Speaker: Marco Tarantino Studente Magistrale in Statistica e Data Science (LM-82), Universita’ degli Studi di Palermo
APPLICAZIONE DI TECNICHE STATISTICHE PER LA PREVISIONE DELLA
TEMPERATURA EFFETTIVA DI STELLE GIOVANI
Uno degli attuali argomenti di discussione riguarda la durata del processo di formazione di cluster stellari: vi sono modelli teorici che ipotizzano un unico evento dal quale si formano tutte le stelle appartenenti ad uno stesso cluster, considerando un processo di formazione rapido, altri modelli che, invece, suppongono piu’ eventi di formazione, tenendo conto di un processo di formazione lento. Uno dei metodi piu’ utilizzati per ricavare l’eta’ delle stelle si basa sul confronto della distribuzione delle stelle sul diagramma H-R con le isocrone teoriche. Una variabile fondamentale per applicare questo metodo e’ la temperatura efficace, la quale si puo’ ricavare in maniera accurata dalla spettroscopia. Tuttavia, nell’epoca delle grandi survey fotometriche, e’ necessario derivare la temperatura efficace basandosi sulla fotometria. L’obiettivo del lavoro consiste nel trovare un modello che permetta di ottenere buone previsioni della temperatura effettiva di stelle giovani, utilizzando variabili fotometriche di stelle ottenute da tre cataloghi differenti, ovvero Gaia, Gaia-ESO Survey, Pan-STARRS e 2MASS. I risultati previsivi evidenziano come il Random Forest rappresenti una buona soluzione per la previsione della temperatura effettiva. Durante questo seminario introdurro’ i cinque approcci statistici applicati nell’analisi svolta, evidenziando in particolare i risultati relativi al Random Forest, risultato il migliore dei cinque.
Speaker: Giorgio La Malfa
Relatore: Marco Miceli
Titolo:Anisotropies in core-collapse supernova explosions: modeling the evolution of a magnetorotational supernova remnant
The core-collapse supernovae (CCSNe) whose explosion is driven by mag-
netorotational instabilities (MRIs) are believed to be viable sources of extremely
interesting astrophysical phenomena, such as hypernovae, super-luminous SNe, magnetars and gamma-ray bursts. Observations of the supernova remnants (SNRs) resulting from these objects pose an important tool for their study. To this end,
numerical simulations offer valuable insights.
Here my goal is to investigate to which extent the supernova remnant (SNR)
of an MR-SN retains memory of the explosion asymmetry, and to study the mor-
phology of the stellar ejecta throughout the evolution.
I performed a three-dimensional (3D) hydrodynamic (HD) simulation of a MR-
SNR, by evolving a state-of-the art MR-SN model available in the literature. The
simulation covers from a few hours after the shock breakout to ∼ 10,000 yrs, with
the adoption of an analytically prescribed circumstellar medium (CSM).
The early outflow jet-like asymmetry, characteristic of MR-SNe, causes the for-
mation of a Mach disk in the equatorial plane. This turns into a torus-like high
pressure region and leads the SNR to a bicone morphology. The pristine bipolar
jet-like structure shows an asymmetry in the ejection time. The simulation indi-
cates that the remnant keeps memory of this asymmetry, presenting a narrower
morphology in the direction of the first ejected jet. The forward shock presents
an elongated morphology, with a polar-to-equatorial ratio of ∼ 1.12 for the first ∼
200 yrs, decaying quite rapidly (down to ∼ 1.06) at ∼ 500 yrs. The stellar ejecta
exhibit a higher ratio of up to 1.20. The ejecta asymmetry increases when they
extend to the forward shock through RT instabilities, deforming the shock since
∼ 200 yrs. A comparison between a proxy of the X-ray emission and an X-ray
Chandra observation of Kes 73 (a SNR hosting a magnetar) indicates some simi-
larities, though a more accurate treatment of the CSM could significantly improve
the agreement with observations.
This first glimpse into the evolution of anisotropies in SNRs originating from
MR-SNe indicates that the SNR keeps memory of the anisotropies in the MR ex-
plosion on a time scale of centuries, rather than millenia.
Speaker: Mario Guarcello (INAF)
Titolo: “EWOCS: status of the project.”
Abstract: “The EWOCS project has the objective of studying star and planet formation, and early stellar evolution, in very young massive clusters (VYMCs). With a mass in excess of 10^4 solar masses, the very few VYMCs known in the Milky Way represent the most accessible examples of starburst regions, which are very rare in our Galaxy today, but common in galaxies experiencing epochs of intense star formation. These regions are characterized by very high stellar density, and they are dominated by a rich and compact ensemble of massive stars that produce an environment dominated by energetic radiation and particles. With a distance of 3.87 kpc and 4.5 kpc, respectively, the Westerlund 1 and 2 clusters are the closest VYMCs to the Sun, and thus the best targets to study how stars and planets form in the most energetic star forming environment known. In this talk, I will present the status and the preliminary results of the EWOCS project, which is mainly based on a 1Msec Chandra/ACIS-I Large Project and a cycle 1 JWST observation of Westerlund 1, a cycle 2 JWST observation of Westerlund 2, and other data at high spatial resolution of the two clusters.”
Google meet: https://meet.google.com/sxz-cctp-tsc?pli=1&authuser=1
Speaker: Maria Kopsacheili (ICE-CSIC)
Titolo: New larger sample of Supernova Remnants in NGC 7793, using MUSE IFS.
Abstract: Study of Supernova Remnant (SNR) demographics and their physical properties (density, temperature, shock velocities) is very important in order to understand their role in galaxies. Many photometric and spectroscopic studies of SNRs have been carried out in our Galaxy but also in extragalactic environments. The most common means for the SNR identification in the optical regime, is the use of the flux ratio of the [S II] (λλ6717, 6731) to Hα (λ6563) emission lines. However, this diagnostic is biased against low excitation SNRs. For this reason, we have developed new diagnostics that combine 2 and 3 emission line ratios along with a Support Vector Machine model, that efficiently differentiate SNRs from HII regions. These diagnostics recover up to 35% of the SNRs that we miss using the traditional diagnostic tool, which is very important in order to obtain more complete samples of SNRs (i.e. SNRs of different physical properties) and consequently to more efficiently explore the feedback processes to the host galaxy. We present the application of these diagnostics on Integral Field Unit (IFU) data of the galaxy NGC 7793. We identify new SNR populations, we construct the distributions of their physical properties and their luminosity functions. Finally, we explore possible correlations between properties of SNRs and those of their environment.