More Like this

1. An observationally driven multifield approach for probing the circum-galactic medium with convolutional neural networks

   Gluck, N; Oppenheimer, BD; Nagai, D; Villaescusa-Navarro, F; Angles-Alcazar, D (December 2023, Monthly notices of the Royal Astronomical Society)

   The circum-galactic medium (CGM) can feasibly be mapped by multiwavelength surveys covering broad swaths of the sky. With multiple large data sets 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 (H I ) radio two-dimensional 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’ data sets comprised of both H I and X-ray maps, providing complementary information about physical CGM properties and impro v ed inferences. Applying eRASS:4 surv e y limits shows that X-ray is not powerful enough to infer individual haloes with masses log ( M halo /M  ) < 12.5. The multifield impro v es 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 developing 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 the physical properties of CGM depends on our ability to interpret these deep-learning models. 

   more » « less
2. An observationally driven multifield approach for probing the circum-galactic medium with convolutional neural networks

   https://doi.org/10.1093/mnras/stad3784  

   Gluck, Naomi; Oppenheimer, Benjamin D; Nagai, Daisuke; Villaescusa-Navarro, Francisco; Anglés-Alcázar, Daniel (December 2023, Monthly Notices of the Royal Astronomical Society)

   ABSTRACT The circum-galactic medium (CGM) can feasibly be mapped by multiwavelength surveys covering broad swaths of the sky. With multiple large data sets 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 (H i) radio two-dimensional 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’ data sets comprised of both H i 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 haloes with masses log (Mhalo/M⊙) < 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 developing 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 the physical properties of CGM depends on our ability to interpret these deep-learning models. 

   more » « less
3. Constraining cosmic ray transport models using circumgalactic medium properties and observables

   https://doi.org/10.1093/mnras/staf1984  

   Lu, Yue Samuel; Kereš, Dušan; Hopkins, Philip F; Ponnada, Sam B; Faucher-Giguère, Claude-André; Hummels, Cameron B (December 2025, Monthly Notices of the Royal Astronomical Society)

   ABSTRACT Cosmic rays (CRs) are a pivotal non-thermal component of galaxy formation and evolution. However, the intricacies of CR physics, particularly how they propagate in the circumgalactic medium (CGM), remain largely unconstrained. In this work, we study CGM properties in FIRE-2 (Feedback In Realistic Environments) simulations of the same Milky Way (MW)–mass halo at $z=0$ with different CR transport models that produce similar diffuse $$\sim$$ GeV $$\gamma$$-ray emission, as an attempt to further constrain CR transport models. We study the gas morphology and thermal properties, and generate synthetic observations of rest-frame UV ion absorption columns and X-ray emission. CRs lower galaxy masses and star formation rates (SFRs) while supporting more cool CGM gas, which boosts the H i and O vi column densities in the CGM, bringing simulations more in line with observations, but there can be large differences between CR transport models and resolution levels. X-ray emission within and close to galaxies is consistent with thermal (free–free and metal-line) emission plus X-ray binaries, while more extended ($$\sim 100\,$$ kpc) CGM emission is potentially dominated by inverse Compton scattering (ICS), motivating future work on the spatially resolved X-ray profiles. Although comparisons with observations are sensitive to sample selection and mimicking the details of observations, and our analysis did not result in strong constraints on CR models, the differences between simulations are significant and could be used as a framework for future studies. 

   more » « less
4. Properties of the circumgalactic medium in cosmic ray-dominated galaxy haloes

   https://doi.org/10.1093/mnras/staa1849  

   Ji, Suoqing; Chan, T K; Hummels, Cameron B; Hopkins, Philip F; Stern, Jonathan; Kereš, Dušan; Quataert, Eliot; Faucher-Giguère, Claude-André; Murray, Norman (August 2020, Monthly Notices of the Royal Astronomical Society)

