Portail "HAL-Francophonie Afrique et Océan Indien"
Not a member yet
43052 research outputs found
Sort by
Evidence of megathrust earthquakes and seismic supercycles in subtropical Japan from millennia-old coral microatolls
No full textInternational audienceAbstract Megathrust earthquakes in subduction zones often go unreported because they are rare and the historical record is short. On the Ryukyu subduction zone of southwestern Japan, unlike neighboring Nankai Trough, the history and future potential of great interplate earthquakes are not well known. While the geodetic measurements on the islands suggest that the plate coupling is very weak, recent observations of slow seismic events as well as offshore geodetic measurements imply the presence of coupled patches along the megathrust. Furthermore, the historical and geological studies indicate evidence of great tsunamis. Here, we use fossil microatolls in Ishigaki island to reconstruct the relative sea level in the Holocene. The coral record reveals several relative emergence episodes clustering between 5-4 and 3-2 thousand years ago (ka). Elastic modeling shows that the observed motions can correspond to coseismic uplift associated with megathrust earthquakes. The clusters of megathrust events suggest possible supercycles of earthquakes with a recurrence interval of more than 2 ka. Such results imply a strong seismic hazard for the upcoming centuries. The devastating 1771 Meiwa earthquake and associated tsunami may mark the onset of the most recent seismic supercycle
Doravirine versus dolutegravir-based regimen in antiretroviral treatment-naive people living with HIV-1 (ANRS0392s ELDORADO): protocol for an international, open-label, randomised, non-inferiority, phase III trial
No full textInternational audienceIntroduction: Increasing evidence suggests that dolutegravir (DTG), endorsed by the WHO since 2018 for first-line antiretroviral therapy (ART), is associated with significant weight gain and potentially also with cardiometabolic disorders. In an effort to expand therapeutic options for people living with HIV (PLHIV), the EvaLuating the non-inferiority of DORAvirine vs DOlutegravir trial aims to compare the virologic efficacy of doravirine (DOR) and DTG-based regimens and to assess their safety, including a focus on cardiometabolic effects.Methods and analysis: This is an international, phase III, multicentre, open-label, non-inferiority, randomised trial that will enrol 610 ART-naïve PLHIV (HIV RNA≥1000 copies/mL at screening) across six countries (Brazil, Cameroon, France, Côte d'Ivoire, Mozambique and Thailand) spanning four continents. Key inclusion criteria include age ≥18 years, confirmed HIV-1 infection with plasma RNA levels ≥1000 copies/mL, indication for ART initiation and no prior ART exposure. Participants will be randomised in a 1:1 ratio to receive either DOR 100 mg once daily in combination with tenofovir disoproxil fumarate (TDF) (300 mg daily) plus lamivudine (3TC) (300 mg daily) or DTG (50 mg daily) in combination with TDF (300 mg once daily) plus either emtricitabine (FTC) (200 mg daily) or 3TC (300 mg daily). Randomisation will be stratified by screening HIV-1 RNA load (≤100 000 or >100 000 copies/mL) and by country. The primary outcome is virological efficacy, defined as the proportion of participants achieving HIV-1 RNA <50 copies/mL at week 48 on the assigned treatment (FDA Snapshot algorithm). Secondary outcomes include cardiometabolic safety endpoints (ie, weight gain, insulin resistance, hypertension, diabetes, waist and hip circumferences, waist-to-hip ratio, fasting glycaemia, insulin and fasting serum lipids), along with mental health, quality of life, virological and immunological parameters. Final data collection is expected by July 2028.Ethics and dissemination: Primary outcome results (week 48) are expected in early 2028. The project was submitted to and approved by national ethics committees and pharmaceutical regulatory authorities in all participating countries: Brazil (CEP INI FIOCRUZ (21.040-900)/CEP HGNI (26.030-380)); Cameroon (CNERSH (2024/09/1717/CE/CNERSH/SP)/Ministry of Public Health (D30-1464/AAR/MINSANTE/SG/DROS/CRC); Côte d'Ivoire: (CNESVS (0018224/MSHPCMU/CNESVS-km)/AIRP (1329/AIRP/DISMP/Om/kbaag); France (CTIS CPP/ANSM (2023-508626-10-00)); Mozambique (CNBS (20/CNBS/25)/ANARME (4635/380/ANARME)); Thailand: (IHRP (08/1944)/Thai FDA: ongoing on 19 January 2026). The trial received authorisation from the French National Commission for Data Protection and Liberties (CNIL) under approval number 924 302. Written informed consent is obtained from all participants prior to any study-specific procedures and trial enrolment, in accordance with the Declaration of Helsinki and applicable national regulations. Study findings will be disseminated through publication in peer-reviewed journals and presentations at national and international scientific conferences. Results will also be communicated to policymakers, healthcare professionals, community stakeholders and study participants through appropriate dissemination activities, including policy briefs, stakeholder meetings and lay summaries on dedicated and easily accessible platforms
