Archive Ouverte d'INRAE
Not a member yet
513551 research outputs found
Sort by
Assimilation de données pour la prévision des débits d’étiage
No full textWithin the framework of WP3 of the ANR CIPRHES project, data assimilation techniques have been developed and/or evaluated at spatial scales ranging from watershed to the entire country of France. General conclusions common to all of these studies emerge. The first is that assimilating observed streamflows improves the simulation of river flows for historical periods. It appears to have a modest but present effect on low flow forecasts. The results from these studies show a very short persistence of the assimilation contribution in forecasting (less than 10 days). Moreover, the results demonstrate a real interest in terms of operational scores (e.g., anticipating the onset of drought periods). Finally, assimilating groundwater table levels does not seem to improve either the simulation of river flows for historical periods or future river flow forecasts.As a perspective, it is necessary to increase the number of assimilated observation points, either to improve the contribution of assimilation itself, or to enhance the method in order to improve its robustness in more varied hydrological contexts. Finally, the modalities for which a transition to operational use in the PREMHYCE platform could be considered in the future still need to be defined.Dans le cadre du WP3 du projet ANR CIPRHES, des techniques d’assimilation ont été développées et/ou évaluées à des échelles spatiales allant du bassin versant jusqu’à la France entière. Il en ressort des conclusions générales communes à l’ensemble de ces travaux. La première est que l’assimilation des débits observés permet d’améliorer la simulation des débits en périodes historiques. Elle semble avoir un effet modeste mais néanmoins présent sur la prévision des débits d’étiage. Les résultats issus de ces travaux montrent une persistance très courte de l’apport de l’assimilation en prévision (inférieure à 10 jours). De plus, les résultats montre un réel intérêt au regard de scores plus opérationnels (exemple : anticiper le démarrage des périodes de sécheresse). Enfin, l’assimilation des niveaux de nappe ne semble pas apporter d’amélioration ni sur la simulation des débits en période historique, ni sur la prévision future des débits.En perspectives de ces travaux, il est nécessaire d’augmenter le nombre de points d’observation assimilés, soit pour améliorer l’apport de l’assimilation en elle-même, soit pour améliorer la méthode pour qu’elle soit robuste dans des contextes hydrologiques plus variés. Enfin, il reste à définir les modalités pour lesquelles un passage en opérationnel dans la plate-forme PREMHYCE pourrait être envisagé dans le futur
Bayesian Adaptive Deep Reinforcement Learning for Personalized Treatment in Multiple Myeloma
No full textPersonalized management of multiple myeloma relies on accurate assessment of the patient’s current disease state, while accounting for its potential evolution, to guide follow-up and treatment adaptation. Piecewise Deterministic Markov Processes (PDMPs) offer an appropriate framework for modelling cancer progression. When treatment or monitoring decisions must be made, these models naturally lead to impulse-controlled PDMPs formulated as continuous-space Partially Observed Markov Decision Processes (POMDPs). Existing solution methods, however, assume perfectly known model parameters of the POMDP, which is unrealistic, given the likely variability between patient responses to treatment. We address this challenge by developing a Bayes-Adaptive POMDP (BAPOMDP) framework for controlling partially-observed \\ PDMPs with uncertain model parameters,enabling the decision-maker to explore a wide range of beliefs about individual patients’ disease dynamics while planning, without requiring large cohort data. We solve the resulting BAPOMDP problem using simulation-based deep reinforcement learning algorithms to compute near-optimal policies in the resulting high-dimensional continuous model. This demonstrates the practical relevance of Bayes-adaptive control for personalized cancer follow-up
The biogenic sulfur cycle in the coupled ocean–sea ice–atmosphere system
