256488 research outputs found
Sort by
Transformations in South-Kivu wetlands: towards sustainable management
No full textWetlands in the eastern Democratic Republic of Congo (D.R.C) are crucial ecosystems that sustain biodiversity, support livelihoods, and regulate hydrological and climatic processes. However, since decades they are facing rapid ecological and social transformations driven by human activities, demographic growth, and land use changes. This research provides a comprehensive analysis of their spatial distribution, ecological characteristics, ecosystem services provided, and degradation dynamics in the South-Kivu Province. It uses an integrated approach combining remote sensing, field surveys, soil analyses, and socio-ecological assessments. First, wetland mapping was conducted using optical and Synthetic Aperture Radar (SAR) imagery integrated with topographic, hydrological, and vegetation indices. Four statistical classifiers (Random Forest “RF”, Artificial Neural Network “ANN”, Boosted Regression Tree “BRT”, and Maximum Entropy “MaxEnt”) were tested, with RF performing best (∼ 95:67%). The results revealed that wetlands cover ∼ 13:5% (898,690 ha) of the province, with significant spatial variability across territories. Second, the study characterized 137 wetlands based on soil, vegetation, and hydrological parameters, identifying four main types: permanently flooded marshes, peatlands, swamps, and seasonally floodplain/inland valleys (FP/IV). Anthropogenic pressures—especially agriculture, brickmaking, and fuelwood collection—have significantly altered their physicochemical and ecological functions. Statistical analyses showed that wetland use, drainage intensity, and location (in agroecological zones “AEZ”) largely explain variations in soil quality and biodiversity. Third, through a survey of 510 households, perceptions of wetland ecosystem services (WES) were analyzed revealing that provisioning and regulating services as the most recognized ES, with variations across wetland types and social groups. They are mainly perceived for their food and materials production, with limited recognition of their roles in biodiversity support and environmental regulation. From the Structural Equation Modeling (SEM) results four latent variables—livelihood, knowledge, personal, and geographical factors—were found to shape perceptions, highlighting the sociocultural intertwining of human–wetland relationships. Fourth, a Wetland soil degradation indicator (WSDI) was developed to quantify ecological degradation. The indicator, combining GIS, remote sensing, and soil profile data, revealed higher degradation in brickmaking areas compared to agricultural and intact wetland zones. Proximity to roads and settlements emerged as the main degradation drivers, confirming a spatial gradient of decline from edges to the wetland core. Finally, landscape analysis from 2000 to 2024 demonstrated a steady conversion of natural wetlands into agricultural and human-modified landscapes, resulting in fragmentation and habitat loss. These transformations reflect broader socioeconomic and environmental dynamics that intertwine governance, equity, and climate pressures. Sustainable wetland management in eastern D.R.C thus requires integrated strategies linking ecological restoration with social participation and adaptive land-use planning. This thesis contributes novel insights into the mapping, classification, and socioecological understanding of wetlands in a tropical, data-scarce region. It provides critical tools and frameworks for guiding conservation, restoration, and more sustainable policies across rural landscapes in the great lakes African region
Design and modelling of a reversible HP/ORC Carnot battery tailored for waste heat integration in flooded mines
No full textpeer reviewedCarnot Batteries (CBs) are a promising option for energy storage, acting as a buffer for the variability from renewables and enabling multi-energy integration and dispatch, converting electricity to heat and back to electricity. Although techno-economic studies report promising costs and high feasibility, especially when components from both cycles are shared in long-term storage, there are few prototypes, and the technology readiness level remains near 4. This paper presents a reversible Rankine-based CB designed for integration with an abandoned flooded mine. The system is under construction, being the largest machine of its type. A physics-based model was developed and validated against manufacturer data to assess performance under realistic constraints. The key focus is the role of auxiliaries and temperature-glide control. By actively modulating secondary-loop pump rotational speed, the Organic Rankine Cycle (ORC) achieves up to a 36 % increase in efficiency and the Heat Pump (HP) mode up to 20 % increase in relative efficiency to a constant-glide strategy. Highlighting that no single pair of glide settings is optimal across the full operating envelope, underscoring the need for adaptive control. Neglecting auxiliaries leads to substantial errors: a relative difference of 24 % in round-trip efficiency (RTE) can be achieved when auxiliaries