Optimal control synthesis for a class of production-regeneration systems -Application to membrane filtration
No full textInternational audienceThis paper studies an optimal control framework for a class of production-regeneration systems, which have to alternate between productive and regenerative phases, such as membrane filtration. These systems are modeled as control-affine systems of triangular forms with a one-dimentional internal state. Pontryagin's Maximum Principle and Green's theorem provide conditions on optimal trajectories, distinguishing bang and singular arcs, where singular arcs are distinguished between turnpike or anti-turnpike. Structural results show that optimal solutions admit at most two switching times and the exhaustive list of admissible structures is provided. The paper further introduces a constructive method to provide optimal feedback synthesis by determining switching and dispersal loci through differential continuation. Numerical applications to membrane filtration processes illustrate the approach for energy minimization and volume maximization cases. An example with non-connected singular locus illustrates the method's generality
Evaluation of upland rice variety mixtures in the Madagascar highlands
No full textSource Agritrop Cirad (https://agritrop.cirad.fr/616353/)International audienceContext or problem Madagascar heavily depends on rice for caloric intake, especially through irrigated farming. In the Vakinankaratra region, rainfed upland rice farming is an important complement but faces challenges such as poor soil fertility and vulnerability to pathogens like Pyricularia oryzae. Objective or research question To address these challenges, we evaluated extended elite lines upland rice varietal mixtures adapted to local conditions, focusing on yield performance, stability, and food security. Methods Four upland rice varieties: Chhomrong Dhan, FOFIFA 172, FOFIFA 173, and FOFIFA 180, well-suited to Madagascar's high-altitude climatic conditions and resistant to Pyricularia oryzae, were evaluated in three experiments conducted in the highlands of Madagascar between 2013 and 2017. The experimental design assessed all variety combinations, considering two treatment factors: 'mixture type' (ranging from pure stands to mixtures of all four varieties) and 'varietal composition' (15 modalities representing different varietal combinations). The analysis included the identification of the best-performing varietal compositions using a mixed-effects linear regression model and land equivalent ratio calculations. Results The analysis revealed that grain yield did not significantly differ among various mixture types, while varietal composition within mixtures had a highly significant effect. No mixture combination yielded more than the best varieties in pure stand although some matched their performance. Three combinations showed a significantly improved land equivalent ratio. Varieties differed in competitive abilities and trait plasticity. Conclusions Contrary to literature suggesting that increased diversity through varietal mixtures enhances production, the study found that the number of varieties in mixtures (mixture type) did not have significant effects. However, it appears possible to identify specific mixture combinations with strong mixing abilities. Implications This study evaluated mixtures of elite rice varieties that are currently available to farmers, under optimal fertility management. In these conditions, varietal mixtures did not demonstrate clear advantages over pure stands. However, results may differ under low-fertility conditions more representative of farmers' fields, or when using a broader genetic diversity. These scenarios warrant further investigation. In such contexts, varietal mixtures could complement other diversification strategies aimed at enhancing the resilience of agricultural systems, particularly in vulnerable regions such as the Madagascar Highlands
Pathogen detection success and trends in lethal sampling during bacterial pathogen research in bats
No full textInternational audiencePathogen surveillance is crucial to understanding the current occurrence and emergence of diseases. Nevertheless, detection success can greatly depend on collection method, source of samples, and diagnostic approach. In this work, we gathered data from published literature on zoonotic bacterial pathogen sampling and detection in bats from studies conducted globally, with a focus on molecular detection methods and specific sample types (e.g., tissue, blood, guano, or urine). Additionally, we explored geographic and taxonomic patterns of lethal collection risks to bats. We found that sample type did not affect pathogen detection, with blood or urine samples from live-caught individuals being equally successful for bacterial pathogen surveillance compared to tissues from lethal sampling across several pathogenic groups. Furthermore, our data showed a strong taxonomic and geographical bias, indicating that certain bat groups and regions are most affected by lethal sampling. Overall, since non-lethal sampling appears to be as effective as invasive and lethal approaches, we recommend minimally invasive sample collection whenever possible. However, lethal sampling of bats for pathogen surveillance appears to be a common practice in published literature despite the availability of less invasive methods. An understanding of how bat collection affects populations and their conservation status is still limited. More sustainable sampling methods are a viable option of minimizing our negative impact on wild and threatened populations, while still conducting effective pathogen surveillance
