An official website of the United States government
Abstract We use IllustrisTNG simulations to explore the dynamic scaling relation between massive clusters and their—central—brightest cluster galaxies (BCGs). The IllustrisTNG-300 simulation we use includes 280 massive clusters from the z = 0 snapshot with M 200 > 10 14 M ⊙ , enabling a robust statistical analysis. We derive the line-of-sight velocity dispersion of the stellar particles of the BCGs ( σ *,BCG ), analogous to the observed BCG stellar velocity dispersion. We also compute the subhalo velocity dispersion to measure the cluster velocity dispersion ( σ cl ). Both σ *,BCG and σ cl are proportional to the cluster halo mass, but the slopes differ slightly. Thus, like the observed relation, σ *,BCG / σ cl declines as a function of σ cl , but the scatter is large. We explore the redshift evolution of the σ *,BCG − σ cl scaling relation for z ≲ 1 in a way that can be compared directly with observations. The scaling relation has a similar slope at high redshift, but the scatter increases because of the large scatter in σ *,BCG . The simulations imply that high-redshift BCGs are dynamically more complex than their low-redshift counterparts.
more »
« less
Finding Universal Relations in Subhalo Properties with Artificial Intelligence
Abstract We use a generic formalism designed to search for relations in high-dimensional spaces to determine if the total mass of a subhalo can be predicted from other internal properties such as velocity dispersion, radius, or star formation rate. We train neural networks using data from the Cosmology and Astrophysics with MachinE Learning Simulations project and show that the model can predict the total mass of a subhalo with high accuracy: more than 99% of the subhalos have a predicted mass within 0.2 dex of their true value. The networks exhibit surprising extrapolation properties, being able to accurately predict the total mass of any type of subhalo containing any kind of galaxy at any redshift from simulations with different cosmologies, astrophysics models, subgrid physics, volumes, and resolutions, indicating that the network may have found a universal relation. We then use different methods to find equations that approximate the relation found by the networks and derive new analytic expressions that predict the total mass of a subhalo from its radius, velocity dispersion, and maximum circular velocity. We show that in some regimes, the analytic expressions are more accurate than the neural networks. The relation found by the neural network and approximated by the analytic equation bear similarities to the virial theorem.
more »
« less
- Award ID(s):
- 2108944
- PAR ID:
- 10331328
- Date Published:
- Journal Name:
- The Astrophysical Journal
- Volume:
- 927
- Issue:
- 1
- ISSN:
- 0004-637X
- Page Range / eLocation ID:
- 85
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract We train graph neural networks to perform field-level likelihood-free inference using galaxy catalogs from state-of-the-art hydrodynamic simulations of the CAMELS project. Our models are rotational, translational, and permutation invariant and do not impose any cut on scale. From galaxy catalogs that only contain 3D positions and radial velocities of ∼1000 galaxies in tiny ( 25 h − 1 Mpc ) 3 volumes our models can infer the value of Ω m with approximately 12% precision. More importantly, by testing the models on galaxy catalogs from thousands of hydrodynamic simulations, each having a different efficiency of supernova and active galactic nucleus feedback, run with five different codes and subgrid models—IllustrisTNG, SIMBA, Astrid, Magneticum, SWIFT-EAGLE—we find that our models are robust to changes in astrophysics, subgrid physics, and subhalo/galaxy finder. Furthermore, we test our models on 1024 simulations that cover a vast region in parameter space—variations in five cosmological and 23 astrophysical parameters—finding that the model extrapolates really well. Our results indicate that the key to building a robust model is the use of both galaxy positions and velocities, suggesting that the network has likely learned an underlying physical relation that does not depend on galaxy formation and is valid on scales larger than ∼10 h −1 kpc.more » « less
-
