Thematic area: Astrophysics and Relativity
Core collapse supernovae (CCSNe) mark the dramatic conclusion of massive stars' lives, serving as captivating laboratories for multi-messenger astrophysics, including neutrino and gravitational wave emissions, as well as the synthesis of heavy elements. Despite their significance, comprehending these events poses formidable challenges across theoretical, observational, and computational realms in astrophysics.
This seminar aims to provide an introductory exploration into the simulation of CCSNe, illustrating the key components of simulations and the prevailing challenges confronting the community. Furthermore, we will look into the observational implications of supernova simulations, giving insights into the alignment between theoretical models and empirical data, as well as observational constraints.
Moreover, the seminar will explore the detectability of core-collapse supernovae, focusing on the brief gravitational waves emitted during the collapse and the convective phases after it.
In 2017, the coordinated detection of gravitational waves by three interferometers (event GW170817) made it possible to triangulate the position of the source and therefore look for electromagnetic signals in that area of the sky. This detection was followed by a multitude of electromagnetic signals, including a short-gamma ray burst and a kilonova. These combined detections paved the way for a new era of multi-messenger astronomy, allowing us to gather more data about the sources and thus better understand the physics that governs them. It is believed that these signals originated from a binary system of neutron stars (BNS). To properly understand the data, fully relativistic, magnetohydrodynamic (GRMHD) numerical studies are necessary. In this seminar, I will present the state-of-the-art of GRMHD simulations of BNS and explore the interplay between magnetic field geometry and the formation of jets that can explain short gamma-ray bursts.
Asteroseismology is the study of the interior of stars through their pulsations and oscillations, much like Seismology studies the interior of the Earth. By delving into stellar pulsations, researchers can obtain relevant physical information about the stars' evolution, chemical composition, and internal processes such as convection or magnetic fields.
Delta Scuti stars are pulsating stars with masses between 1.5-2.5 solar masses which exhibit the so-called linear and non-linear oscillations. Their power spectrum is very complex, implying the interaction of several physics processes within them.
The aim of this seminar is to give a brief introduction to the field of Asteroseismology and to present the main characteristics of the Delta Scuti Stars, as well as to detail the importance of the non-linear interactions within them.
One of the most promising and challenging future gravitational wave (GW) sources are core-collapse supernovae. The oscillation modes of the proto-neutron star (PNS) and the stalled accretion shock will be excited triggering the GW emission. Due to the stochastic nature of these signals, it is not possible to use template matching techniques. An alternative way to analyze the signal is to perform asteroseismology in order to infer properties of the PNS. The oscillations can be described by a system of partial differential equations (PDEs), which can be solved as an eigenvalue problem. In that frame, the eigenvalues are the characteristic frequencies of the oscillation modes. We introduce a machine learning technique, the Physics Informed Neural Networks (PINNs), that simplifies the implementation of differential equations and complex boundary conditions making them suitable PDE solvers. Here we demonstrate an eigenvalue solver consisting of PINNs, which is used for the first time in the context of asteroseismology.
In this study, we present a machine learning model, particularly a Hierarchical Bayesian Neural Network (HBNN), designed to predict stellar parameters such as mass, radius, and age, with a particular focus on the last one. Our goal is to emphasize the advantages of transitioning from classical machine learning predictors to probabilistic models, which usually provide a more realistic representation of the data and showcase their robustness in handling uncertainties. Leveraging statistically hierarchical architectures, our Bayesian NN automatizes the learning process, eliminating the need for manual exploration of parameter relationships. The application of our model to Chemical Clocks, a method for dating main sequence stars, demonstrates its ability to handle observational uncertainties and propagate errors effectively. The results demonstrate the model's capability to predict stellar ages with a mean absolute error of less than 1 Gyr in the testing set, showing its effectiveness in addressing the challenges posed by some stellar dating methods.
Thanks to the James Webb Space Telescope (JWST), we are looking at the cosmos as never before, receiving cutting-edge imaging and spectroscopic data, which allows us to investigate the properties of distant, early (z>7) galaxies. However, this process is not simple and a keen knowledge of it is necessary, as it requires the understanding of multiple pipelines and tools, and how to combine them to get quality results that show us how the first structures of the universe were. In this talk, we will talk about how the JWST pipeline (and its different versions) works, how it can be combined with other pipelines, and what other software we need to use and create in order to obtain the physical properties of high-z galaxies observed with JWST NIRSpec-IFU.