An Observationally Driven Multifield Approach for Probing the Circum-Galactic Medium with Convolutional Neural Networks
Authors:
Naomi Gluck,
Benjamin D. Oppenheimer,
Daisuke Nagai,
Francisco Villaescusa-Navarro,
Daniel Anglés-Alcázar
Abstract:
The circum-galactic medium (CGM) can feasibly be mapped by multiwavelength surveys covering broad swaths of the sky. With multiple large datasets becoming available in the near future, we develop a likelihood-free Deep Learning technique using convolutional neural networks (CNNs) to infer broad-scale physical properties of a galaxy's CGM and its halo mass for the first time. Using CAMELS (Cosmolog…
▽ More
The circum-galactic medium (CGM) can feasibly be mapped by multiwavelength surveys covering broad swaths of the sky. With multiple large datasets becoming available in the near future, we develop a likelihood-free Deep Learning technique using convolutional neural networks (CNNs) to infer broad-scale physical properties of a galaxy's CGM and its halo mass for the first time. Using CAMELS (Cosmology and Astrophysics with MachinE Learning Simulations) data, including IllustrisTNG, SIMBA, and Astrid models, we train CNNs on Soft X-ray and 21-cm (HI) radio 2D maps to trace hot and cool gas, respectively, around galaxies, groups, and clusters. Our CNNs offer the unique ability to train and test on ''multifield'' datasets comprised of both HI and X-ray maps, providing complementary information about physical CGM properties and improved inferences. Applying eRASS:4 survey limits shows that X-ray is not powerful enough to infer individual halos with masses $\log(M_{\rm{halo}}/M_{\odot}) < 12.5$. The multifield improves the inference for all halo masses. Generally, the CNN trained and tested on Astrid (SIMBA) can most (least) accurately infer CGM properties. Cross-simulation analysis -- training on one galaxy formation model and testing on another -- highlights the challenges of develo** CNNs trained on a single model to marginalize over astrophysical uncertainties and perform robust inferences on real data. The next crucial step in improving the resulting inferences on physical CGM properties hinges on our ability to interpret these deep-learning models.
△ Less
Submitted 16 January, 2024; v1 submitted 14 September, 2023;
originally announced September 2023.
Enhanced mass-loss rate evolution of stars with $\gtrsim 18 M_\odot$ and missing optically-observed type II core-collapse supernovae
Authors:
Roni Anna Gofman,
Naomi Gluck,
Noam Soker
Abstract:
We evolve stellar models with zero-age main sequence (ZAMS) mass of $M_{\rm ZAMS} \gtrsim 18 M_\odot$ under the assumption that they experience an enhanced mass-loss rate when crossing the instability strip at high luminosities and conclude that most of them end as type Ibc supernovae (SNe Ibc) or dust-obscured SNe II. We explore what level of enhanced mass-loss rate during the instability strip w…
▽ More
We evolve stellar models with zero-age main sequence (ZAMS) mass of $M_{\rm ZAMS} \gtrsim 18 M_\odot$ under the assumption that they experience an enhanced mass-loss rate when crossing the instability strip at high luminosities and conclude that most of them end as type Ibc supernovae (SNe Ibc) or dust-obscured SNe II. We explore what level of enhanced mass-loss rate during the instability strip would be necessary to explain the `red supergiant (RSG) problem'. This problem refers to the dearth of observed core-collapse supernovae progenitors with $M_{\rm ZAMS} \gtrsim 18 M_\odot$. Namely, we examine what enhanced mass loss rate could make it possible for all these stars actually to explode as CCSNe. We find that the mass-loss rate should increase by a factor of at least about ten. We reach this conclusion by analyzing the hydrogen mass in the stellar envelope and the optical depth of the dusty wind at the explosion, and crudely estimate that under our assumptions only about a fifth of these stars explode as unobscured SNe II and SNe IIb. About 10-15 per cent end as obscured SNe II that are infrared-bright but visibly very faint, and the rest, about 65-70 per cent, end as SNe Ibc. However, the statistical uncertainties are still too significant to decide whether many stars with $M_{\rm ZAMS} \gtrsim 18M_\odot$ do not explode as expected in the neutrino driven explosion mechanism, or whether all of them explode as CCSNe, as expected by the jittering jets explosion mechanism.
△ Less
Submitted 16 April, 2020; v1 submitted 17 October, 2019;
originally announced October 2019.
