univ-antilles.fr/hal-00767463/file/ccod0010.pdf

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
Ecological signs of Earth's biosphere forewarn an alarming trajectory towards a global mass-extinction. Assessing species performance and susceptibilities to decline is essential to comprehend and reverse this trend. Yet it is challenging, given difficulties associated with quantifying individual and population processes that are variable across time, space, and life-stages. We describe a new approach to estimating and comparing species performances that combines empirical data, a novel theoretical consideration of population dynamics, and modern statistics. Our approach allows for a more realistic continuous representation of individual performances along development stages while taking into account non-linearity, and natural variability as captured by spatio-temporally replicated observations. We illustrate its application in a coral meta-assemblage composed of populations of the three major reef-building taxa Acropora, Pocillopora, Porites. Using a unique set of highly replicated observations of individual coral dynamics under various environmental conditions, we show how taxa differ in their investment in recruitment and size-specific aptitude for growth and survival, notably through different use of clonal shrinkage, fragmentation, fission, and fusion processes. Our results reveal contrasting life-history trade-offs among taxa which, along with differing patterns of density-dependent recruitment, modulate species responses to decline. These differences in coral life history traits reflect opposing life-strategies, imply regulation at differing life-stages, and explain divergence in species trajectories. Our findings indicate a high potential for resilience in Pocillopora and Porites populations, thanks respectively to a sustained recruitment that promotes demographic elasticity through replacement of individuals, and a steady resistance to mortality which confers persistence through lingering of individuals. Resilience in Acropora appears more arbitrary, given high susceptibility to perturbations and dependency of recruitment on presence of established local populations. We identify management actions that can complement Acropora's life history and benefit recovery of its populations following mortality events. Our regression-modelling approach to quantifying and comparing species performances in different population processes is applicable to all taxa, as illustrated even those with complex clonal life histories, and can be implemented at wide spatio-temporal and taxonomic coverage. It can promote more accurate representation of species dynamics in both descriptive and predictive modelling approaches. The semi-parametric contrast curve method we develop facilitates comparing response variables along continuous explicative metrics while accounting for multiple sources of complexity in empirical data. It should widely benefit investigations in ecology and quantitative science.
ISSN: 1476-945X

hal-01212337
https://hal-univ-perp.archives-ouvertes.fr/hal-01212337
DOI : 10.1016/j.ecocom.2015.07.003

Éditeur(s) : HAL CCSD
Résumé : Background Acute splenic sequestration is one of the major clinical forms of sickle cell disease. This potentially life-threatening complication is defined by a rapid sequestration of sickle red blood cells in the spleen. Approximately 30% of young patients will develop this complication and 49% of them will have recurrences. Methods The decision tree method is a statistical method that can act as a medical decision tool in the case of a complex biological problem. The decision tree method is used here to determine the variables involved in the severity and recurrence of acute splenic sequestration crises. Results A prior decision tree shows that platelet level, variation in spleen size, patient age and hygrometry also called relative humidity define acute splenic sequestration severity. A second one shows that spleen size variation, atmospheric pressure and patient age define recurrence. Conclusions For the first time, biological and environmental parameters are combined to discover some interesting rules which can lead to more accurate evaluations of acute splenic sequestration severity and recurrence. Thus, this method provides a sharp diagnostic tool which could improve medical treatment for better patient care.
https://hal.archives-ouvertes.fr/hal-00201460

hal-00201460
https://hal.archives-ouvertes.fr/hal-00201460
https://hal.archives-ouvertes.fr/hal-00201460v2/document
https://hal.archives-ouvertes.fr/hal-00201460/file/DT_Marianne_v2.pdf

Éditeur(s) : HAL CCSD Wiley
Résumé :  
Animals, especially insects, are principal pollen vectors of tropical trees and have behavior patterns that affect gene dispersal. Here, we explore complex pollination systems using a new simulation model Eco-Gene and considering, among other factors, flowering synchrony, spatial distribution of trees, degree of selfing, population densities, pollinator flight distances, pollen deposition, and pollinator response to floral display size. Sensitivity analyses using two contrasting tree data sets (Jacaranda copaia and Dipteryx odorata) determined the importance of each parameter on three response variables: the proportion of seeds from self-pollination, effective population size, and pollen dispersal. Spatial considerations and attractiveness of floral displays were prominent features determining the population genetic result of pollinators, and some biological implications of the results are discussed.
ISSN: 0006-3606

hal-01032091
https://hal.archives-ouvertes.fr/hal-01032091
DOI : 10.1111/j.1744-7429.2004.tb00309.x

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