149544 research outputs found
Sort by
National road traffic noise estimation with ensemble learning and multimodal geodata
No full textThe European Noise Directive mandates the mapping of noise – high, continuous sound pressure levels considered to be a major health threat. However, the strictest rulesets apply to specific regions only and the majority of residential areas are unmapped. Transfer learning was deployed to close spatial data gaps between the official, strategic road traffic noise maps. The three most suitable hyperparameter configurations achieved weighted Kappa values (a measure of ordinal agreement) ranging between 0.889 and 0.956 during repeated cross-validation. The best model achieved an overall classification accuracy of 90.7 % when tested against held-out samples. 7.8 % of predictions exhibited minor deviations within ± 5 dB(A). The model was subsequently deployed to predict road traffic noise across Germany at 10 x 10 Meter resolution for 2017. The results suggest a total of 13.1 million people exposed to yearly averaged road traffic noise (Lden) above 55 dB(A) and stress need for improved noise policies
Remote Sensing Capabilities for Identifying Hydrological Tipping Points and Associated Urban Risks in the Andes
No full textGRAPHCSVAE: Graph Categorical Structured Variational Autoencoder for Spatiotemporal Auditing of Physical Vulnerability Towards Sustainable Post-Disaster Risk Reduction
Get PDFIn the aftermath of disasters, many institutions worldwide face challenges in continually monitoring changes in disaster risk, limiting the ability of key decision-makers to assess progress towards the UN Sendai Framework for Disaster Risk Reduction 2015-2030. While numerous efforts have substantially advanced the large-scale modeling of hazard and exposure through Earth observation and data-driven methods, progress remains limited in modeling another equally important yet challenging element of the risk equation: physical vulnerability. To address this gap, we introduce Graph Categorical Structured Variational Autoencoder (GraphCSVAE), a novel probabilistic data-driven framework for modeling physical vulnerability by integrating deep learning, graph representation, and categorical probabilistic inference, using time-series satellite-derived datasets and prior expert belief systems. We introduce a weakly supervised first-order transition matrix that reflects the changes in the spatiotemporal distribution of physical vulnerability in two disaster-stricken and socioeconomically disadvantaged areas: (1) the cyclone-impacted coastal Khurushkul community in Bangladesh and (2) the mudslide-affected city of Freetown in Sierra Leone. Our work reveals post-disaster regional dynamics in physical vulnerability, offering valuable insights into localized spatiotemporal auditing and sustainable strategies for post-disaster risk reduction
RANS Turbulence Modeling for Aerospace Applications. Development, Validation and Prospects
Get PDFThis talk gives an overview over selected topics in the development of RANS turbulence modeling for aerospace applications. The basic calibration of RANS models is revisited and implications for non-canonical flows with streamwise pressure gradients, surface curvature and vortical flows are outlined. The aim is to bring together the view from RANS modeling with experimental investigations and theoretical as well as empirical findings of turbulent flows regarding the mean velocity field and the Reynolds stresses. Then some comments and implications for data-driven turbulence modeling approaches are given. Finally, some thoughts for future prospects are given
ExcitationSolve
No full textEs handelt sich um eine Software-Publikation, die in einer Journal Publikation (https://www.nature.com/articles/s42005-025-02375-9) verwendet wurde, um Grundzustände von verschiedenen Molkülsystemen mit dem Variational Quantum Eigensolver (VQE) zu bestimmen.