   ABSTRACT We investigate the impact of cosmic rays (CRs) on the circumgalactic medium (CGM) in FIRE-2 simulations, for ultra-faint dwarf through Milky Way (MW)-mass haloes hosting star-forming (SF) galaxies. Our CR treatment includes injection by supernovae, anisotropic streaming and diffusion along magnetic field lines, and collisional and streaming losses, with constant parallel diffusivity $$\kappa \sim 3\times 10^{29}\, \mathrm{cm^2\ s^{-1}}$$ chosen to match γ-ray observations. With this, CRs become more important at larger halo masses and lower redshifts, and dominate the pressure in the CGM in MW-mass haloes at z ≲ 1–2. The gas in these ‘CR-dominated’ haloes differs significantly from runs without CRs: the gas is primarily cool (a few $${\sim}10^{4}\,$$ K), and the cool phase is volume-filling and has a thermal pressure below that needed for virial or local thermal pressure balance. Ionization of the ‘low’ and ‘mid’ ions in this diffuse cool gas is dominated by photoionization, with O vi columns $${\gtrsim}10^{14.5}\, \mathrm{cm^{-2}}$$ at distances $${\gtrsim}150\, \mathrm{kpc}$$. CR and thermal gas pressure are locally anticorrelated, maintaining total pressure balance, and the CGM gas density profile is determined by the balance of CR pressure gradients and gravity. Neglecting CRs, the same haloes are primarily warm/hot ($$T\gtrsim 10^{5}\,$$K) with thermal pressure balancing gravity, collisional ionization dominates, O vi columns are lower and Ne viii higher, and the cool phase is confined to dense filaments in local thermal pressure equilibrium with the hot phase. 

   more » « less
5. Cooling flows as a reference solution for the hot circumgalactic medium

   https://doi.org/10.1093/mnras/staf786  

   Sultan, Imran; Faucher-Giguère, Claude-André; Stern, Jonathan; Rotshtein, Shaked; Byrne, Lindsey; Wijers, Nastasha (May 2025, Monthly Notices of the Royal Astronomical Society)

   ABSTRACT The circumgalactic medium (CGM) in $$\gtrsim 10^{12}\ \mathrm{M}_{\odot }$$ haloes is dominated by a hot phase ($$T \gtrsim 10^{6}$$ K). While many models exist for the hot gas structure, there is as yet no consensus. We compare cooling flow models, in which the hot CGM flows inwards due to radiative cooling, to the CGM of $$\sim 10^{12}{\,\rm to\,}10^{13}\ \mathrm{M}_{\odot }$$ haloes in galaxy formation simulations from the Feedback in Realistic Environments (FIRE) project at $$z\sim 0$$. The simulations include realistic cosmological evolution and feedback from stars but neglect AGN feedback. At both mass scales, CGM inflows are typically dominated by the hot phase rather than by the ‘precipitation’ of cold gas. Despite being highly idealized, we find that cooling flows describe $$\sim 10^{13}\ \mathrm{M}_{\odot }$$ haloes very well, with median agreement in the density and temperature profiles of $$\sim 20{{\ \rm per\ cent}}$$ and $$\sim 10{{\ \rm per\ cent}}$$, respectively. This indicates that stellar feedback has little impact on CGM scales in those haloes. For $$\sim 10^{12}\ \mathrm{M}_{\odot }$$ haloes, the thermodynamic profiles are also accurately reproduced in the outer CGM. For some of these lower-mass haloes, cooling flows significantly overpredict the hot gas density in the inner CGM. This could be due to multidimensional angular momentum effects not well captured by our one-dimensional cooling flow models and/or to the larger cold gas fractions in these regions. Turbulence, which contributes $$\sim 10{\!-\!}40{{\ \rm per\ cent}}$$ of the total pressure, must be included to accurately reproduce the temperature profiles. Overall, cooling flows predict entropy profiles in better agreement with the FIRE simulations than other idealized models in the literature. 

   more » « less