No full textInternational audiencePolar oceans and sea-ice regions are global hot spots for the production of biogenic volatile methylated sulfur (VMS) compounds: dimethyl sulfide (DMS) and methanethiol (MeSH). VMS compounds make important contributions to atmospheric particle formation and cloud property modulation, especially when polar atmospheres are pristine. As a result, the polar biogenic sulfur cycle may induce significant climate feedback in response to ongoing sea ice decline. However, polar VMS production, emission, and atmospheric oxidation processes remain poorly represented in current numerical models, hampering assessments of their radiative impacts and, in turn, implementation of targeted observations necessary for providing predictive understanding of changes in the ocean–sea ice–atmosphere (OIA) system. We synthesize current knowledge of the polar biogenic sulfur cycle and its representation in models. To untangle the existing gaps and provide a roadmap toward predictive understanding, we identify key features of sea ice habitats for biological VMS production, sea ice physical features that enhance or suppress VMS emissions, and atmospheric VMS oxidation at low temperatures that controls the contribution of oxidation products to particle formation or growth. These features are tightly coupled, emphasizing the need for coordinated efforts across disciplines that span the OIA interface, and among observational, experimental, and modeling communities. We recommend 4 priority research areas: (1) model representation of biological VMS production at the sea ice bottom and surface; (2) improved quantification of cloud condensation nuclei (CCN) sensitivity to VMS emissions with updated gas phase and multiphase oxidation chemistry at low temperatures; (3) better spatial and seasonal quantification of MeSH abundance and its biological and chemical controls in sea-ice environments; and (4) assessment of the contribution of episodic extreme VMS emissions during sea ice breakup for the polar CCN budget
Past2ECO : Des cultures du passé aux cultures du futur : utilisation de solutions numériques pour mobiliser la diversité génétique adaptative pour l'agroécologie.
No full textNational audienc
Stationary imbibition with evaporation through a flattened triangular channel
No full textInternational audienceA study of the stationary flow in a channel of flattened triangular cross section subject to evaporation on one end and in contact with a liquid reservoir on the other end is presented taking into account the liquid film flow in the corners of the channel and in the adsorbed thin liquid film together with transport by diffusion of the vapor in the gas phase. Three main regimes are distinguished depending on the inlet liquid pressure, namely, the pure corner flow regime, the regime with partial bulk invasion, and the regime where the channel is fully invaded by the liquid. The various regimes are summarized in a phase diagram that depend on the inlet liquid pressure and external air relative humidity. Models are derived for each regime combining the physical description of each transport mechanism. Numerical solutions are obtained using a simple bisection method. Results show that the mass transfer rate through the channel can vary over several orders of magnitude depending on the inlet pressure. Situations where the mass transfer rate is independent of the external air relative humidity are exhibited both for the pure corner flow and partial bulk invasion regimes
Sociologie des goûts homosexuels
No full textInternational audienceComment se fabriquent les préférences amoureuses et sexuelles ? Selon quelles modalités socialisatrices ? Et suivant quels processus vont-elles se différencier en fonction du genre, de la classe sociale et des parcours individuels ?C’est à ces questions que ce livre s’attache à répondre, en s’inscrivant dans la lignée de travaux ayant montré que la sociologie peut appréhender des objets dont on pense a priori qu’elle n’a rien à en dire, car ils se situeraient hors du social. Ce travail montre que les socialisations qui fabriquent les goûts amoureux peuvent faire l’objet du même type d’analyse sociologique que les goûts culturels. En se concentrant sur une population d’individus homosexuels, il étudie les processus de construction des dispositions à aimer amoureusement et sexuellement les personnes de même sexe que soi.L’enquête repose sur des entretiens longs et répétés avec des femmes et des hommes appartenant à différentes classes sociales, permettant de retracer leurs parcours
How a periodic thermal environment can trigger robust coexistence
No full textThe Competitive Exclusion Principle is a cornerstone of ecology, predicting that n species cannot coexist on fewer than n limiting resources. While this principle is robustly validated in constant environments like the chemostat, it fails to account for the vast biodiversity observed in nature. In this study, we challenge this classical paradigm by investigating how periodic fluctuations in a thermal environment -rather than artificial variations in operating parameters -influence the outcome of single-resource competition. Combining mathematical analysis based on Floquet theory with numerical exploration, we demonstrate that periodic temperature variations can trigger robust, stable coexistence over a wide parameter space. We show that the domain of coexistence expands with increased nutrient enrichment (larger Sin) and slower environmental cycles. However, we also identify a critical trade-off: slower fluctuations induce high-amplitude population oscillations, which may increase the risk of extinction during periodic troughs. Our findings provide a biologically realistic mechanism for biodiversity and suggest that environmental periodicity is a fundamental driver of species coexistence in natural systems