Abstract Recent work has pointed out the potential existence of a tight relation between the cosmological parameter Ω m , at fixed Ω b , and the properties of individual galaxies in state-of-the-art cosmological hydrodynamic simulations. In this paper, we investigate whether such a relation also holds for galaxies from simulations run with a different code that makes use of a distinct subgrid physics: Astrid. We also find that in this case, neural networks are able to infer the value of Ω m with a ∼10% precision from the properties of individual galaxies, while accounting for astrophysics uncertainties, as modeled in Cosmology and Astrophysics with MachinE Learning (CAMELS). This tight relationship is present at all considered redshifts, z ≤ 3, and the stellar mass, the stellar metallicity, and the maximum circular velocity are among the most important galaxy properties behind the relation. In order to use this method with real galaxies, one needs to quantify its robustness: the accuracy of the model when tested on galaxies generated by codes different from the one used for training. We quantify the robustness of the models by testing them on galaxies from four different codes: IllustrisTNG, SIMBA, Astrid, and Magneticum. We show that the models perform well on a large fraction of the galaxies, but fail dramatically on a small fraction of them. Removing these outliers significantly improves the accuracy of the models across simulation codes.more » « less
-
Abstract We explore the redshift evolution of the dynamical properties of massive clusters and their brightest cluster galaxies (BCGs) at z < 2 based on the IllustrisTNG-300 simulation. We select 270 massive clusters with M 200 < 10 14 M ⊙ at z = 0 and trace their progenitors based on merger trees. From 67 redshift snapshots covering z < 2, we compute the 3D subhalo velocity dispersion as a cluster velocity dispersion ( σ cl ). We also calculate the 3D stellar velocity dispersion of the BCGs ( σ *,BCG ). Both σ cl and σ *,BCG increase as the universe ages. The BCG velocity dispersion grows more slowly than the cluster velocity dispersion. Furthermore, the redshift evolution of the BCG velocity dispersion shows dramatic changes at some redshifts resulting from dynamical interaction with neighboring galaxies (major mergers). We show that σ *,BCG is comparable with σ cl at z > 1, offering an interesting observational test. The simulated redshift evolution of σ cl and σ *,BCG generally agrees with an observed cluster sample for z < 0.3, but with large scatter. Future large spectroscopic surveys reaching to high redshift will test the implications of the simulations for the mass evolution of both clusters and their BCGs.more » « less
-
ABSTRACT A variety of observational campaigns seek to test dark matter models by measuring dark matter subhaloes at low masses. Despite their predicted lack of stars, these subhaloes may be detectable through gravitational lensing or via their gravitational perturbations on stellar streams. To set measurable expectations for subhalo populations within Lambda cold dark matter, we examine 11 Milky Way (MW)-mass haloes from the FIRE-2 baryonic simulations, quantifying the counts and orbital fluxes for subhaloes with properties relevant to stellar stream interactions: masses down to $$10^{6}\, \text{M}_\odot$$, distances ≲50 kpc of the galactic centre, across z = 0 − 1 (tlookback = 0–8 Gyr). We provide fits to our results and their dependence on subhalo mass, distance, and lookback time, for use in (semi)analytical models. A typical MW-mass halo contains ≈16 subhaloes $$\gt 10^{7}\, \text{M}_\odot$$ (≈1 subhalo $$\gt 10^{8}\, \text{M}_\odot$$) within 50 kpc at z ≈ 0. We compare our results with dark matter-only versions of the same simulations: because they lack a central galaxy potential, they overpredict subhalo counts by 2–10×, more so at smaller distances. Subhalo counts around a given MW-mass galaxy declined over time, being ≈10× higher at z = 1 than at z ≈ 0. Subhaloes have nearly isotropic orbital velocity distributions at z ≈ 0. Across our simulations, we also identified 4 analogues of Large Magellanic Cloud satellite passages; these analogues enhance subhalo counts by 1.4–2.1 times, significantly increasing the expected subhalo population around the MW today. Our results imply an interaction rate of ∼5 per Gyr for a stream like GD-1, sufficient to make subhalo–stream interactions a promising method of measuring dark subhaloes.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.3847/1538-4357/ac4d30
