Repository for Publications and Research Data
News from the ETH Library
Recently Added
Modeling Brightness Temperature of Prominences on the Solar Disk Using ALMA Single-Dish Observations
Item type: Journal Article
Matković F.; Brajša R.; Benz A.O.; et al. (2025)
Prominences (PRs) are among the most common solar phenomena, yet their full physical picture, particularly their chromospheric mm emission, remains incomplete. The new Atacama Large Millimeter/submillimeter Array (ALMA) presents an opportunity to study PRs at mm and sub-mm wavelengths through a combination of measurements and theoretical modeling. We utilize ALMA single-dish measurements alongside data from other radio instruments to model the PR brightness temperature through adaptation and modification of the 1D semi-empirical Avrett–Tian–Landi–Curdt–Wülser (ATLCW) quiet-Sun (QS) model. The calculated and measured PR brightness temperatures were found to be lower than the measured QS value and predictions from the unperturbed ATLCW QS model across the ALMA wavelength range, consistent with PRs appearing in absorption. The PR density was found to be 60–163 times higher and temperature 155–163 times lower than the QS level, aligning with previous measurements. A key finding emerged with the non-hydrostatic equilibrium assumption, yielding a more physically consistent PR brightness temperature. This suggests that PR stability is most likely maintained by its magnetic field obeying magnetostatic conditions rather than by pure hydrostatic equilibrium, supporting recent studies. Additionally, our results confirm that thermal bremsstrahlung is the dominant radiation mechanism for PRs at mm and sub-mm wavelengths.
On the Dark-Energy Enigma
Item type: Journal Article
Chamseddine A.H.; Fröhlich J. (2025)
A model is presented that provides an explanation for the presence of (Dark Matter and) Dark Energy in the universe. A key idea is to express the volume form of the Lorentzian metric on space–time in terms of a positive function of a new scalar field multiplying a certain four-form given by the wedge product of the differential of the mimetic scalar field and a certain closed three-form. An ansatz for this three-form related to one commonly used to determine the winding number of a map from a three-dimensional hypersurface to a three-sphere is discussed. An action functional depending on the space–time metric, the new scalar field, the mimetic scalar and the three-form is proposed, and the field equations are derived. Special solutions of these equations for a Friedmann–Lemaître universe are presented.
Playback experiments highlight the importance of nearest-neighbor distance and social information for nest site selection in the House Martin (Delichon urbicum)
Item type: Journal Article
Brühlmann R.; Michler S.P.M.; Grüebler M.U.; et al. (2025)
Understanding nest site selection is crucial for species conservation. Bird conservation often involves installing nesting aids to increase nest site availability and induce colonization of unoccupied sites. However, prospecting individuals must find nesting aids, which may be facilitated by social information. Here, we investigated the effectiveness of artificial nests and playback in the declining, migratory House Martin Delichon urbicum. We selected unoccupied sites with artificial nests along a distance gradient to occupied sites and broadcasted conspecific vocalizations during prospection times of House Martins in both the post- and the following pre-breeding periods. Visitation and colonization rates increased considerably in proximity to occupied sites. Playback during the post-breeding and pre-breeding periods enhanced visitation rates, while pre-breeding-only and post-breeding-only playback had smaller positive effects. Colonization rate increased exclusively with pre-breeding-only playback. Colonized playback and non-playback sites had similar breeding success, indicating that playback did not create ecological traps by attracting House Martins to suboptimal sites. Hence, broadcasting conspecific vocalizations informs prospecting birds of nest site availability, thereby increasing visitation, and to some degree, colonization of unoccupied House Martin sites. To boost colonization, we recommend installing artificial House Martin nests within approximately 500 m of occupied sites and using playback of conspecific vocalizations.
Hands on Introduction to Generative Machine Learning
Item type: Conference Paper
Sharma R.; Tang M. (2025)
This is an intermediate level course for attendees to gain a strong understanding of the basic principles of generative AI. The course will help build intuition around several topics with easy-to-understand explanations and examples from some of the prevalent algorithms and models including Autoencoders, CNN, Diffusion Models, and Transformers.
Laser Ultrasound Super-Resolution Imaging for Multi-Parametric Non-Invasive Volumetric Characterization of Brain Cancer
Item type: Journal Article
Nozdriukhin D.; Chen Y.; Lyu S.; et al. (2025)
Accurate characterization of brain tumor heterogeneities and microvascular dynamics remains an unmet need in cancer research driving the development of innovative imaging methods. Here, laser ultrasound super-resolution imaging (LUSSI) is introduced, a hybrid modality combining laser-generated ultrasound with optoacoustic (OA) contrast to enable functional volumetric (3D) super-resolution imaging of vascular structures in vivo. The system integrates a laser ultrasound (LUS) emitter and a spherical array transducer to localize microbubble responses in 3D, further enabling concurrent acquisition of multi-spectral (multi-wavelength) OA data. The capabilities of LUSSI to assess tumor heterogeneities are demonstrated in a murine glioblastoma model, including visualization of intra-tumoral vasculature, identification of compressed vessels, and detection of functional abnormalities such as hemorrhages and hypoxic regions. OA imaging further enables resolving oxygenated and deoxygenated forms of hemoglobin, revealing heterogeneous perfusion within the tumor core. LUSSI is further shown to resolve whole-brain vascular networks with ≈35 µm resolution and quantify microvascular blood flow and oxygenation, supporting applications in neuroscience. With its non-invasive, multi-parametric capabilities and proven compatibility with clinical imaging components, LUSSI offers a promising platform for studying vascular pathology, monitoring therapy, and advancing translational imaging strategies.