The calf holobiont under challenge: longitudinal microbiome-pathogen dynamics and respiratory health
No full textThe Bovine Respiratory Disease Complex (BRDC) is a major health and welfare challenge driven by multifactorial interactions among pathogenic viruses and bacteria, host and environmental factors, and the respiratory and gut microbiomes. Many implicated bacterial pathogens are also commensals of the respiratory tract, complicating diagnosis and prevention. Growing evidence indicates the roles of the respiratory and gut microbiomes in BRDC, yet their combined effects remain poorly understood. In this study, we dynamically followed the microbiome, pathogen load, and host response in 30 calves over 147 days under commercial rearing conditions. Nearly half developed BRDC, with elevated fever, cough, and lung sound scores, and peak symptoms at day 58 after feedlot arrival. Pathogen detection in nasal cavities revealed distinct patterns: Mycoplasma bovis and BCoV peaked during the first two weeks, IDV and Histophilus somni after one month, Mannheimia haemolytica after two, while Pasteurella multocida peaked after one month and persisted. Importantly, M. haemolytica and Pasteurella multocida loads correlated with higher BRDC scores, whereas BCoV and Mycoplasma bovis were associated with diarrhea, suggesting systemic effects. Nasal beta-diversity diverged between groups at the symptomatic window, and healthy animals exhibited higher fecal diversity and evenness early in life. The respiratory pathobionts Pasteurella and Corynebacterium were enriched in diseased calves, whereas potentially protective families (Lachnospiraceae, Oscillospiraceae) were more abundant in healthy ones. Multivariate analyses confirmed that antibiotic treatments and short-chain fatty acids, especially isovalerate and isobutyrate, further modulated both fecal and nasal microbiomes, with consistently stronger impacts in diseased animals. Together, these findings demonstrate that BRDC outcomes are shaped not by pathogen burden alone, but by the interplay among respiratory and digestive microbiotas, pathogens, environment, and host factors. Our study highlights the importance of a holobiont perspective that integrates both gut and respiratory microbiotas to better elucidate the complexity of BRDC. Such an inclusive framework may provide new insights into disease mechanisms and inform the development of innovative therapeutic strategies
Projet Agro'Deep : les outils d'annotation d'objets
No full textAgro 'Deep is a remote visual data processing platform using deep neural networks that offers services for the detection, segmentation and identification of organs, plants and landscapes for digital agriculture Network learning and validation require annotated image datasets. Two digital tools are available to facilitate preliminary annotation tasks: (1) Annotator for drawing and labelling bounding boxes on images, and (2) Checker for visually checking annotation/detection pairing and correcting dataset annotations.Agro 'Deep est une plateforme de traitement visuel à distance utilisant des réseaux neuronaux profonds qui offre des services de détection, de segmentation et d'identification d'organes, de plantes et de paysages pour l'agriculture numérique. L'apprentissage et la validation en réseau nécessitent des ensembles de données d'images annotées. Deux outils numériques sont disponibles pour faciliter les tâches d'annotation préliminaires : (1) Annotator pour dessiner et étiqueter des cadres de sélection sur les images, et (2) Checker pour vérifier visuellement les associations annotation/détection et corriger les annotations des ensembles de données.</div
Proteomic profile data of Klebsormidium nitens alga grown in control and saline conditions
No full textInternational audienceKlebsormidium nitens (K. nitens) is an alga of the charophyte class used as a model for studying the adaptation of plants to terrestrial life. Its genome has been completely sequenced and 16,215 protein-coding genes have been predicted. In this study, the proteome of K. nitens grown under standard conditions or after salt stress was investigated. A total of 1190 proteins were experimentally confirmed and 922 of them were classified according to their cellular location, molecular or biological function. Of these 922 proteins, 62 and 124 were specifically found in the control and salt-treated samples, respectively. However, no specific function or location could not be assigned on the basis of the primary sequences. A protein-protein interaction network based on the 124 proteins found in saline conditions was constructed using STRING analysis. All the data are accessible and are of interest for phycologists, as well as for evolutionary plant biologists, and provide a foundation for future studies investigating how K. nitens responds to salt stress