Portail HAL UNIV-RENNES
Not a member yet
180610 research outputs found
Sort by
Improving human papillomavirus vaccination coverage in France: a need to act but not to rush
No full textInternational audienceSuboptimal vaccine uptake is a worldwide public health issue. We need (i) to develop interventional population health research on vaccine uptake and (ii) to ensure better use of the research results by public health authorities before developing and implementing a national strategy
New age constraints for the Perdasdefogu Basin, Italy: implications for early Permian vertebrate palaeobiogeography and Variscan Belt dynamics
No full textInternational audienceDuring the late Carboniferous and early Permian (ca. 320 to 270 Ma), the climate of the Earthwent through a global warming by evolving from an icehouse to a greenhouse state. Thisclimate change was coeval with the final stages of the Variscan orogeny, and both climate andtectonic contributed to reshape landscapes, drainage patterns, and dispersal routes of varioustaxa across Pangea. The Perdasdefogu Basin, Sardinia, Italy, was located to the south of theVariscan Belt and comprises a rich fossil record. As such, determining the age of fossil remainsfrom this basin bear clues to reconstruct potential dispersal routes across the Variscan beltand discuss the role of climate and tectonics in the dispersal of taxa between the north andthe south of the belt. This study reports new sedimentological, palaeontological andgeochronological data from the Perdasdefogu Basin, Sardinia, Italy. Sedimentologicalinvestigations showed that the basin corresponded to a permanent lake initially infilled byepiclastic and volcaniclastic material, then by carbonates, and finally covered by thick felsicvolcanic series. Lacustrine sediments yielded various fossils including plant and temnospondylamphibian remains. Geochronological investigations, performed by U-Pb analyses on zircon,indicate that the Perdasdefogu Basin was filled during the early Permian, between297.9±0.9/3.1 and 295.7±0.7/3.0 Ma (without/with systematic uncertainties propagated,respectively), and covered shortly after, at 295.9±1.1/3.1 Ma, by volcanic series. These ageconstraints show that the temnospondyls of the Perdasdefogu Basin, located in the southernside of the Variscan belt, were coeval to those of early Permian basins located in the northernside of the belt. This demonstrates a rather large areal distribution for these amphibian fossils,which encompassed the northern, inner and southern parts of the Variscan belt. This in turnimplies probable hydrographic connections that allowed for vertebrate migration betweenbasins located across the belt. While hydrographic networks represent an obvious dispersalmechanism for ichthyofaunas, geochronological and palaeontological data for thePerdasdefogu Basin suggest that this dispersal mechanism also holds for tetrapods, at leastfor aquatic and semi-aquatic groups. The reorganisation of hydrographic networks due to thecombined effects of climate change and Variscan tectonics could then have represented adriving mechanism for the migration of tetrapods in central Pangea during the late Carboniferous and early Permian
Mud mounds or not mud mounds? That is the question for the early Devonian red marbles of the Minervois (France)
No full textInternational audienceCarbonate mud mounds are currently the less understood bodies among the reefal structures,while they exist since the Precambrian and they still range from tropical to polar latitudes inthe modern oceans. Carbonate mud mounds are dome-shaped biosedimentary buildupsdominated by fine-grained carbonates supporting various types of fabrics, such as thrombolitic,fenestral, or stromatactoid structures. The geological record provides the key to understandingthe diversity of ancient and today’s carbonate mounds, their unconventional biosedimentaryprocesses including the formation of their fabrics.The Minervois red marble quarries are known to reflect the occurrence of at least sevencarbonate mounds dated from the Emsian (Early Devonian). However, evidence for theirpresence is rather incomplete. Our study aims to consolidate their temporal and environmentalframeworks and to investigate the processes of their origin and those behind their famousstromatactoid cavities.Our study is based on geological mapping to identify the geometry of the carbonate mudmounds. We then analyzed sedimentary facies, constrained the various local tectonic phases,and carried out geochemical, isotopic and mineralogical analyses on 60 matrix, cavity, faultand vein-type objects from six main quarries.Our study shows that only a few published mounds exhibit the typical mudstone towackestone texture. On the contrary, they often show a low organic content associated withhigh bioclastic and detrital contents. The rare unfragmented macrofossils often show signs ofreworking. In addition, the dome shape of some published mud mounds is induced by thesecondary folds of the larger Minervois nappe fold. Therefore, only a few quarries may reflectthe presence of Emsian mud mounds with a very flat, at best low-relief geometry (flanksbetween 5 and 10°) in the Minervois. It is important to highlight that stromatactoid cavities,supposedly characterizing a mud mound, occur in other facies and stratigraphic levels (up tothe Famennian Griottes Formation).Whenever the stromatactoid cavities occur within or outside a mud mound, theirdevelopment correlates with sediment grain size and inversely with stratonomy. We show thattheir morphology and infill are not dependent on a particular sedimentary texture or the age ofthe rocks. We suggest that these cavities may be due to the decomposition/dissolution oftransported organisms, to fluid escape phenomena within the sediment, and/or to the action ofbottom currents opening breaches in the indurating micritic sediment. We show that thecavities have a polyphase filling of matrix micrite and radiaxial calcite precipitated inoxygenated seawater circulation.We propose that calcitic sealing of cavities occurs within the oxic zone below thewater/sediment interface during early diagenesis. Thus, cavities were mainly filled by different generations of calcitic cement and micritic sediment during the uppermost Emsian and Eifelian(duration <~15 Ma), then remobilized by Givetian and Frasnian cone waters during the openingof sedimentary dykes. During the Variscan orogeny in the Carboniferous, hydrothermal events,including manganese oxide veins, impacted the evolution of cavity fillings and the carbonatefabric of the mounds