are omitted, resulting in unrealistic performance values and, consequently, an unrealistic feasibility. With auxiliaries and constraints included, the modelled charge–discharge RTE ranges from 22.8 % to 34.7 %, lower than conventional storage but consistent with reported limits for CB technology. However, CBs can also supply industrial heat, reject heat to district heating networks, and/or deliver cooling, making RTE efficiency an incomplete metric for this technology. The analysis indicates that efficiency depends more on operating conditions than on component selection. This highlights that, for CBs connected to low-temperature storage, auxiliary components are decisive for performance. Achieving high efficiency requires water pumps with high part-load efficiency (including both pump and motor), refrigerant pumps capable of high efficiency at low net positive suction head, and the deployment of active control laws governing charge management and pump operation.WEFORMING Buildings as Efficient Interoperable Formers of Clean Energy Ecosystem
Human Dignity on Trial: Welfare Judges, Immigration Politics and Social Change
No full textpeer reviewedBuilding on ethnographic fieldwork in welfare hearings in French-speaking Belgium, this article explores how judges decide between irregular migrants claiming social assistance and the public welfare administrations refusing such claims. Investigating these cases helps to analyze how members of the bench establish truthfulness and ponder the social and political consequences of their decisions. In these contexts, irregular migrants, despite being the more disadvantaged party to the case, regularly win against the state. At the theoretical level, this article provides a counterpoint to two general trends in sociolegal and migration studies. First, it nuances the idea that judicial proceedings generally tend to further or reproduce inequalities by showing how courts can, under certain conditions, help uphold migrants' rights against the state. Second, it highlights the importance of law and formal institutions in the governance of precarious migrants
Post-treatments on carbon xerogels to improve their performance as negative electrodes of Na-ion batteries
Full text linkpeer reviewedA carbon xerogel (CX) with ∼2 μm nodules was synthesized via polycondensation of resorcinol with formaldehyde in water, followed by pyrolysis at 800 °C. The resulting sample underwent surface treatments using Chemical Vapor Deposition (CVD) and/or CO2 activation in order to mask the micropores with a secondary carbon layer or develop additional micropores, respectively. This strategy aimed at understanding the impact of surface modification and closed micropores on the performance of hard carbons as negative electrode materials for Na-ion battery. On the one hand, the coating deposited by CVD was found to display more graphitic-like domains and to close the CX microporosity, leading to enhanced Initial Coulombic Efficiency (ICE) and reversible capacity. On the other hand, due to its very high accessible surface area, the activated sample showed very low ICE (18 %) and reversible capacity (62 mAh g−1). However, once the activated sample was covered with a secondary carbon layer by CVD, the capacity reached 294 mAh g−1 with a high ICE of around 88 %, and an enhanced insertion plateau at low voltage was observed. Additionally, this activated-coated sample showed a high-rate capability and much greater stability than the other samples upon cycling. Such surface treatments provide an effective strategy for both understanding the impact of hard carbon surface properties on Na storage and optimizing their performance for negative electrodes in Na-ion batteries
How grooves control droplet growth, transport and release
Full text linkWater management on surfaces underpins applications ranging from atmospheric water harvesting to heat exchange and surface cleaning. Many existing strategies rely on chemical coatings or micro-texturing, which can be fragile, costly, or difficult to scale. This thesis explores an alternative approach based on simple geometric features, focusing on whether grooves alone can collect, guide, and release small volumes of water on vertical substrates.
We investigate four representative systems that span different flow configurations and degrees of confinement. On fibers and fiber bundles, we show that grooves naturally appearing between strands reorganize droplet dynamics by modifying the film left behind the droplet, reducing dissipation and increasing sliding speed. Under condensation on a vertical plate, we demonstrate that groove spacing selects the drainage pathway: large spacings favor gravitational shedding, while small ones confine droplets to the plateaus and redirect transport into the grooves. At the lower edge of such plates, groove geometry determines the disposition and frequency of droplet dripping. Finally, in a minimal configuration consisting of two parallel grooves, we show that geometry alone can stabilize a thin water film over more than one hundred capillary lengths. At groove termini, the film breaks and releases a droplet through a cyclic sequence of events.