Die Software ist öffentlich, unter dem Namen ExcitationSolve, zugänglich auf GitHub: https://github.com/dlr-wf/ExcitationSolv
Multi spectral investigation of volcanic deposits in Vulcano, Italy, and the PETRAS campaign/summer school
No full textThe volcanic settings of the Aeolian Islands, Italy, offer access to various types of volcanic terrains, with diverse morphology and mineralogy. The current signs of volcanic activity on the island of Vulcano, whose last eruption is dated to 1890, include e.g. the presence of large fumarole fields on ground and underwater. The dry landscape and easy access to layers of older and more recent volcanic material, in addition to the possibility of investigating secondary minerals, make this site a very promising analogue for the Moon, Mars, Venus and other planetary environments. For instance, the active and acidic fumarolic sites provide extreme acid alteration conditions of volcanic deposits, which also have been a key process at local and regional scales throughout Martian geologic history. For these reasons, Vulcano is a remarkable training site to test instruments, rovers, or data processing techniques for planetary exploration and has a high astrobiological potential pertaining to the questions of life detection, using spectroscopy techniques, and habitability, looking at microbial colonization in extreme environments. Since 2019, we combined three different field instruments to investigate volcanic deposits: a portable visible and near infrared spectro-radiometer, a portable LIBS (laser-induced breakdown spectroscopy) instrument, and a portable Raman system; thus, combining mineralogical, elemental and molecular information. This combination significantly improved the characterization of the spectral properties of the volcanic deposits in the study area and therefore offered a more comprehensive view into their alteration processes. In short, LIBS, or any instrument solving the elemental composition of the rocks, is essential to study the composition of lava rocks and to distinguish between different types, while VIS-NIR and Raman spectrometers are powerful tools for the detection, discrimination and characterization of alteration and hydrothermally formed minerals. In parallel to the field campaigns, and building on previous editions going on since 2015, the PETRAS Summer school (Planetology, Exploration, Terrestrial analogs, Robotics, Astrobiology and, Spectroscopy) was launched in 2024. PETRAS involves volcanologists, geophysicists, astrobiologists, electronic/mechanical engineers and computer scientists to provide a 360-degree vision of the different skills needed by the planetary sciences community. Results on the combination of VIS-NIR, LIBS, and Raman will be presented, showing the complementarity of the techniques, and lessons learned from the field campaigns, especially on the alignment of measurements. Data processing techniques are also in focus to help in the interpretation of the results (Stephan et al. under review). Finally, a brief overview of the recent and current PETRAS campaigns/summer school will be presented to show the unique multi-disciplinary aspects and opportunities for young researchers as well as different research groups interested in all aspects of planetary exploration from science to technology, outreach, and dissemination
The space experiments BIOMEX and BioSigN to prepare for in situ life detection missions and habitability studies
No full textThe search for life in our Solar System is at the centre of several current and future robotic missions to Mars and beyond, to the icy moons of Jupiter and Saturn. New instruments, in the context of space exploration, have been sent to Mars on recent rover missions, such as Raman spectrometers particularly suited for the fast and non-destructive identification of biomolecules embedded in minerals. International and national efforts have also been focusing on bringing samples back from Mars to be analysed with state-of-the-art laboratory instruments on Earth. And two missions en route to Jupiter will teach us more about the environment and potential organic content of another excellent target for finding life beyond Earth, Europa. However, little is known about the stability of putative biosignatures in the Martian or space environment. To support and prepare these current and future exploration missions to Mars, Europa and Enceladus, exposure experiments in low Earth orbit, using the International Space Station (ISS) are crucial. For instance, during the BIOMEX experiment (2014-2016), biomolecules and microorganisms were exposed for 16 months to a simulated Martian environment in LEO. This environment was provided in the EXPOSE-R2 module, outside the ISS and comprised of UV and ionizing radiation, a Mars-like atmosphere, extreme temperature cycles, and analogues of Martian regolith. Seven of those biomolecules were analysed post-flight using Raman spectroscopy; all remained detectable. The next ESA space exposure experiment BioSigN (Biosignatures and habitable niches) will follow in the next few years extending the range of samples to icy moons relevant organisms and molecules. These experiments, and the international efforts required to achieve them, are crucial to advance our knowledge on the detectability of putative traces of life outside of Earth, and on the potential habitable conditions of our neighbouring planets and moons. To maximize the scientific outputs, the space experiments are always connected to the results obtained on ground from planetary analogue field sites and planetary simulation facilities
Application relevant load cycles for PEMFC component development and hybrid system optimization
Get PDFPolymer electrolyte membrane fuel cells (PEMFCs) are one of the key technologies for the decarbonization of transportation. In contrast to light-duty vehicles (LDV), durability requirements for heavy-duty (HD) transport applications are challenging. To overcome these difficulties, material improvements must be complemented by optimized operation strategies. Furthermore, specific testing protocols in application relevant conditions are required. Harmonized testing protocols already exist for automotive applications. However, for HD applications, such protocols are still pending, which represents a bottleneck for further development. A semi-empirical model is used in a novel systematic top-down methodology to generate application related power demand cycles in PEMFC of an HD hybrid vehicle is presented and demonstrated in detail for the case of PEMFC in HD transport. The resulting load profile, applicable at single cell and stack levels, is proposed as a starting point for a harmonization of open-source HD load cycling and testing protocols for PEMFC component development. Furthermore, the method is also used to evaluate the impact of the energy storage system (ESS) capacity and of the hybridization strategy parameters on the PEMFC stack power demand dynamic and efficiency; providing up to 2 % efficiency increase and a 50 % reduction in FC load changes