Students’ Perceptions and Expectations of Higher Education in Burundi: The Case of the University of Burundi
No full textInternational audienc
Landscapes of uncertainty: Mangrove rice farmers' perceptions of rainfall variability and climate change adaptation in three coastal regions of Guinea-Bissau, West Africa
No full textInternational audienceRainfall variability poses growing challenges to mangrove rice farming in Guinea-Bissau, threatening both local livelihoods and food security. This study investigates how farmers in the coastal regions of Cacheu, Oio, and Tombali perceive changing rainfall patterns and adapt their cropping calendars accordingly. Using a companion modelling approach, we co-designed a serious game with 45 mangrove rice farmers from 13 villages, complemented by group interviews and participant observation. The results reveal that farmers possess detailed, experience-based knowledge of rainfall variability-particularly regarding the delayed onset, mid-season dry spells, and early cessation of rains-which directly influences key agricultural decisions such as nursery establishment, transplanting, and harvesting. Through the serious game, participants collectively identified and validated four main adaptation strategies: (1) flexible adjustments to cropping systems (e.g., shifting to direct seeding in erratic years), (2) strategic use of short-and medium-cycle rice varieties, (3) enhanced water management through reinforced dykes and drainage innovations, and (4) diversification into complementary livelihood activities such as cashew cultivation and fishing. The serious game proved to be an effective tool for facilitating dialogue, sharing local knowledge, and collectively refining context-specific adaptation pathways. These findings underscore the value of integrating farmers' ecological knowledge with participatory methodologies to support resilient agricultural planning under climate uncertainty.</div
DroidHunter : une détection robuste basée sur la vision contre les logiciels malveillants Android obfusqués
No full textInternational audienceDue to their large popularity, Android smartphones are often targeted by malware attacks. Several strategies exist to detect malwarecode. However, we show that they are insufficient when dealingwith obfuscation techniques. DroidHunter is our novel methodfor detecting Android malwares. DroidHunter leverages opcodesand their parameters, transforming those into RGB images andspecific encoding techniques. The generated images are then usedto train two different classification models based on support vector machines, convolutional neural networks, and a vision-basedtransformer. We evaluate DroidHunter on several datasets withup to 476,937 APKs from multiple sources. With detection ratesfrom 98.65% to 99.94%, DroidHunter overcomes nine state-of-the-art malware detection techniques, including Drebin, MaMadroid,DexRay. Moreover, DroidHunter demonstrates strong resilienceagainst hidden malware with detection rates up to 98.98%, andshows robustness on newly emerging threats, achieving an AUT of0.89 on recent malware samples. We release our code to the researchcommunity, with instructions to reproduce our evaluation availableat: https://zenodo.org/doi/10.5281/zenodo.10977166.En raison de leur grande popularité, les smartphones Android sont souvent la cible d'attaques de logiciels malveillants. Il existe plusieurs stratégies pour détecter les codes malveillants. Cependant, nous montrons qu'elles sont insuffisantes lorsqu'il s'agit de techniques d'obfuscation. DroidHunter est notre nouvelle méthode de détection des logiciels malveillants Android. DroidHunter exploite les codes opérationnels et leurs paramètres, en les transformant en images RVB et en techniques d'encodage spécifiques. Les images générées sont ensuite utilisées pour entraîner deux modèles de classification différents basés sur des machines à vecteurs de support, des réseaux neuronaux convolutifs et un transformateur basé sur la vision. Nous évaluons DroidHunter sur plusieurs ensembles de données comprenant jusqu'à 476 937 APK provenant de multiples sources. Avec des taux de détection compris entre 98,65 % et 99,94 %, DroidHunter surpasse neuf techniques de détection de logiciels malveillants de pointe, notamment Drebin, MaMadroid etACM Asia Conference on Computer and Communications Security (ASIA CCS '26)DexRay. De plus, DroidHunter fait preuve d'une forte résilienceface aux logiciels malveillants cachés, avec des taux de détection pouvant atteindre 98,98 %, et se montre robuste face aux nouvelles menaces émergentes, avec un AUT de 0,89 sur des échantillons de logiciels malveillants récents. Nous publions notre code à la communauté des chercheurs, avec des instructions pour reproduire notre évaluation disponibles à l'adresse suivante : https://zenodo.org/doi/10.5281/zenodo.10977166