Across these systems, a common principle emerges: groove acts as a minimal feature that structures the flow. Grooves define where liquid accumulates, how it moves, and when it detaches, enabling robust control without coatings or complex fabrication. These findings suggest that simple geometric design can serve as the foundation for scalable, passive, and durable water-handling surfaces, and they outline the key operations needed for a future geometry-driven millifluidic platform.Daphnée12. Responsible consumption and production6. Clean water and sanitatio
Convolutional neural network-based mapping of material micro-structures to deep material networks for non-linear mechanical response prediction
No full textpeer reviewedData-driven approaches make the development of surrogates of complex heterogeneous material responses possible. After being trained using a previously generated data-set during an offline stage, the surrogates can be used as a material law to conduct structural simulations during the online stage. Nevertheless, in view of accounting for the uncertainty and variability of the heterogeneous materials, the surrogates should be able to account for the micro-structure variability, which remains a challenge. Among the possible surrogate candidates, (Interaction-Based-)Deep-Material Networks ((IB-)DMN) offer the advantage that they can extrapolate the response for new material model parameters of the heterogeneous material phases and for arbitrary loading histories outside of their training range. This advantage results from their thermodynamics consistency and from the fact that the IB-DMN learnable parameters represent solely the micro-structure organization and not the phases material response. However, a trained IB-DMN remains an image of a given micro-structure spatial organization realization in terms of clustering etc. A new microstructure realization thus requires a new training process, limiting the interest of the IB-DMN for stochastic multi-scale analyses. In order to address this limitation, we define the learnable or topological arameters of the IB-DMN from a combination of convolution encoder and neural network, with the micro-structure image serving as input data. After training a CNN encoder-decoder,
the encoder part allows extracting the feature vectors of the heterogeneous material directly from micro-structure images. These feature vectors then serve as input of a trained feedforward neural network (FNN) that predicts the topological parameters of the IB-DMN, yielding a “Image to IB-DMN” framework. The methodology is first illustrated in the context of Unidirectional (UD) composites, for which Stochastic Volumes Elements (SVEs) serve as images of the micro-structure realizations. In a second step we show that the machine learning tools can be trained by considering simultaneously composite families of different inclusion shapes such as circular, elliptical and squared. Despite training considering only elastic data, the predictions for a complex pressure-sensitive elasto-plastic model remain accurate. These results demonstrate the complementary roles of the two networks: the CNN encoder–decoder efficiently extracts reduced feature vectors from micro-structure images with diverse inclusion geometries, and the FNN accurately maps these features to the topological parameters of the IB-DMN, establishing a robust, end-to-end image-to-model framework capable of generalizing across different micro-structural configurations.This research has been funded by the Walloon Region under the agreement no. 2010092-CARBOBRAKE in the context of the M-ERA.Net Join Call 2020 funded by the European Union under the Grant Agreement no. 958174.9. Industry, innovation and infrastructur
Recension de Louis Rouquayrol "Descartes et la culture des esprits" (Honoré Champion, 2025)
No full texteditorial reviewe
La minería en Madre de Dios: Efectos sobre la cultura harakbut y la vitalidad del idioma
No full textpeer reviewe
Development of molybdenum doped cerium oxide passive counter electrodes by surfactant-assisted ultrasonic spray pyrolysis
No full textpeer reviewedNumerous optoelectronic systems, such as electrochromic smart windows, require efficient counter electrodes for their functional operation. Herein, cerium oxide (CeO2) based layers are considered as optically-neutral compounds of high electrochemical activity. Their deposition as thin films onto conducting glass substrates is carried out via surfactant-assisted ultrasonic spray pyrolysis, while further considering heteroelement doping with molybdenum (0–10 %at.). Highly transparent and homogeneous films are accordingly produced, demonstrating important ion storage abilities, especially in the optimal case (6 %at. Mo), bearing a 28 mC cm 2 charging capacity, together with 90+% transmittance over a large optical range. Morpho-structural characterizations additionally highlight a high homogeneity in the deposited layers, owing to the presence of the surfactant species, and enhancing the transmittance of the films. Moreover, the substitution of Ce4+ ions by Mo6+ in the crystal lattice is shown to create additional oxygen vacancies in the layers, contributing to the observed increase in charging capacity. Altogether, excellent optical and electrochemical performances are obtained from such Modoped CeO2 formulations, surpassing most of the current related literature. Finally, proof-of-concept electrochromic devices, combining Mo-doped CeO2 optically-neutral electrodes with WO3 films and involving either liquid- or solid, gel-based electrolytes, display great performances of large optical contrasts, fast kinetics, and good coloration efficiencies
Simulating SRSLY: Sensitivity Resolved SubvoxeL sepctroscopY
Full text linkpeer reviewedMagnetic resonance spectroscopy employs localization in an approximately rectangular voxel, which commonly includes contributions from various tissues due to non-rectangular in-vivo tissues geometry. The suggested technique aims to resolve spectral contribution from different spatially distinct regions based on multichannel data, smilar to the imaging Sensitivity Encoding technique. Simulations are performed using spectral data acquired on a phantom and multichannel coil sensitivity profiles. Simulations show feasibility of the suggested spectral separation technique and demonstrate limitations related to chemical shift displacement. Real data does not yet show full spectral separation of the components