A foreign cauldron full of imports: Elemental and lead isotope analyses of bronze objects from the Hassle hoard (Late Bronze Age Period VI, HaC–HaD), Sweden
No full textInternational audienceThis study uses metal analyses to examine one of the most import-rich metalwork finds in the Scandinavian final Bronze Age—the Hassle hoard (probably deposited c. 500 BCE)—containing two ciste a cordoni/Rippenzisten, two Hallstatt Mindelheim swords (one with bronze pommel), two small hooks and twelve bronze-iron ornamental discs in a large bronze cauldron, all submerged in the now-drained river Äverstaån. Twelve samples from a selection of eight objects in the Hassle hoard have been analysed, primarily for chemical composition and lead isotopic signature. The aim is to study their alloy recipes, and origins and potential production areas, also allowing for comparison of similarities and differences between objects in the assemblage. Apart from the sword pommel, with a typical fahlore signature, the alloys in the Hassle objects differ considerably from most of the alloys in previously analysed, contemporary Nordic Bronze Age objects. The cauldron is very atypic for the area, so far without any clear match in relevant isotopic datasets. The analytical results support a foreign origin, both for the metal and for the objects. The results show that the objects, even those from typologically comparable groups, are made of several alloys, from raw materials of different origins. As relevant reference material for objects of similar types is so far limited, these analyses contribute to build dataset for future comparisons. The results underline the complexity in both raw material and alloys circulating in Late Bronze Age Scandinavia and Early Iron Age Europe
Saint-Bonnet-de-Mure (69), Le Château, Chemin de la Grange Haute. Un site d'ensilage en plaine du Velin: Rapport Final d'Opération d'Archéologie Préventive
No full textL’opération de fouille archéologique située sur la commune de Saint-Bonnet-de-Mure, au lieu-dit « le Château » s’est déroulée entre la fin de l’été et l’automne 2021. Initialement prévue pour une durée de six semaines, du 06 septembre au 15 octobre, elle s’est vue prolongée par l’activation de deux tranches conditionnelles jusqu’au 5 novembre. Elle a couvert une surface de 4800 m² et accuse une pente assez forte du sud vers le nord, de l’ordre de 6 % en moyenne.À la suite du premier décapage mécanique, un total de 179 anomalies a été repérées. Leur traitement nous a poussé à annuler 30 d’entre elles. En dehors des traces d’une limite parcellaire vraisemblablement contemporaine située au centre de l’emprise, deux phases d’occupation principales sont caractérisées, l’une médiévale, l’autre moderne. La plus grande partie des vestiges identifiés renvoient au Moyen Âge central, dont la chronologie peut être précisée, entre la fin du XIe et le début du XIIIe siècle d’après l’étude céramologique. La réalisation de deux séries d’analyses au radiocarbone permet également d’envisager deux phases d’occupation, l’une située entre la fin du XIe et le milieu du XIIe siècle, l’autre au XIIIe siècle. Les vestiges médiévaux sont constitués par un total de 47 silos répartis en quatre batteries réparties sur l’ensemble de la partie orientale de l’emprise de fouille et constituant autant de zones de stockage, complétés par quelques rares éléments isolés. Il apparaît une certaine hétérogénéité dans le nombre de structures constitutives de chaque ensemble, qui peut varier de cinq à dix silos. De même, tant les volumes de stockage que la forme des structures est relativement variable. Les éléments les plus massifs sont implantés dans la batterie la plus septentrionale (Batterie 1), qui serait, au regard des résultats des analyses radiocarbone, à associer à la première moitié du XIIIe siècle. Les silos qui composent les autres batteries présentent des volumes variables, parfois très importants, mais compris, en règle générale, entre 600 L et 1200 L. Plusieurs formes de chambres de stockage ont été identifiées. La plupart apparaissent globulaire avec un rapport diamètre/hauteur proche de 1, mais certains silos présentent une forme plus allongée marquée par une hauteur supérieure au diamètre de la chambre de stockage. La batterie 2 située au centre de l’emprise de fouille est la seule à disposer de structures plus ramassées, de forme générale lenticulaire, caractérisées par un diamètre très important et une hauteur limitée. Enfin, certains éléments de la batterie 1, au nord, présentent une forme particulière, tronconique ou en « bouteille à long col ». Les différentes formes semblent ainsi cohabiter sans que leur mise en œuvre soit liée à une spatialisation ou une répartition chronologique précise. Plusieurs structures complètent cette occupation, notamment trois silos isolés, dont les formes globulaires, les volumes et les comblements rendent les résultats d’étude moins riche. Nous pouvons également relever la présence d’un four en partie septentrionale. Celui-ci devrait vraisemblablement être complété par plusieurs éléments analogues situés hors de l’emprise de fouille au regard des dynamiques de comblement identifiées dans certaines structures. Sa vocation, uniquement domestique ou également liée au traitement des récoltes n’apparaît pas clairement et il est possible que les deux fonctions soient complémentaires. Il apparaît cependant, au vu de son architecture simple et de la rubéfaction toute limité qu’il conserve, que son usage ait été relativement limité, tant en termes de durée d’utilisation que d’intensité. Une fosse charbonneuse située en partie méridionale de l’emprise de fouille pourrait également être liée au traitement des récolte, avec une fonction de séchoir à grains, mais les arguments manquent pour nous assurer de cette interprétation. Enfin, deux excavations ont été caractérisées. L’une, subcirculaire, est située en partie médiane du site et dispose de deux trous de poteaux aménagés dans son fond, mais d’éventuelles traces de superstructures manquent pour s’assurer de son identification comme fond de cabane. La seconde, située en partie sud-est de l’emprise de fouille, correspond à un espace excavé sur plus de 1 m, rectangulaire, doté de multiples aménagements périmétraux qui permettent de restituer une charpente dotée d’un faîtage. Un petit foyer est aménagé sur son fond, possiblement doté d’une hotte, tandis que son accès est permis par une entrée placée dans l’angle nord-est du creusement principal. Il dispose d’un sol couvert d’un plancher dans un deuxième état d’utilisation, avant de servir de dépotoir rapidement après son abandon et la disparition de sa toiture.Une autre occupation, associée cette fois à la période moderne, a également été identifiée. Elle est marquée par l’aménagement d’un réseau de drains en tuyaux de terre cuite dans la partie septentrionale de l’emprise de fouille. Trois de ces linéaires sont implantés dans un axe sud-est/nord-ouest, tandis que le quatrième accuse les écoulements des premiers et les évacue selon un axe sud-ouest/nord-est. Dans un second temps vraisemblablement proche, une série de fosse, éventuels silos, sont aménagés en quinconce le long de ce dernier. Cette batterie compte alors dix structures relativement arasées, mais dont l’une présente un léger départ de sape qui suggère une fonction de stockage. Les carporestes y sont rares et il est difficile de définir quelles denrées pouvaient y être stockées, mais leur faible profondeur d’implantation suggère un entreposage somme tout très court. Une dernière fosse située en partie occidentale, sans connexion avec ces éléments, a livré un liard à l’effigie de Louis XIV et pourrait permettre de restreindre l’intervalle de datation fourni par radiocarbone pour cette phase d’occupation à l’époque moderne
Eco-ES 2025: 3rd workshop on Eco-design and circular economy of Electronic Systems
No full textAs this document is a collection of abstracts, to cite a single abstract, please report only the authors and the title of the contributionInternational audienceMany environmental impacts arise from electronics and, among them, electronic waste is rising quickly to the top of priority issues. Because of the number of electronic systems growing exponentially and their short life cycles, Europe asks for strong and rapid actions towards a sustainable development and use of electronics. Eco-design and circular economy applied to electronic systems are thus becoming major challenges for the electronic community to respond to the dangers for the environment; besides the exponential increase in electronic waste generation, other issues are the depletion of resources, the contribution to climate change and the poor resiliency to supply-chain. Electronic designers willing to engage in eco-design face several difficulties, related in particular to a limited knowledge or a lack of data to estimate the environmental impact from the design phase. Other difficulties come from the challenging application of circular economy strategies to electronics and the uncertain extension of the service lifetime of the system or parts of the system, owing to the variability in user behavior and business models.The objective of the workshop Eco-ES is to gather experts from both academia and industry, covering a wide scope in the environmental sustainability of electronics. Eco-ES is proposed as a workshop of the conference DATE since 2023. In the third edition at DATE 2025, the event hosted a keynote of Jean-Pierre Raskin (UCL), an invited talk of Marc Jouss´e (Retronix) and talks selected from an open call. This document collects the abstracts of the talks
GEMMify: A Unified Model for Layer-Wise DNN Performance Prediction on Edge GPUs
No full textInternational audiencePerformance prediction techniques enable early design space exploration to create and deploy cost-effective Deep Neural Networks (DNNs). Specifically, layer-wise DNN performance predictions produce a fine-grained performance breakdown of DNNs, proving to be valuable for automated model optimization. However, current methods embed a predictive model for each type of DNN layer, which increases their complexity and lowers their practicality when targeting novel neural networks. Using similarities between General Matrix Multiplication (GEMM) operations and main DNN layers, such as convolutional (CONV) and fully connected (FC) layers, we introduce a novel method for predicting the layer-wise power, latency, and energy consumption of DNNs. The proposed method uses GEMM dimensions as common features for the main layers, allowing performance prediction using a single model. Experiments on the Jetson Orin edge device show that the proposed method achieves state-of-the-art Mean Absolute Percentage Error (MAPE) ranging from 9.77% to 15.28% while offering widely applicable models
